Information Technology — HyperText Markup Language (ISO-HTML) <stitle>Technologies de l'information — Langage de balisage d'hypertexte (ISO-HTML) </titlep> <toc level=1> <foreword> <stdsrc>  <jtcsc>1 18 </stdsrc> <p> This &text; was prepared by the working group WG8 of SC18. The New Work Item Proposal was approved at the SC18 level in November 1996. <p> The New Work Item Proposal was approved at the JTC1 level in whenever 199x and assigned project number is 02.xx. <p> SC18/WG8 works on this project in liaison with the Internet Engineering Task Force and the World Wide Web Consortium. The text of this &text; follows closely the &dave; specification edited by David Raggett <cptr>URL:http://www.w3.org/pub/WWW/TR</cptr>, and the W3C Working Draft dated 1996-04-22 describing work in progress on "Inserting objects into HTML" <cptr>URL:http://www.w3.org/pub/WWW/TR/WD-object.html</cptr>. The World Wide Web Consortium's Recommendation is based on specifications prepared by the Internet Engineering Task Force: <ul> <li>IETF Proposed Standard RFC 1866 "HyperText Markup Language — 2.0" edited by Tim Berners-Lee and Daniel W. Connolly, September 22nd, 1995. <cptr>URL:ftp://ds.internic.net/rfc/rfc1866.txt</cptr> <li>IETF RFC 1867 "Form-based File Upload in HTML" edited by E. Nebel and L. Masinter, November 1995. <cptr>URL:ftp://ds.internic.net/rfc/rfc1867.txt</cptr> <li>IETF RFC 1942 "HTML Tables" edited by Dave Raggett, May 1996. <cptr>URL:ftp://ds.internic.net/rfc/rfc1942.txt</cptr> <li>IETF Internet Draft "Internationalization of the HyperText Markup Language" edited by François Yergeau, Gavin T. Nicol, Glenn Adams and Martin J. Dürst, January, 1997. <cptr>URL:ftp://ds.internic.net/rfc/rfc2070.txt</cptr> </ul> <replace>all previous working documents <repcom>which are now obsolete. </foreword> <intro id=intro> <p> The HyperText Markup Language is an SGML application conforming to International Standard ISO 8879 — Standard Generalized Markup Language. It provides a simple way of structuring hypertext documents which refer to one another and which collectively create an enormous "web" which continues to grow and evolve as many hypertext authors add and modify documents. <p> The web has expanded and browser developers have added additional features to the markup language such as new tags and new semantics for the tags. As a result, many documents have been created which can only be rendered faithfully on a limited number of browsers. Normal web practice is to hide any syntactic problems detected by the browsers and thus the reader is not always aware that the page being browsed is not faithful to the original authored document. <p> This &text; has been developed in an effort to ensure that it will remain possible for an author to produce simple hypertext for the web and be confident that a conforming browser will be able to render the document faithfully. This specification represents a core of the language to be supported by all conforming browsers and provides techniques for extending the core that are SGML conformant and represent good SGML practice. <p> The language defined by this &text; differs slightly from the &dave; specification. It omits all deprecated features of the language, and some other features whose role is purely cosmetic. This is done in preparation for the expected future introduction of style sheets. Certain optional facilities such as markup omission of the document element and the major elements have been removed to produce more robust texts in keeping with recognized good SGML practice. This does not reduce in any way the expressive power of the language. A minimal ISO-HTML document has the form: <cptr keep=off type=source width=column> <!DOCTYPE ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language//EN" [<!ENTITY % ISOhtme PUBLIC "ISO 15xxx//ENTITIES CD//EN" > <!ENTITY % ISOlat1 PUBLIC "ISO 8879:1986//ENTITIES Added Latin 1//EN" > %ISOlat1;%ISOhtme; ]> <ISO-HTML> <HEAD> <TITLE>Les unités de base</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <p> This &text; follows the current World Wide Web convention of presenting element and attribute names in upper case, although there is no formal requirement for the practice. <p> In order to support world wide use of the markup language, the internationalization facilities specified by the IETF have been included in this &text; in anticipation of their expected inclusion in the W3C specifications. <p> The elements of the ISO-HTML language are specified in this &text; in alphabetical order but may also be reviewed in the following order which reflects more closely the structure of the DTD: <sl compact> <li>The ISO-HTML element <hdref refid=iso-html page=yes> <sl compact> <li>The HEAD element <hdref refid=head page=yes> <sl compact> <li>The BASE element <hdref refid=base page=yes> <li>The ISINDEX element <hdref refid=isindex page=yes> <li>The LINK element <hdref refid=link page=yes> <li>The META element <hdref refid=meta page=yes> <li>The TITLE element <hdref refid=title page=yes> </sl> <li>The BODY element <hdref refid=body page=yes> <sl compact> <li>The ADDRESS element <hdref refid=address page=yes> <li>The block elements parameter entity <cptr>%block;</cptr> <sl compact> <li>The BLOCKQUOTE element <hdref refid=blockquote page=yes> <li>The DIV element <hdref refid=div page=yes> <li>The DL element <hdref refid=dl page=yes> <sl compact> <li>The DT element <hdref refid=dt page=yes> <li>The DD element <hdref refid=dd page=yes> </sl> <li>The FORM element <hdref refid=form page=yes> <li>The HR element <hdref refid=hr page=yes> <li>The ISINDEX element <hdref refid=isindex page=yes> <li>The OL element <hdref refid=ol page=yes> <sl compact> <li>The LI element <hdref refid=li page=yes> </sl> <li>The P element <hdref refid=p> page=yes> <li>The PRE element <hdref refid=pre page=yes> <li>The TABLE element <hdref refid=table page=yes> <sl compact> <li>The CAPTION element <hdref refid=caption page=yes> <li>The TR element <hdref refid=tr page=yes> <li>The TH element <hdref refid=th page=yes> <li>The TD element <hdref refid=td page=yes> </sl> <li>The UL element <hdref refid=ul page=yes> </sl> <li>The text elements parameter entity <cptr>%text;</cptr> <sl compact> <li>The physical style elements parameter entity <cptr>%physical.styles;</cptr> <sl compact> <li>The B element <hdref refid=b page=yes> <li>The BIG element <hdref refid=big page=yes> <li>The I element <hdref refid=i page=yes> <li>The SMALL element <hdref refid=small page=yes> <li>The STRIKE element <hdref refid=strike page=yes> <li>The SUB element <hdref refid=sub page=yes> <li>The SUP element <hdref refid=sup page=yes> <li>The TT element <hdref refid=tt page=yes> <li>The U element <hdref refid=u page=yes> </sl> <li>The logical style elements parameter entity <cptr>%logical.styles;</cptr> <sl compact> <li>The CITE element <hdref refid=cite page=yes> <li>The CODE element <hdref refid=code page=yes> <li>The DFN element <hdref refid=dfn page=yes> <li>The EM element <hdref refid=em page=yes> <li>The KBD element <hdref refid=kbd page=yes> <li>The Q element <hdref refid=q page=yes> <li>The SAMP element <hdref refid=samp page=yes> <li>The STRONG element <hdref refid=strong page=yes> <li>The VAR element <hdref refid=var page=yes> </sl> <li>The form elements, parameter entity <cptr>%form;</cptr> <sl compact> <li>The INPUT element <hdref refid=input page=yes> <li>The SELECT element <hdref refid=select page=yes> <sl compact> <li>The OPTION element <hdref refid=option page=yes> </sl> <li>The TEXTAREA element <hdref refid=textarea page=yes> </sl> <li>The special elements, parameter entity <cptr>%special;</cptr> <sl compact> <li>The A element <hdref refid=a page=yes> <li>The BDO element <hdref refid=bdo page=yes> <li>The BR element <hdref refid=br page=yes> <li>The IMG element <hdref refid=img page=yes> <li>The OBJECT element <hdref refid=object page=yes> <sl compact> <li>The PARAM element <hdref refid=param page=yes> </sl> <li>The MAP element <hdref refid=map page=yes> <sl compact> <li>The AREA element <hdref refid=area page=yes> </sl> <li>The SPAN element <hdref refid=span page=yes> </sl> </sl> <li>The sectioning elements <sl compact> <li>The H1 element <hdref refid=h1 page=yes> <li>The B1 element <hdref refid=b1 page=yes> <li>The H2 element <hdref refid=h2 page=yes> <li>The B2 element <hdref refid=b2 page=yes> <li>The H3 element <hdref refid=h3 page=yes> <li>The B3 element <hdref refid=b3 page=yes> <li>The H4 element <hdref refid=h4 page=yes> <li>The B4 element <hdref refid=b4 page=yes> <li>The H5 element <hdref refid=h5 page=yes> <li>The B5 element <hdref refid=b5 page=yes> <li>The H6 element <hdref refid=h6 page=yes> <li>The B6 element <hdref refid=b6 page=yes> </sl> </sl> </sl> </sl> <h2 id=markup>This document <p> This &text; represents the editors' best efforts to capture the continuing development of the language. The editors' have copied text from <ul> <li>the IETF RFC 1866 <bibref refid=html2.0>, <li>the &dave; specification <bibref refid=html3.2>, <li>the W3C Working Draft dated 1996-04-22 describing work in progress on <cit>Inserting objects into HTML</cit> <bibref refid=w3c.object>, <li>the SGML Handbook <bibref refid=goldfarb> where needed to ensure consistency with those specifications, <li>the IETF HTML internationalisation specification <bibref refid=rfci18n>. </ul> <h3 id=source>Source markup <p> This document was prepared using ISO 8879 based technology. Details of the DTD used to structure this document may be found in <cit>ISO/IEC TR 9573-11 Information processing — SGML support facilities — Techniques for using SGML — Part 11: Application at ISO Central Secretariat for International Standards and Reports</cit>. <body cols=2> <scope id=scope> <p> The scope of this &text; is a conforming application of SGML, ISO 8879:1986, which provides a simple and stable markup language for documents to be published on the World Wide Web. This &text; also provides techniques whereby application, regional or industry-specific extensions may be added in a way which follows good SGML practice. <h2 id=specif>Specificity of the scope <p> Since it is expected that a wide range of products, user applications, recommendations and other standards may use this standard, the scope focuses on the general structuring aspects and provides only sufficient semantics to ensure that the structures are rendered in a familiar way. <p> The scope excludes any standardization of models, services, systems, protocols or applications which are likely to make use of the ISO-HTML language. This specification does not define the "look and feel" of any conforming product. </scope> <conf id=conf> <h2 id=docconf>Conforming ISO-HTML documents <p> A document which conforms to this &text; shall <ol> <li>be a conforming SGML document, <note>&NB; This requirement is itself required by ISO 8879:1986 subclause 15.2.2 [Goldfarb p.480]. </note> <li>conform to the requirements of this &text;. </ol> <h2 id=sysconf>Conforming ISO-HTML systems <p> A conforming ISO-HTML system is a conforming SGML system which is able to process any conforming ISO-HTML document. <note> This requires the system to be capable of processing any conforming ISO-HTML document that is not inconsistent with its "system declaration". In order to promote the maximum capability for document interchange among systems, every conforming system must be able to support the reference concrete syntax and the ISO-HTML capacity set [Goldfarb p.481]. </note> <h3 id=sysdoc>Documentation of conforming ISO-HTML systems <p> Systems conforming to this &text; shall display the ISO-HTML identification text prominently: <ol compact> <li>In a prominent location in the front matter of all publications (normally the title page and the cover page), <li>On all identifying display screens of programs, <li>In all promotional and training material. </ol> and in the national language of the documentation. <p> The ISO-HTML system identification text is: <lq>ISO-HTML, an SGML system conforming to International Standard ISO 8879 — Standard Generalised Markup Language.</lq> <note> These requirements are intended to help users apply knowledge gained on one SGML system to the use of other systems, not to inhibit friendly documentation and human-computer interfaces [Goldfarb p.486]. </note> <note>&NB; The editors have deliberately omitted any reference to the national language to be used for the identification text. The issue is too complex, and not worth the trouble [Goldfarb p.486 subclause 15.5.1 line 6]. </note> <p> The documentation shall distinguish SGML constructs from ISO-HTML conventions and system functions, and shall identify the SGML constructs as being part of the Standard Generalized Markup Language. <nl> <li>The objective of this requirement is for the user to be aware of which constructs are common to all SGML systems, and which are unique to ISO-HTML. This will reduce the experienced user's learning time for a new system or application. <li>&NB; This requirement is itself required by ISO 8879:1986 subclause 15.5.2 [Goldfarb p.486 line 23]. </nl> <p> The documentation shall cite ISO 8879:1986 as a reference for supported SGML constructs that are not specifically documented for the system. <p> For example, if, for simplicity's sake, only a subset of some function is presented in the documentation (such as omitting some of the options of the entity declaration), it shall be stated clearly that other options exist and can be found in the SGML International Standard [Goldfarb p.487]. <note>&NB; Do we need to include text to cover the requirement of ISO 8879:1986 subclause 15.5.3 "Terminology"? [Goldfarb p.487]. </note> <h3 id=sysvalid>Validating ISO-HTML systems <p> In addition to being a conforming ISO-HTML system, a system is also a validating ISO-HTML system if <ol> <li>It is a validating SGML parser as defined by ISO 8879:1986 subclause 15.4; and <li>It finds and reports an ISO-HTML error if one exits; and <li>Does not report an ISO-HTML error where none exists. </ol> <note>&NB; The &dave; specification does not say what the reportable HTML errors are. </note> <h2 id=confi18n>Character set conformance <note>&NB; We have taken the same approach here as was taken by ISO 2022, and this subclause is based on ISO 2022 clause 3. </note> <p> The SGML declaration provided with this &text; calls for the use of the Basic Multilingual Plane of the ISO/IEC 10646 Universal Multiple-Octet Coded Character Set (UCS). ISO/IEC 10646 addresses whole classes of provisions and it is not intended by this &text; that they are all implemented in any user agent. As a result it is only practicable to envisage limited conformance to ISO/IEC 10646 as defined in this subclause. <p> Under limited conformance, the following is required: <ol> <li>When the characters described by ISO/IEC 10646 are used, they shall be implemented by the control functions, and with the meanings and coded representation specified in ISO/IEC 10646. <li>When two systems with different levels of implementation of the ISO/IEC 10646 coded character set are required to communicate with one another, they shall do so using using that part of the coded character set they have in common. <nl> <li>This &text; does not discuss the operation of the HTTP protocol <bibref refid=http>. </nl> <li>If a server is unable to express a document using the limited character set supported by the user agent, it should instead deliver a document in the limited character set explaining the impossibility. <li>Code positions that are either reserved for registration or reserved for future standardization shall not be used. <li>No registered escape sequence shall be used with a meaning different from that defined by ISO/IEC 10646. </ol> <p> The UTF-1 transformation format of ISO/IEC 10646, registered by IANA as ISO-10646-UTF-1, has been removed from the ISO/IEC 10646 and should not be used. </conf>  <refs apply=all> <refref>ISO 639:1988 <refref>ISO 646:1991 <refref>ISO 3166:1993 <refref>ISO 8859-1:1987 <refref>ISO 8879:1986 <refref>ISO 10646-1:1993 </refs> <defs apply=all id=defs> <defref>ISO 8879:1986 <dl> <dt id=d.justifying>Justifying element <dd>An element which carries an &a.align; attribute used to specify horizontal justification of its content relative to the current left and right margins. The &a.align; attribute takes one of the following values: <ul compact> <li>LEFT: text is left justified, ragged right. <li>CENTER: text is centered, ragged left and ragged right. <li>CENTRE: a permitted alternative specification of CENTER. <li>RIGHT: text is right justified, ragged left. </ul> <p> The justifying elements specified by this &text; are &justifying;. <dt id=d.boxing>Boxing element <dd>An element which carries an &a.align; attribute used to specify the position of a box around the visible rendered area. <p> The boxing elements specified by this &text; are &boxing;. <dt id=d.browser>Browser <dd>A User Agent whose main function is to present documents to the user. <dt id=d.character>Character <dd source="RFC1866">An atom of information, for example a letter or a digit. Graphic characters have associated glyphs, whereas control characters have associated processing semantics. <p> The multiple definitions and techniques for the representation of characters may be the source of confusion. The following table shows some of the ideas involved: <fig id=chars place=fixed width=page> <figbody> <cptr keep=off type=source width=page> The set of characters in this column forms the character repertoire |-------------------->| | This function called| |"coded character set"| | by RFC 1866 | ,---|---------------------|---------------------------------, | Code | Bit | Character |ISOlat1 |Numeric |Glyph| |posit- |pattern| name |entity |character| | | ion |(hexa) | |ref. |reference| | |-------+-------+------------------+--------+---------+-----| | 0 | 00 |Unused | | � | | | ... | | | | | | | 121 | 79 |SMALL LETTER y | | y | y | | 122 | 7A |SMALL LETTER z | | z | z | | 123 | 7B |LEFT CURLY BRACKET| | { | { | | ... | | | | | | | 199 | C7 |CAPITAL C, CEDILLA|Ç| Ç | | | 200 | C8 |CAPITAL E, GRAVE |È| È | | | 201 | C9 |CAPITAL E, ACUTE |É| É | | | ... | | | | | | | 255 | FF |SMALL y, UMLAUT |ÿ | ÿ | | '--|-------|-------------|----------------------------------' | | | | |------------>| | | This function is called the "character encoding | | scheme" by RFC 1866. | |<------------| | | This function is called "character set" by SGML | | | Called the "code set" by SGML Called the "character number" by SGML </cptr> <figcap>Illustration of some character representation definitions <figdesc>This illustration is based on the character set defined by ISO 8859-1:1987 "8-bit single-byte coded graphic character sets", Part 1: Latin alphabet No. 1. </fig> <dt id=d.ces>Character encoding scheme <dd source="RFC1866">A function whose domain is the set of sequences of octets, and whose range is the set of sequences of characters from a character repertoire; that is, a sequence of octets and a character encoding scheme determines a sequence of characters. <dt id=d.charrep>Character repertoire <dd source="RFC1866">A finite set of characters; eg. the range of a coded character set. <dt id=d.codepos>Code position <dd source="RFC1866">An integer; a coded character set and a code position from its domain determine a character. <dt id=d.ccs>Coded character set <dd source="RFC1866">A function whose domain is a subset of the integers and whose range is a character repertoire. That is, for some set of integers (usually of the form {0, 1, 2, ..., N}), a coded character set and an integer in that set determine a character. Conversely, a character and a coded character set determine the character's code position (or, in rare cases, a few code positions). <dt id=d.CRLF>CRLF <dd source="RFC1521">The sequence of the two ISO 646:1983 characters CR (13) and LF (10) which, taken together, in this order, denote a line break. <dt id=d.fds>Form data set <dd source="RFC 1866">A sequence of name/value pairs; the names are given by an ISO-HTML document and the values are given by the user. <dt id=d.fragid>Fragment identifier <dd source="RFC1866">The portion of an &a.href; attribute following the `<cptr>#</cptr>' character which may modify the presentation of the destination of a hyperlink. <dt id=d.HTMLbrowser>ISO-HTML browser <dd>Browser which presents ISO-HTML documents. <dt id=d.document>ISO-HTML document <dd>A document structured in accordance with this &text;. <dt id=d.hlink>Hyperlink <dd>A relationship between two anchors, called the source and the target. The link goes from the source to the target. The source is also known as the tail, and the target is also known as the destination or head. </dl> </defs> <symabb id=abbrev> <p> The following symbols and abbreviations are used in this &text;. <syml> <sym>ISO-HTML <desc>Pertaining to this standard. <sym>HTML 3.2 <desc>A specification for the Hypertext Markup Language developed by the World Wide Web Consortium. <sym>HTTP <desc>IETF RFC 2068 HyperText Transfer Protocol. <sym>HyTime <desc>One or more architectural forms as provided by <refref>ISO 10744:-</refref>. <sym>IANA <desc>Internet Assigned Numbers Authority. IANA is the central coordinator for the assignment of unique parameter values for Internet protocols. The IANA is chartered by the Internet Society (ISOC) and the Federal Network Council (FNC) to act as the clearinghouse to assign and coordinate the use of numerous Internet protocol parameters <bibref refid=rfc1700>. <sym>IETF <desc>Internet Engineering Task Force. <sym>RFC <desc>Request for Comments. An Internet Engineering Task Force specification. <sym>SGML <desc>Notation provided through use of <refref>ISO 8879:1986</refref>. <sym>URI <desc>Universal Resource Identifier as defined by <bibref refid=uri>. <sym>URL <desc>Uniform Resource Locator as defined by <bibref refid=url>. <sym>WWW <desc>World Wide Web <sym>W3C <desc>World Wide Web Consortium, founded in 1994 to develop common standards for the evolution of the World Wide Web. It is an industry consortium, hosted by the Massachusetts Institute of Technology Laboratory for Computer Science (MIT/LCS) in the United States, the Institut National de Recherche en Informatique et en Automatique (INRIA) in Europe and the Keio University Shonan Fujisawa Campus in Asia <cptr>http://www.w3.org/</cptr>. </syml> </symabb> <h1 id=requ>Requirements <p> This &text; has been designed to satisfy the following requirements, as documented in ISO/IEC JTC1/SC18/WG8 N1852 dated 24 June 1996: <ul> <li>Provide a minimum presentation architecture for SGML applications. <li>Clarify relationships between ISO-HTML and SGML details, such as SGML declaration and minimization. <li>Allow ISO-HTML to be used in environments where ISO standards are required. <li>Allow ISO-HTML to be used as a base architecture for other SGML applications. <li>Documents conforming to this &text; should be viewable by browsers that conform to the &dave; (or an appropriate later version of that specification). <li>Define useful subsets of the &dave; (or an appropriate later version of that specification). </ul>  <h1 id=i18n>Internationalization <p> The early applications of HTML on the World-Wide Web were seriously restricted by their reliance on the ISO 8859-1:1987 coded character set, which is appropriate only for Western European languages. This &text; provides facilities for a fuller internationalization, based on RFC 2070 <bibref refid=rfci18n>. The following subclauses specify the general internationalization features. <h2 id=a.justify>The ALIGN attribute when used for justification <p> The internationalization facilities use the &a.align; attribute to specify the horizontal alignment to be applied to content within the scope of the justifying elements: &justifying;. <p> The &a.align; justification attribute takes the following values: <dl> <dt>ALIGN=left <dd>Render "flush left" or left justified. The right edge is ragged. <dt>ALIGN=center <dd>Render "centered". The left and right edges are ragged. <dt>ALIGN=centre <dd>A permitted alternative specification of ALIGN=center. <dt>ALIGN=right <dd>Render "flush right" or right justified. The left edge is ragged. <dt>ALIGN=justify <dd>Render fully justified on the left and right. There are no ragged edges. </dl> <p> In the case of nested justification specifications, the innermost containing specification shall apply. If a justifying &a.align; attribute is omitted, the value of the next outer justifying specification is used. If the &a.align; attribute is omitted from all containing justifying elements, the value "left" shall be used. <note> There are elements defined by this &text; which have an &a.align; attribute for historical reasons, but are not justifying elements. E.g. the &caption; element and the boxing elements &boxing;. </note> <h3 id=align.ex>Examples of justifying behaviour <p> <ol> <li>In this example all the containing justification specifications are present, and the innermost applies. <cptr keep=off type=source width=column> <DIV ALIGN=left> <OL ALIGN=justify> <LI ALIGN=right>Important sentence ... </OL> </DIV> </cptr> The <q>Important sentence</q> is right aligned. <li>In this example only some of the containing justification specifications are present, and the innermost specification applies. <cptr keep=off type=source width=column> <DIV ALIGN=left> <OL> <LI ALIGN=justify> <P>Important sentence ... </OL> </DIV> </cptr> The <q>Important sentence</q> is fully justified. <li>In this example none of the containing justification specifications are present, hence the default value applies. <cptr keep=off type=source width=column> <DIV> <P>Important sentence ... </DIV> </cptr> The <q>Important sentence</q> is left justified. </ol> <h2 id=a.class>The CLASS attribute <p> It is expected that many ISO-HTML documents will be created by automatic processes from SGML documents structured according to richer document type definitions, eg. the DTD for ISO documents specified in ISO TR 9573. Conversion to the ISO-HTML structure may lead to the loss of some information about document content. For example the notes in an ISO document are tagged using the ¬e; element, rather than the &p; element. A rendering system might typically apply the ISO note style to the content of a ¬e; element, making it distinct in the output presentation. There may be a need to retain this type of information during the conversion to ISO-HTML so that browsers with the capability may provide a distinctive rendering. <p> The &a.class; attribute provides a means of attaching a sequence of SGML names to the element indicating the class to which the element belongs: typically the source element name if the document is the result of an automatic conversion from another document type. <note> This &text; does not define values for the &a.class; attribute. </note>  <h3 id=class.ex>An example <p> A conversion process might convert <cptr><note>This text does not define values for the class attribute.</note></cptr> to <cptr><P CLASS=note>This text does not define values for the class attribute.</P></cptr> <h2 id=a.dir>The DIR attribute <p> The &a.dir; attribute indicates the writing direction of text. With block-type elements the &a.dir; attribute indicates the base writing direction of the text in the block. If the attribute is omitted, the writing directional is inherited from the parent element if present. With inline elements, it indicates the element start a new writing direction embedding level. Directional-embedding is used to handle nested changes in writing direction. A common need for embedding characters is to handle text that has been pasted from one bidirectional context to another. <h3 id=dir.ex>An example <p> Here is an example of a case where direction-embedding is needed: <p> The following latin (shown in upper case) and arabic (shown in lower case) letters in backing store with the specified writing direction embeddings (LRE is shorthand for <SPAN DIR=ltr>, RLE for <SPAN DIR=rtl> and PDF for </SPAN>): <cptr keep=off type=source width=column> LRE A B RLE a b LRE C D PDF c d PDF E F PDF </cptr> result in the following rendering with square brackets showing the directional transitions: <cptr keep=off type=source width=column> [ A B [ d c [ C D ] b a ] E F ] </cptr> <note> If the &span; elements with &a.dir; attributes were not used, this example would result in the following rendering, assuming an inherited direction of "ltr". <cptr keep=off type=source width=column> [ A B [ b a [ C D ] d c ] E F ] </cptr> Notice that <cptr>b a</cptr> is on the left and <cptr>d c</cptr> is on the right unlike the example where writing direction embedding levels are used. Without writing direction embedding characters there would be at most two levels: a base directional level and a single counterflow level. </note> <h2 id=a.id>The ID attribute <p> When documents with the same information content in different languages are created, there may be a need to identify elements in the documents as being the "same" element, differing only in the language employed. <p> The &a.id; attribute provides a means of identifying the "same" element in different documents. Within each document, the elements' &a.id; attribute is set to the same value. <h2 id=a.lang>The LANG attribute <p> The &a.lang; attribute identifies a natural language spoken, sung, written or otherwise used by human beings for communication of information between people. Computer languages are explicitly excluded. The value of the &a.lang; attribute is referred to as the "language tag". <p> The syntax and registry of language tags used by this &text; is the same as that defined by RFC 1766 <bibref refid=rfc1766>, where a language tag is composed of one or more parts: a primary language tag and a possible empty series of subtags. For example, using RFC 822 EBNF: <cptr keep=off type=source width=column> language-tag = primary-tag *( "-" subtag ) primary-tag = 1*8ALPHA subtag = 1*8ALPHA </cptr> Whitespace is not allowed within the tag and all tags are case insensitive. The namespace of language tags is administered by the IANA. Example tags include: <cptr>en</cptr>, <cptr>en-US</cptr>, <cptr>en-cockney</cptr>, <cptr>i-cherokee</cptr> and <cptr>x-pig-latin</cptr>. <p> Two-letter primary-tags are reserved for ISO 639 language abbreviations. This &text; does not specify three-letter primary-tags, however their description may be found in the "Ethnologue" <bibref refid=ethno>. Any two-letter initial sub-tag is an ISO 3166 country name. <p> In the context of this &text;, a language tag is not to be interpreted as a single token, per RFC 1766, but as a hierarchy. For example, a user agent that adjusts its rendering according to language should recognize that it has a match when a language tag in a style sheet entry matches the initial portion of the language tag of an element. An exact match should be preferred. This interpretation allows an element marked up as, for instance, <cptr>en-US</cptr> to trigger styles corresponding to, in order of preference, US-english (<cptr>en-US</cptr>) or `international' English (<cptr>en</cptr>). <note> Using the language tag as a hierarchy does not imply that all languages with a common prefix will be understood by those fluent in one or more of the languages; it simply allows the user to request commonality when desired. </note> <note> It is intended that any new element introduced in later versions or extensions to ISO-HTML will admit the &a.lang; attribute, unless there is a compelling reason not to do so. </note> <p> The rendering of elements is meant to be controlled (in part) by the &a.lang; attribute. Specific user preferences set within a browser should override the &a.lang; attribute which in turn overrides the value specified by the &a.lang; attribute of any enclosing element. If none of these are set, a suitable default, perhaps controlled by the user's locale, should be used to control rendering. <h2 id=unknown>User agent unable to handle characters <p> If a user agent is unable to handle a character due, for example, to a lack of a resource such as a font, this should be indicated to the end user, clearly but unobtrusively. <nl> <li>Since some documents may contain many characters that cannot be rendered, showing an alert for each one is excessive. <li>If a numeric representation of a missing character is given, its hexadecimal, not decimal, form is to be preferred since this is the form used in character set standards. </nl> <h2 id=bytesex>Byte order <p> When an ISO-HTML text is transmitted directly in a multibyte representation, this &text; recommends: <ol> <li>That it be transmitted in big-endian byte order — high order byte first. <li>That the document always begin with a ZERO-WIDTH NON-BREAKING SPACE character (hexadecimal FEFF) which, when byte-reversed becomes number FFFE, a character guaranteed never to be assigned. Thus a user agent receiving an FFFE as the first two octets of a text would know that bytes have to be reversed for the remainder of the text. </ol> <h2 id=namedch>Use of named character references <p> It is often convenient for an author to use named character references to specify characters which are not available on the author's keyboard. Wherever named characters are used for characters specified by ISO 10646, the name used shall be that specified in ISO TR 9573. <note> ISO 9573 defines names for all the characters specified by ISO 10646. The intention of this requirement is to facilitate interchange. </note> <p> Example: <cptr><EM>Liberté Égalité Fraternité!</EM></cptr> <h1 id=useDTD>The DTD <p> The DTD provided by this &text; has a public document type definition with the invocation: <cptr keep=off type=source width=column> <!ENTITY % ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language//EN"> %ISO-HTML; </cptr> <h2 id=nosubset>Alternate public document type definition <p> Many popular user agents do not support the use this &text; makes of the document type declaration subset. To facilitate the use of ISO-HTML an alternate public document type definition is provided to specify a DTD which is totally self contained and contains all of the features of ISO-HTML without using the document type declaration subset. <cptr keep=off type=source width=column> <!ENTITY % ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language Core//EN"> %ISO-HTML; </cptr> <p> With the use of this public document type definition the minimum document becomes: <cptr keep=off type=source width=column> <!DOCTYPE ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language Core//EN"> <ISO-HTML> <HEAD> <TITLE>Les unités de base</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <h1 id=copyr>Copyright protection <p> This document makes several references to industry and proprietary standards, products and publications. Such references are not normative, and do not imply endorsement by the ISO, IEC, or their national member bodies or affiliates. Any brand names or trademarks mentioned are the property of their respective owners. <p> The formal SGML definitions are part of the text of this &text; and are protected by copyright held by the IETF and the W3C. <h1 id=linking>Hyperlinks <note>&NB; This clause is based entirely on clause 7 in RFC 1866 HTML 2.0 <bibref refid=html2.0>. </note> <p> Hyperlinks provide the mechanism which ties the web together and gives it an overall structure. They are the threads of gossamer glueing documents together. They have two ends, known as the <hp1>source anchor</hp1> and the <hp1>target anchor</hp1> which are both identified by their address: an absolute Uniform Resource Identifier (URI) <bibref refid=uri>, optionally followed by a `<cptr>#</cptr>' character and a further sequence of characters called a fragment identifier. For example: <cptr keep=off type=source width=column> http://www.xyz.org/hypertext/WWW/TheProject.html http://www.xyz.org/hypertext/WWW/TheProject.html#Overview </cptr> <note>&NB; In RFC 1866, source anchors are called <hp1>tail</hp1> anchors, and target anchors are called <hp1>head</hp1> anchors as viewed from the current position in the hypertext document by a user agent. That is, prior to selecting a hypertext link its destination is seen as a target, but after selection and rendering the same position subsequently appears as a source. </note>  <p> In an anchor address, the URL refers to a resource; it may be used in a variety of information retrieval protocols to obtain an entity that represents the resource, such as an ISO-HTML document. The fragment identifier, if present, marks some position in the resource. <p> Each of the following constructs provides the target anchor of a hyperlink or set of hyperlinks: <ul compact> <li>The &a; element with the &a.href; attribute present. <li>The &img; element. <li>The &link; element. <li>The &isindex; element. <li>The &form; element with &a.method;<cptr>=get</cptr>. <li>The &input; element with the &a.src; attribute present. </ul> These markup constructs refer to source anchors by URL, in either absolute or relative form, or to a fragment identifier, or to both. <p> In the case of a relative URL, the actual address referenced by the user agent is constructed by combining the specified relative address with an absolute base URL as in RFC 1808 <bibref refid=relurl>. The base address is taken from the document's &base; element, if present; otherwise it is determined as in RFC 1808 <bibref refid=relurl>. <h2 id=access>Accessing resources <p> Once the address specified by the target anchor is determined, the user agent may obtain a representation of the resource (now the source). <p> For example, if the base address of a document is <cptr>http://host/x/y.html</cptr> and the document contains: <cptr keep=off type=source width=column> <IMG SRC="../icons/abc.jpeg"> </cptr> then the user agent uses the URL <cptr>http://host/icons/abc.jpeg</cptr> to access the resource. <note> The user agent constructs the destination URL by discarding the file identifier portion of the base address (if present). </note> <h2 id=activ>Activation of hyperlinks <p> An ISO-HTML user agent allows the user to navigate the content of a document and request activation of hyperlinks denoted by &a; elements. To activate a link, the user agent obtains a representation of the resource identified by the address specified in the source anchor. If the representation is another ISO-HTML document, navigation may begin again with this new document. <h2 id=simulimg>Simultaneous presentation of image resources <p> An ISO-HTML user agent may activate hyperlinks indicated by &img; elements concurrently with processing a document; that is, image hyperlinks may be processed without explicit request by the user. Image resources should be embedded in the presentation at the point of the source anchor, that is the &img; element. The same also holds for &input; elements within forms. <h2 id=fragid>Fragment identifiers <p> Any characters following a `<cptr>#</cptr>' character in a hypertext address constitute a fragment identifier. In particular, an address of the form <cptr>#fragment</cptr> refers to a target anchor in the document. <p> The meaning of fragment identifiers depends on the media type of the representation of the target anchor's resource. For <cptr>text/html</cptr> representations, it refers to the &a; element whose &a.name; attribute value is the same as the fragment identifier. Character matching is case sensitive. The target document shall have exactly one such element. The user agent should render the source anchor element by scrolling to and/or highlighting the text at the destination position. <p> For example, if the base URL is <cptr>http://host/x/</cptr> and the user activates the link denoted by the following markup: <cptr keep=off type=source width=column> <P>See: <A> HREF="app1.html#bananas">Appendix 1</A> for more details on bananas. </cptr> then the user agent accesses the resource identified by <cptr>http://host/x/app1.html</cptr> and assuming the resource is represented using the <cptr>text/html</cptr> media type, the user agent shall locate the &a; element whose &a.name; attribute has the value <cptr>bananas</cptr> and begin navigation at that point. <h1 id=addon>The extend and addon interfaces <p> This &text; does not define facilities for specifying style in the body of a conforming document. However, authors of simple documents may wish to make use of such a facility and suppliers of user agents may wish to add the facility to the ISO-HTML document architecture. This is possible, and conveniently done using one of the <hp1>extend</hp1> or <hp1>addon</hp1> interfaces. Many extensions and addons may be defined in the future: this clause defines the general <hp1>extend</hp1> and <hp1>addon</hp1> interface mechanism and provides an example of its use. <p> There are three types of interfaces available to document designers: <dl> <dt>New element interface <dd>The &head.extend; and &body.extend; interfaces which allow the addition of new elements and their attributes. <dt>Additional attribute interface <dd>The &a.addons; interfaces which allow the addition of further attributes to existing elements. <dt>Notation interface <dd>The use of <hp1>notations</hp1> as defined by ISO 8879:1986. </dl> <note> This &text; does not define any uses of the <hp1>extend</hp1> or <hp1>addon</hp1> interfaces, although an example is given to assist readers in their understanding of the SGML mechanisms used. </note> <note>&NB; This mention of the use of SGML notations does not extend the proposed ISO-HTML standard. The full power of SGML is available to all conforming applications [see Goldfarb 15.2.2 p.480]. </note> <h2 id=addon.conf>Conformance of extensions <p> No <hp1>extend</hp1> or <hp1>addon</hp1> shall cause a document to be non-conforming if that document is conforming in the absence of the <hp1>extend</hp1> or <hp1>addon</hp1>. <note> This &text; does not address the question of interdependencies between possibly incompatible extensions defined by other specifications. </note> <h2 id=extend>The new element interface <p> The &head.extend; and &body.extend; interfaces provide a means of extending the architecture of a document to include additional elements and attributes. The &head.extend; interface allows additions within the head of a document, at any point at which the &meta; element may be used. The &body.extend; interface allows additions within the body of a document at any point where an &a; element might be used. The technique used is to extend the content models which contain the &meta; and &a; elements. <p> By default the interfaces contain a &nop; element which has no semantic value, but the interface may be redefined to include an additional DTD fragment in the document architecture. <p> The procedure for adding new elements is as follows: <ol> <li>Place the definitions of the new elements and their attributes in a DTD fragment and assign a public text identifier. The definitions in this DTD fragment may make use of the entities defined in the ISO-HTML entity set. <note> The fragment may be in a single file, or in two or more files, for example one file for parameter entity definitions and another for element definitions. </note> <li>Place a new definition for the &head.extend; or &body.extend; parameter entity in the document type declaration subset. The definition should provide a content model for the new elements. Typically it is of the form <cptr>ELEMENT1 | ELEMENT2 ...</cptr>. The definition will override the default definition &nop; provided by the &text; DTD entity set [see Goldfarb p.404 line 32]. More than one extension may be defined in the &head.extend; or &body.extend; parameter entities with declarations of the form <cptr>EXTEND1.EL1 | EXTEND1.EL2 | ... | EXTEND2.EL1 ...</cptr>. <li>Place a parameter entity definition (or definitions) for the DTD fragment(s) containing definitions of the new elements and attributes in the document type declaration subset. <li>Invoke the &head.extend; or &body.extend; parameter entity. <li>Invoke the parameter entities for the DTD fragments at the end of the document type declaration subset. </ol> <h2 id=a.addon>The additional attribute interface <p> The &a.addons; interfaces provide a means of extending the set of attributes of an existing element. By default, the interfaces contain an empty string, ie. no additional attributes, but by redefining the interface, new attributes may be added to extend the document architecture. <p> The procedure for adding new attributes to an existing element X is as follows: <ol> <li>Place the definitions of the new attributes for the X element in an &X.addon; parameter entity in a DTD fragment and assign it a public text identifier. The definitions in this DTD fragment may make use of the entities defined in the ISO-HTML entity set. <note> This DTD fragment may be shared with the one defined in <hdref refid=extend page=yes>. </note> <p> This definition will override the default definition provided by the &text; DTD entity set [see Goldfarb p.404 line 32]. <li>Place a parameter entity definition for the DTD fragment in the document type declaration subset. <li>Invoke the parameter entities for the DTD fragments at the end of the document type declaration subset. </ol> <h2 id=subset>Use of the document type declaration subset <p> The extension techniques provided by this &text; make use of the document type declaration subset which is a part of all SGML documents. However many historic World Wide Web browsers do not process the subset, so it is recommended that for wide public use, the designer of an extension consider repackaging the extension with the parameter entity set and the document type definition provided by this &text; as a single entity (file) which can be referenced by a suitable public text identifier in the DOCTYPE declaration. <p> Such a repackaging will reduce the risk of error by users of the extension, and will allow the authority defining the package to provide better documentation and quality assurance, particularly when the package contains several extensions. <h2 id=nop.addon>The NOP attribute <p> A small number of the elements defined by this specification have a single &a.nop; attribute assigned to them. <p> This attribute is defined in the DTD to ensure that the DTD remains syntactically correct when the <hp1>addon</hp1> interface is not used. This attribute has no meaning and is not intended for use by authors or authoring systems. <h2 id=addonex>Example of addon definitions <p> <note>&NB; See the file <cptr>sample2.iso-html</cptr> for an example of a complete document type declaration subset. <p> These files will be cleaned up and placed in an informative annex. </note> <h1 id=a.box>Elements rendered as rectangles <p> The &boxing; elements defined by this &text; are usually rendered as rectangles with boxes around the visible area. Each of the elements which creates such a box is provided with an &a.align; attribute to specify how the rendered item is positioned with respect to the baseline of the text flow in which it occurs. A user agent may take into account only that text which has been placed on the line prior to the placement of the item in deciding the alignment of the enclosing box. <p> Each of the elements described in this clause use some of the attribute values from the following list: <dl> <dt>ALIGN=top <dd>Aligns the top of the rectangle with the top of the text rendered in the current line. <dt>ALIGN=texttop <dd>A permitted alternative specification for &a.align;=top. <dt>ALIGN=middle <dd>Aligns the middle of the rectangle with the current baseline. <dt>ALIGN=textmiddle <dd>Aligns the middle of the rectangle with a horizontal line drawn through the middle of the text rendered in the current line. <dt>ALIGN=bottom <dd>Aligns the bottom of the rectangle with the current baseline. <dt>ALIGN=baseline <dd>A permitted alternative specification for &a.align;=bottom. <dt>ALIGN=textbottom <dd>Aligns the bottom of the rectangle with a horizontal line drawn through the bottom of the text rendered in the current line. <dt>ALIGN=left <dd>Places the rectangle to the right of the current left margin, temporarily changing this margin. Subsequent text may be flowed along the rectangle's righthand side. <dt>ALIGN=center <dd>Places the rectangle after the end of the current line, centered between the left and right margins. Subsequent text starts at the beginning of the next line. <dt>ALIGN=centre <dd>A permitted alternative specification for &a.align;=center. <dt>ALIGN=right <dd>Places the rectangle to the left of the current right margin, temporarily changing this margin so that subsequent text is flowed along the rectangle's lefthand side. </dl> <note> The &caption; element and the justifying elements <hdref refid=a.justify page=yes> also have an &a.align; attribute, but in these cases its meaning is different. </note> <h1 id=a>The A element — Source and target anchors <h2 id=a.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT A - - (%text;)* -(A) > <!ATTLIST A CHARSET NAME #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED HREF %URL; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME CDATA #IMPLIED REL CDATA #IMPLIED REV CDATA #IMPLIED TITLE CDATA #IMPLIED %a.addon; > </cptr> <note>&NB; The attributes &a.rel; and &a.rev; are declared by &dave; to have type CDATA but &dan; declares them as NAMES. Which do we choose? </note> <h2 id=a.dsc>Description <p> The &a; element defines hyperlink source and target anchors (see clause <hdref refid=linking page=yes>). A source anchor requires the &a.href; attribute which provides the address of the target, while a target anchor requires the &a.name; attribute which identifies it as a location that may be used as a target in source anchors. At least one of the &a.href and &a.name attributes shall be present. <p> Good practice dictates that anchors should not be attached to markup, thus <cptr><H1><A NAME="xxx">Heading</A></H1></cptr> is preferred to <cptr><A NAME="xxx"><H1>Heading</H1></a></cptr>. <note> In the example given, the deprecated usage is also forbidden by the DTD, since it would not respect the required correct nesting of sections. </note> <p> Both start and end tags are required. <h2 id=a.att>Attributes <p> The attributes of the &a; element are: <dl> <dt>CHARSET <dd>A hint to the user agent concerning the character encoding scheme used by the resource pointed to by the link. It should be the appropriate value of the MIME charset parameter for that resource <bibref refid=mime>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>HREF <dd>Gives the URL of the source anchor of a hyperlink. For example: <cptr keep=off type=source width=column> The way to <A HREF="hands-on.html">happiness</A>. </cptr> <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>Gives the name of the anchor, and makes it available as a target for a hyperlink. The value shall be unique for the scope of the current document. For example: <cptr keep=off type=source width=column> <H2><A NAME=mit>Hacker's Paradise</A></H2> </cptr> <dt>REL <dd>The &a.rel; attribute gives the relationship(s) described by the hyperlink. The value is a whitespace separated list of relationship names. The semantics of link relationships are not specified in this document. <dt>REV <dd>Same as the &a.rel; attribute, but the semantics of the relationship are in the reverse direction. A link from A to B with <cptr>REL="X"</cptr> expresses the same relationship as a link from B to A with <cptr>REV="X"</cptr>. An anchor may have both &a.rel; and &a.rev; attributes. <dt>TITLE <dd>Suggests a title for the destination resource — advisory only. The &a.title; attribute may be used: <ul> <li>for display prior to accessing the destination resource, for example, as a margin note or in a small box while the mouse is over the anchor, or while the document is being loaded; <li>for resources that do not include a title, such as graphics, plain text and Gopher menus, for use as a window title. </ul> </dl> <h2 id=a.ex>Example <cptr keep=off type=source width=column> ... Please see <A NAME="intro" HREF="#details">here</A> for further details, and then <A HREF="http://www.acme.firm">go here</A> to apply for the job. ... <H3><A NAME="details">Job description</A></H3> ... <P> <A HREF="#intro">Return to introduction.</A> </cptr> <h1 id=address>The ADDRESS element — Author's address <h2 id=address.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % justify "(left|center|centre|right|justify)" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ADDRESS - - (%text;|P)+ > <!ATTLIST ADDRESS ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %address.addon; > </cptr> <h2 id=address.dsc>Description <p> The &address; element contains such information as address, signature and authorship, often at the beginning or end of the body of a document. <p> The &address element requires start and end tags. User agents should render the content with paragraph-breaks before and after. In the absence of a style-sheet, the &address; element should be rendered in an italic typeface and may be indented. <h2 id=address.att>Attributes <p> The &address; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=address.ex>Example <p> <cptr keep=off type=source width=column> <ADDRESS> Madame l'Éditrice en Chef<BR> Le Clarion du Pays<BR> 1, rue Rotative<BR> Ruritanie<BR> Téléphone: +800 l2 34 56 789 </ADDRESS> </cptr> <h1 id=area>The AREA element — Region in image map <h2 id=area.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % shape "(rect|circle|poly)" > <!ELEMENT AREA - O EMPTY > <!ATTLIST AREA ALT CDATA #REQUIRED COORDS CDATA #IMPLIED HREF %URL; #IMPLIED NOHREF (nohref) #IMPLIED SHAPE %shape; rect %area.addon; > </cptr> <h2 id=area.dsc>Description <p> The &area; element describes a region in an image map. It has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=area.att>Attributes <p> The &area; element has the following set of attributes: <dl> <dt>ALT <dd>Specifies a text label which may be displayed in a status line (when a pointing device is moved over the corresponding region in an image), or used for constructing a textual menu for non-graphical user agents. Authors are strongly recommended to provide meaningful &a.alt; attributes to support interoperability with speech-based or text-only agents. <dt>COORDS <dd>Specifies the size and position of a region in an image. See the &a.shape; attribute for further details. <dt>HREF <dd>Specifies the target hypertext link (behaviour) associated with a region. <dt>NOHREF <dd>Specifies that there is no target hypertext link (behaviour) associated with a region. This allows authors to create holes in regions. <dt>SHAPE <dd>Specifies the shape of a region and takes one of the following values: <dl> <dt>SHAPE=rect <dd>The region is rectangular and the &a.coords; attribute specifies the position of its top left and bottom right corners: &a.coords;="<hp1>left-x, top-y, right-x, bottom-y</hp1>", measured (in pixels) from the top left corner of the image. <dt>SHAPE=circle <dd>The region is circular and the &a.coords; attribute specifies the position of its centre and the length of its radius: &a.coords;="<hp1>centre-x,centre-y,radius</hp1>". <dt>SHAPE=poly <dd>The region is a polygon and the &a.coords; attribute specifies the position of its corners listed in clockwise order: &a.coords;="<hp1>x1,y1, x2,y2, ...</hp1>". </dl> If an <hp1>x</hp1> or <hp1>y</hp1> value is given with a percent sign as suffix, the value will be interpreted as a percentage of the image's width or height, respectively. For example, <cptr>SHAPE=rect COORDS="0,0, 50%,100%"</cptr>. </dl> <note>One of &a.href; or &a.nohref; shall be specified.</note> <h2 id=area.ex>Example <cptr keep=off type=source width=column> <AREA NOHREF SHAPE=circle COORDS="50%,50%,3" ALT="Bulls Eye"> </cptr> <p> See also the example of a &e.map; element <hdref refid=map page=yes>. <h1 id=b>The B element — Bold character style <h2 id=b.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "B | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=b.dsc>Description <p> The contents of a &b; element should be rendered in a bold text style. <p> Both start and end tags are required. <h2 id=b.att>Attributes <p> The &b; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=b.ex>Example <p> <cptr keep=off type=source width=column> Thank you for <B>NOT SMOKING</B> </cptr> <h1 id=base>The BASE element — Base URL specification <h2 id=base.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % URL "CDATA" > <!ELEMENT BASE - O EMPTY > <!ATTLIST BASE HREF %URL; #REQUIRED %base.addon; > </cptr> <h2 id=base.dsc>Description <p> The &base; element provides a base URL for dereferencing relative URLs, using the rules given by the URL specification <bibref refid=url>. <p> In the absence of a &base; element the document URL should be used. <note> This is not necessarily the same as the URL used to request the document, as the base URL may be overridden by an HTTP header accompanying a document. </note> <note>&NB; This clause appears to contain a normative reference to RFC 1738. How should this be handled? </note> <p> The start tag is required, but the end tag shall be omitted. <h2 id=base.att>Attributes <p> The &base; element has the following required attribute: <dl> <dt>HREF <dd>The value is a Uniform Resource Locator as defined by IETF RFC 1808 and IETF RFC 1738. </dl> <h2 id=base.ex>Example <p> Given the &base; specification <cptr keep=off type=source width=column> <BASE HREF="http://www.acme.com/intro.html"> </cptr> the image refernced by <cptr keep=off type=source width=column> <IMG SRC="icons/logo.jpeg"> </cptr> is dereferenced to <cptr keep=off type=source width=column> http://www.acme.com/icons/logo.jpeg </cptr> <h1 id=bdo>The BDO element — Bidirectional override <h2 id=bdo.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BDO - - (%text;)+ > <!ATTLIST BDO DIR %dirn; #REQUIRED LANG NAME #IMPLIED %bdo.addon; > </cptr> <h2 id=bdo.dsc>Description <p> The &bdo; element contains unusual pieces of text in which directionality cannot be resolved from the context in an unambiguous fashion. For example in part numbers, formulas, telephone numbers, punctuation and other texts whose directionality cannot be determined by context alone. <p> Both start and end tags are required. <h2 id=bdo.att>Attributes <p> The &bdo; element has the following attributes: <dl> <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=bdo.ex>Example <p> <cptr keep=off type=source width=column> <p>A palindrome appears the same whether rendered left-to-write: <BDO DIR=ltr>evil rats on no star live</BDO> or right-to-left: <BDO DIR=rtl>evil rats on no star live</BDO>. </cptr> <h1 id=big>The BIG element — Big character style <h2 id=big.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | BIG | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=big.dsc>Description <p> The contents of a &big; element should be rendered in a larger font than would otherwise be used. <p> Both start and end tags are required. <h2 id=big.att>Attributes <p> The &big; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=big.ex>Example <cptr keep=off type=source width=column> <BIG>What the large print giveth,</BIG> <SMALL>the small print taketh away.</SMALL> </cptr> <h1 id=blockquote>The BLOCKQUOTE element — Block quotation <h2 id=blockquote.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BLOCKQUOTE - - %section.content; > <!ATTLIST BLOCKQUOTE ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %blockquote.addon; > </cptr> <h2 id=blockquote.dsc>Description <p> The &blockquote; element contains a quotation. A typical rendering might be with a slight left and right indent, and/or an italic font. The &blockquote; usually provides space above and below the quoted text. <p> Single font renditions may reflect the quotation style on Internet mail by putting a vertical line of graphic characters, such as the greater than symbol (>), in the margin. <p> Both start and end tags are required. <h2 id=blockquote.att>Attributes <p> The &blockquote; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=blockquote.ex>Example <cptr keep=off type=source width=column> <BLOCKQUOTE>Afin de réaliser les objectifs visés à l'article 129B, la Communauté : <p> met en oeuvre toute action qui peut s'avérer nécessaire pour assurer l'interoperabilité des réseaux, en particulier dans le domaine de l'harmonisation des normes techniques ; </BLOCKQUOTE> Traité sur l'Union Européenne, Article 129 C. </cptr> <h1 id=body>The BODY element — Document body <h2 id=body.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BODY - - (%section.content;, (H1,B1)* ) > <!ATTLIST BODY CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %body.addon; > </cptr> <h2 id=body.dsc>Description <p> The &body; element contains the textual flow of the document. <p> Both start and end tags are required. <h2 id=body.att>Attributes <p> The &body; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=body.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &body; element. <h1 id=br>The BR element — Line break <h2 id=br.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT BR - O EMPTY > <!ATTLIST BR NOP (nop) #IMPLIED %br.addon; > </cptr> <h2 id=br.dsc>Description <p> The &br; element specifies a line break between words. There is no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=br.att>Attributes <p> This &text; defines no attributes for the &br; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h2 id=br.ex>Example <cptr keep=off type=source width=column> <ADDRESS> ISO/IEC Copyright Office<BR> P.O. Box 56<BR> 1211 Geneva 20<BR> Switzerland </ADDRESS> </cptr> <h1 id=b1>The B1 element — Major section body <h2 id=b1.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B1 O O (%section.content;, (H2,B2)* ) > <!ATTLIST B1 NOP (nop) #IMPLIED %b1.addon; > </cptr> <h2 id=b1.dsc>Description <p> The &b1; element provides a formal definition for the body of a major section. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b1.att>Attributes <p> This &text; defines no attributes for the &b1; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b2>The B2 element — Section body <h2 id=b2.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B2 O O (%section.content;, (H3,B3)* ) > <!ATTLIST B2 NOP (nop) #IMPLIED %b2.addon; > </cptr> <h2 id=b2.dsc>Description <p> The &b2; element provides a formal definition for the body of a section. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b2.att>Attributes <p> This &text; defines no attributes for the &b2; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b3>The B3 element — Subsection body <h2 id=b3.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B3 O O (%section.content;, (H4,B4)* ) > <!ATTLIST B3 NOP (nop) #IMPLIED %b3.addon; > </cptr> <h2 id=b3.dsc>Description <p> The &b3; element provides a formal definition for the body of a subsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b3.att>Attributes <p> This &text; defines no attributes for the &b3; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b4>The B4 element — Subsubsection body <h2 id=b4.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B4 O O (%section.content;, (H5,B5)* ) > <!ATTLIST B4 NOP (nop) #IMPLIED %b4.addon; > </cptr> <h2 id=b4.dsc>Description <p> The &b4; element provides a formal definition for the body of a subsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b4.att>Attributes <p> This &text; defines no attributes for the &b4; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b5>The B5 element — Subsubsubsection body <h2 id=b5.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B5 O O (%section.content;, (H6,B6)* ) > <!ATTLIST B5 NOP (nop) #IMPLIED %b5.addon; > </cptr> <h2 id=b5.dsc>Description <p> The &b5; element provides a formal definition for the body of a subsubsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b5.att>Attributes <p> This &text; defines no attributes for the &b5; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b6>The B6 element — Minor subsubsubsection body <h2 id=b6.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B6 O O (%section.content;) > <!ATTLIST B6 NOP (nop) #IMPLIED %b6.addon; > </cptr> <h2 id=b6.dsc>Description <p> The &b6; element provides a formal definition for the body of a minor subsubsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b6.att>Attributes <p> This &text; defines no attributes for the &b6; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=caption>The CAPTION element — Table caption <h2 id=caption.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT CAPTION - - (%text;)+ > <!ATTLIST CAPTION ALIGN (top|bottom) #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %caption.addon; > </cptr> <h2 id=caption.dsc>Description <p> The &caption; element is used only in tables, where it contains a caption for a table. <p> Both start and end tags are required. <h2 id=caption.att>Attributes <p> The &caption; element has the following attributes: <dl> <dt>ALIGN <dd>The &a.align; attribute has value <cptr>top</cptr> or <cptr>bottom</cptr> and indicates the preferred placement of the caption with respect to the table. If the &a.align; attribute is omitted, browsers usually place the caption at the top of the table. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <note> &caption; is not a justifying element. Authors requiring horizontal text alignment should consider including a justifying element such as &p; within the content of the &caption;. </note> <h2 id=caption.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &caption; element. <h1 id=cite>The CITE element — Citation <h2 id=cite.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "CITE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=cite.dsc>Description <p> The &cite; element contains a citation or a reference to other sources. <p> Both start and end tags are required. <h2 id=cite.att>Attributes <p> The &cite; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=cite.ex>Example <cptr><CITE LANG="en-US">The buck stops here.</CITE></cptr> <h1 id=code>The CODE element — Program code <h2 id=code.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | CODE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=code.dsc>Description <p> The &code; element contains extracts from program code. <p> Both start and end tags are required. <h2 id=code.att>Attributes <p> The &code; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <note> The &a.lang; attribute is used for specifying human languages, not computer languages. </note> </dl> <h2 id=code.ex>Example <p> The list append function may be coded elegantly in Prolog as follows: <cptr keep=off type=source width=column> <CODE>append([],L,L) .</CODE> <CODE>append([X|L1],L2,[X|L3]) :- append(L1,L2,L3) .</CODE> </cptr> <h1 id=dd>The DD element — Definition data <h2 id=dd.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DD - O %section.content; > <!ATTLIST DD CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dd.addon; > </cptr> <h2 id=dd.dsc>Description <p> The ⅆ element is used only in definition lists <hdref refid=dl page=yes> and contains a definition for a term marked with a &dt; element. The definition is usually formatted as an indented paragraph after the term. <p> The start tag is required, but the end tag may be omitted. <h2 id=dd.att>Attributes <p> The ⅆ element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dd.ex>Example <p> See <hdref refid=dl page=yes> for an example of the use of the ⅆ element. <h1 id=dfn>The DFN element — Defining instance <h2 id=dfn.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | DFN | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=dfn.dsc>Description <p> The contents of the &dfn; element are the defining instance of a term. <p> Both start and end tags are required. <h2 id=dfn.att>Attributes <p> The &dfn; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dfn.ex>Example <p> <cptr keep=off type=source width=column> ICAO <DFN>International Civil Aviation Organization</DFN> responsible ... </cptr> <h1 id=div>The DIV element — Document division <h2 id=div.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT DIV - - %section.content; > <!ATTLIST DIV ALIGN %justify; #IMPLIED %div.addon; > </cptr> <h2 id=div.dsc>Description <p> The ÷ element may be used to structure documents into divisions. This &text; makes no recommendation for the presentation of divisions. <p> Both start and end tags are required. <h2 id=div.att>Attributes <p> The ÷ element has the following attribute: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. </dl> <h2 id=div.ex>Example <cptr keep=off type=source width=column> <BODY> ... <DIV> English Dictionary Terms <DL> ... </DL> </DIV> <DIV> Japanese Dictionary Terms <DL> ... </DL> </DIV> ... </BODY> </cptr> <h1 id=dl>The DL element — Definition list <h2 id=dl.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DL - - (DT|DD)+ > <!ELEMENT DT - O (%text;)+ > <!ELEMENT DD - O %section.content; > <!ATTLIST DL CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dl.addon; > </cptr> <h2 id=dl.dsc>Description <p> The &dl; element provides structure for a definition list: a list of terms and corresponding definitions. <p> The content is a non empty sequence of intermixed &dt; and ⅆ elements, usually in pairs. Multiple &dt; may be paired with a single ⅆ element and vice versa. <p> Definition lists are usually formatted with the term flushed to the margin and the definition, formatted paragraph style, indented to follow the term. If the &dt; term does not fit in the &dt; column (often one third of the display area), it may be extended across the page with the ⅆ content moved onto the next line, or it may be wrapped onto successive lines in the &dt; column. <p> Both start and end tags are required. <h2 id=dl.att>Attributes <p> The &dl; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &dt;, ⅆ groupings. The &a.compact; attribute may also imply a reduction of the width of the &dt; column. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dl.ex>Example <cptr keep=off type=source width=column> <DL compact> <DT>Center <DT>Centre <DD>A point equidistant from all points on the surface of a sphere <DD>In some field sports the player who holds the middle position on the field, court, or forward line </DL> </cptr> <h1 id=dt>The DT element — Definition term <h2 id=dt.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DT - O (%text;)+ > <!ATTLIST DT CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dt.addon; > </cptr> <h2 id=dt.dsc>Description <p> The &dt; element is used only in definition lists <hdref refid=dl page=yes> and contains a term to be defined. The term is typically formatted flush to the margin. If the &dt; term does not fit in the &dt; column (often one third of the display area), it may be extended across the page with the ⅆ content moved onto the next line, or it may be wrapped onto successive lines in the &dt; column. <p> The start tag is required, but the end tag may be omitted. <h2 id=dt.att>Attributes <p> The &dt; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dt.ex>Example <p> See <hdref refid=dl page=yes> for an example of the use of the &dt; element. <h1 id=em>The EM element — Emphasized text <h2 id=em.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | EM | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=em.dsc>Description <p> The contents of the &em; element should be emphasized and are usually rendered in italics. <p> Both start and end tags are required. <h2 id=em.att>Attributes <p> The &em; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=em.ex>Example <cptr>Trespassers <EM>will</EM> be prosecuted!</cptr> <h1 id=form>The FORM element — Forms <h2 id=form.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % form.fields "INPUT | SELECT | TEXTAREA" > <!ENTITY % form.text "%text; | %form.fields;" > <!ENTITY % form.content "(%block; | %form.text;)+" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT FORM - - %form.content; -(FORM) > <!ATTLIST FORM ACTION %URL; #REQUIRED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ENCTYPE %Content-Type; "application/x-www-form-urlencoded" ID ID #IMPLIED LANG NAME #IMPLIED METHOD %HTTP-Method; get %form.addon; > </cptr> <h2 id=form.dsc>Description <p> The &form; element provides a template for a form data set and an associated method and action URL. A form data set is a sequence of name/value pair fields. The names are specified in the &a.name; attributes of form input elements, and the values are provided with initial values by various forms of markup, and may subsequently be edited by the user. The resulting form data set is used to access an information service as defined by the &a.action; and &a.method; attributes. <p> Form elements can be mixed in with document structuring elements. For example, a &form; element may contain lists which contain &input; elements. This gives considerable flexibility in designing the layout of forms. &form; elements do not contain other &form; elements. <p> Usually &form; elements contain a sequence of &input; elements along with additional document structure. <p> Both start and end tags are required. <h3 id=form.submit>Form submission <p> A user agent usually begins processing by presenting a document with the fields in their initial state. The user is allowed to modify the content of fields, constrained by the field type. When the user indicates that the form should be submitted (using for example a submit button), the form data set is processed as defined by its &a.method;, &a.action; URL and &a.enctype; attributes. <p> Where there is only one single-line text input field in a form, the user agent should accept "Enter" in that field as a request to submit the form. <h2 id=form.att>Attributes <p> The attributes of the &form; element are: <dl> <dt>ACTION <dd>Specifies the action URL for the form. The action URL of a form defaults to the base URL of the document <hdref refid=linking page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ENCTYPE <dd>Specifies the media type used to encode the name/value pairs for transport, in case the protocol itself does not impose a format. The default encoding for all forms is <cptr>application/x-www-form-urlencoded</cptr> A form data set is represented in this media type as follows: <ol> <li>The form field names and values are escaped; spaces are replaced by `<cptr>+</cptr>', and then reserved characters are escaped as per RFC 1738 <bibref refid=url>; that is non-alphanumeric characters are replaced by <cptr>%HH</cptr>, a percent sign and two hexadecimal digits representing the ASCII code of the character. Line breaks, as in multi-line text field values, are represented as CRLF pairs, i.e. %0D%0A. <note>&NB; How do we handle the reference to "ASCII"? </note> <li>The fields are listed in the order they appear in the document with the name separated from the value by `<cptr>=</cptr>' and the pairs separated from each other by `<cptr>;</cptr>'. Fields with null values may be omitted. In particular, unselected radio buttons and checkboxes should not appear in the encoded data, but hidden fields with &a.value; attributes present should. <note>&NB; RFC1866 allows the use of `<cptr>&</cptr>' instead of `<cptr>;</cptr>', but notes that this leads to ambiguities, since `<cptr>&</cptr>' is an SGML entity reference delimiter, and encourages the use of `<cptr>;</cptr>'. </note> </ol> <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>METHOD <dd>Selects a method for accessing the action URL. The set of applicable methods is a function of the scheme of the action URL specified for the form <hdref refid=form.submit page=yes>. The &a.method; attribute has one of the following values: <dl> <dt>METHOD=get — Forms with no server side effects <dd>If the processing of a form is idempotent (ie. it has no lasting observable effect on the state of the world), then the form method should be <cptr>get</cptr>. Many database searches have no visible side-effects and make ideal applications for query forms. <p> To process a form whose action URL is an HTTP URL and whose method is <cptr>get</cptr>, the user agent starts with the action URL and appends a `<cptr>?</cptr>' and the form data set, in <cptr>application/x-www-form-urlencoded</cptr> format. The user agent then traverses the link to this URL as if it were an anchor. <note> The URL encoding may result in very long identifiers, which cause some historical HTTP server implementations to exhibit defective behaviour. As a result, some ISO-HTML forms may be written using <cptr>METHOD=post</cptr> even though the form submission has no server side effects. </note> <dt>METHOD=post — Forms with server side effects <dd>If the service associated with the processing of a form has side effects (for example modification of a data base or subscription to a service), the method should be `post'. <p> To process a form whose action URL is an HTTP URL and whose method is <cptr>post</cptr>, the user agent conducts an HTTP `post' transaction using the action URL, and a message body of type <cptr>application/x-www-form-urlencoded</cptr> format. The user agent should display the response from the HTTP `post' interaction just as it would display the response from an HTTP `get'. </dl> </dl> <h2 id=form.ex>Example <cptr keep=off type=source width=column> <H1>Acme Company Inc</H1> <H2>Smallsville Agency</H2> <FORM METHOD=post ACTION="http://www.acme.firm/job"> <B>Job Application Form</B> <BR> Your Name: <INPUT TYPE=text NAME=name VALUE="Yosemite Sam"> <BR> Do you have a valid driver licence <BR> <INPUT TYPE=radio NAME=driver VALUE=yes>Yes <INPUT TYPE=radio NAME=driver VALUE=no CHECKED>No <BR> When you are finished, you may submit this application: <INPUT TYPE=submit VALUE="Apply"> <BR> You may clear the application form and start over at any time: <INPUT TYPE=reset VALUE="New Form"> <BR> <EM>Acme is an equal opportunity employer<EM> </FORM> </cptr> <h1 id=head>The HEAD element — Document header <h2 id=head.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT HEAD - - (TITLE & ISINDEX? & BASE?) +(LINK | META | %head.extend;) > <!ATTLIST HEAD CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %head.addon; > </cptr> <h2 id=head.dsc>Description <p> The &head; element defines the head of an ISO-HTML document, and contains an unordered collection of general information about the document. <p> Both start and end tags are required. <p> The content model of the &head; element also provides the &head.extend; interface. See <hdref refid=extend page=yes>. <h2 id=head.att>Attributes <p> The &head; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=head.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &head; element. <h1 id=hr>The HR element — Horizontal rule <h2 id=hr.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT HR - O EMPTY > <!ATTLIST HR ALIGN %justify; #IMPLIED DIR %dirn; #IMPLIED %hr.addon; > </cptr> <h2 id=hr.dsc>Description <p> The &hr; element specifies a horizontal separator in a document. Visual user agents usually present it as a full width horizontal rule, while speech-based user agents might present it as a pause. <p> The start tag is required, but the end tag shall be omitted. <h2 id=hr.att>Attributes <p> The &hr; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. </dl> <h2 id=hr.ex>Example <cptr keep=off type=source width=column> ... <H5>Monday Activities</H5> <P> ... <HR> <H5>Tuesday Activities</H5> <P> ... <HR> ... </cptr> <h1 id=h1>The H1 element — Major section header <h2 id=h1.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H1 - - (%text;)* > <!ATTLIST H1 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h1.addon; > </cptr> <h2 id=h1.dsc>Description <p> The &h1; element specifies the beginning of a major section of a document and contains the title of that major section. <p> Both start and end tags are required. <p> This &text; requires the correct nesting of sections. The &h1; element shall not be followed by an &h3;, &h4;, &h5;, or &h6; element without an intervening &h2; element. <h2 id=h1.att>Attributes <p> The &h1; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h1.ex>Example <cptr keep=off type=source width=column> <H1>Continent</H1> ... <H2>Country</H2> ... <H3>Province</H3> ... <H4>County</H5> ... <H5>City</H4> ... <H3>State</H3> ... <H4>City</H4> ... </cptr> <h1 id=h2>The H2 element — Section header <h2 id=h2.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H2 - - (%text;)* > <!ATTLIST H2 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h2.addon; > </cptr> <h2 id=h2.dsc>Description <p> The &h2; element specifies the beginning of a section of a document and contains the title of the section. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h2; element shall not be followed by an &h4;, &h5;, or &h6; element without an intervening &h3; element. An &h2; element shall be preceded by an &h1; element. <h2 id=h2.att>Attributes <p> The &h2; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h2.ex>Example <p> See <hdref refid=h1 page=yes> for an example of the use of header elements. <h1 id=h3>The H3 element — Subsection header <h2 id=h3.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H3 - - (%text;)* > <!ATTLIST H3 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h3.addon; > </cptr> <h2 id=h3.dsc>Description <p> The &h3; element specifies the beginning of a subsection of a document and contains the title of the subsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h3; element shall not be followed by an &h5; or &h6; element without an intervening &h4; element. An &h3; element shall be preceded by an &h1; element. <h2 id=h3.att>Attributes <p> The &h3; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h3.ex>Example <p> See <hdref refid=h1 page=yes> for an example of the use of header elements. <h1 id=h4>The H4 element — Subsubsection header <h2 id=h4.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H4 - - (%text;)* > <!ATTLIST H4 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h4.addon; > </cptr> <h2 id=h4.dsc>Description <p> The &h4; element specifies the beginning of a subsubsection of a document and contains the title of the subsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h4; element shall not be followed by an &h6; element without an intervening &h5; element. An &h4; element shall be preceded by an &h1; element. <h2 id=h4.att>Attributes <p> The &h4; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h4.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=h5>The H5 element — Subsubsubsection header <h2 id=h5.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H5 - - (%text;)* > <!ATTLIST H5 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h5.addon; > </cptr> <h2 id=h5.dsc>Description <p> The &h5; element specifies the beginning of a subsubsubsection of a document and contains the title of the subsubsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. An &h5; element shall be preceded by an &h1; element. <h2 id=h5.att>Attributes <p> The &h5; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h5.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=h6>The H6 element — Minor subsubsubsection header <h2 id=h6.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H6 - - (%text;)* > <!ATTLIST H6 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h6.addon; > </cptr> <h2 id=h6.dsc>Description <p> The &h6; element specifies the beginning of a minor subsubsubsection of a document and contains the title of the minor subsubsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. An &h6; element shall be preceded by an &h1; element. <h2 id=h6.att>Attributes <p> The &h6; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h6.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=i>The I element — Italic character style <h2 id=i.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | I | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=i.dsc>Description <p> The contents of an &i; element should be rendered in an italic text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=i.att>Attributes <p> The &i; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=i.ex>Example <cptr>Following her <I LANG=fr>affaire</I> with the well known rugby player, ...</cptr> <h1 id=img>The IMG element — Inline images <h2 id=img.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT IMG - O EMPTY > <!ATTLIST IMG ALIGN (top|middle|bottom|left|right) #IMPLIED ALT CDATA #REQUIRED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED ISMAP (ismap) #IMPLIED LANG NAME #IMPLIED SRC %URL; #REQUIRED USEMAP %URL; #IMPLIED %img.addon; > </cptr> <h2 id=img.dsc>Description <p> The &img; element refers to an image or icon via a hyperlink, see <hdref refid=linking page=yes>. <p> Non-graphical user agents should process the value of the &a.alt; attribute as an alternative to processing the image resource indicated by the &a.src; attribute. <p> The &img; element has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=img.att>Attributes <p> The &img; attributes are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes> <dt>ALT <dd>The &a.alt; attribute is required, and specifies text that may be used in place of the referenced image resource. Use of this attribute is vital for interoperability with speech-based and text only user agents. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ISMAP <dd>If the &a.ismap; attribute is present in an &img; element, the &img; element shall be contained in an &a; element with an &a.href; attribute present. This construct represents a set of hyperlinks. The user can choose from the set by selecting a pixel on the image. The user agent computes the target URL by appending the character `<cptr>?</cptr> and the comma separated x and y coordinates of the selected pixel to the URL given in the &a; element. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>SRC <dd>The &a.src; attribute is required, and specifies the URL of the image resource. <note> In practice, the media types of the image resources are limited to a few raster graphic formats: typically <cptr>image/jpeg</cptr>, <cptr>image/gif</cptr>. In particular, <cptr>text/html</cptr> resources are not intended to be used for images. </note> <dt>USEMAP <dd>Specifies a URL fragment identifier for a client-side image map defined with the &e.map; element, see <hdref refid=map page=yes>. </dl> <h2 id=img.ex>Example <p> If the document contains: <cptr keep=off type=source width=column> ... <ISO-HTML><HEAD><TITLE>ImageMap Example</TITLE> <BASE HREF="http://host/index"></HEAD> <BODY> <P>Select a point of interest in the following image: <A HREF="/cgi-bin/imagemap"> <IMG ismap SRC="atlas.jpeg" ALT="Atlas of central Europe"> </A> ... </cptr> and the user selects the upper-leftmost pixel in the image, the chosen hyperlink has the URL <cptr>http://host/cgi-bin/imagemap?0,0</cptr> <h1 id=input>The INPUT element — User input field <h2 id=input.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % InputType "(checkbox|file|hidden|image| password|radio|reset|submit|text)" > <!ELEMENT INPUT - O EMPTY > <!ATTLIST INPUT CHECKED (checked) #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED MAXLENGTH NUMBER #IMPLIED NAME CDATA #IMPLIED SIZE CDATA #IMPLIED SRC %URL; #IMPLIED TYPE %InputType; text VALUE CDATA #IMPLIED %input.addon; > </cptr> <h2 id=input.dsc>Description <p> The &input; element is used in forms and represents a field for user input. <p> The start tag is required, but the end tag shall be omitted. <h2 id=input.att>Attributes <p> The &a.type; attribute of the &input; element discriminates between several different types of input field. The set of applicable attributes depends on the value of the &a.type; attribute as specified in the following subclauses. By default the value of the &a.type; attribute is TEXT. <h3 id=i.checkbox>TYPE=CHECKBOX <p> An &input; element with TYPE=CHECKBOX specifies a boolean choice. A set of &input; elements in the same &form; element with the same &a.name; attribute value represents an n-of-many choice. <p> The other attribute values are as follows: <dl> <dt>CHECKED <dd>Used to mark the corresponding input item as initially selected. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies the value to be returned if this input item is selected. </dl> <p> Example: <cptr keep=off type=source width=column> Which flavours do you like? <INPUT TYPE=CHECKBOX NAME=flavour VALUE=van CHECKED>Vanilla <BR> <INPUT TYPE=CHECKBOX NAME=flavour VALUE=str>Strawberry <BR> <INPUT TYPE=CHECKBOX NAME=flavour VALUE=pec CHECKED>Pecan <BR> </cptr> Note that more than one flavour may be checked. <h3 id=i.file>TYPE=FILE <p> An &input; element with TYPE=FILE provides a means for users to attach a file to a form's contents. It is typically rendered by a text field and an associated button which when selected invokes a file browser to select a file name. The file name can also be entered directly in the text field. See RFC 1867 for further details <bibref refid=file>. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Prefered size of the visible input line in average character widths. Users should be able to type more characters than this limit with the text scrolling through the field to keep the input cursor in view. </dl> <p> Example: <cptr keep=off type=source width=column> <FORM ACTION="http://server.rec/cgi/handle" ENCTYPE="multipart/form-data" METHOD=post> What is your name? <INPUT TYPE=TEXT NAME=submitter><BR> Which file are you sending? <INPUT TYPE=FILE NAME=pics> </FORM> </cptr> The user types "Joe Sixpack" in the name field, and selects an image file "football.jpeg" for the answer to 'Which file are you sending?'. <h3 id=i.hidden>TYPE=HIDDEN <p> An &input; element with TYPE=HIDDEN declares that fields should not be rendered — they are hidden from the user. The user does not interact with the field; instead, the &a.value; attribute specifies the value of the field. The &a.name; and &a.value; attributes are required, and are returned to the server when the form is submitted. <note> This input element may be used to handle state information in a form. </note> <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies a value for the field. </dl> <p> Example: <cptr keep=off type=source width=column> <INPUT TYPE=HIDDEN NAME=customer VALUE="827364"> </cptr> <h3 id=i.image>TYPE=IMAGE <p> An &input; element with TYPE=IMAGE specifies an image resource to display. If a point in the image is selected, the x and y coordinates of that point are returned to the server. The names of the fields used for returning the x and y pixel coordinates are the specified name of the field with ".x" and ".y" appended. <p> TYPE=IMAGE implies TYPE=SUBMIT processing, that is when a pixel is chosen, the form as a whole is submitted. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SRC <dd>This attribute is required and specifies the source of the image. </dl> <p> Example: <cptr keep=off type=source width=column> Choose a point on the map: <INPUT TYPE=IMAGE SRC="world.jpeg" NAME=place> </cptr> <p> When the user selects a point on the map, say with pixel coordinates (129,436), the input fields become <cptr>place.x=129,place.y=436</cptr>. See <hdref refid=form.submit page=yes>. <h3 id=i.password>TYPE=PASSWORD <p> An &input; element with TYPE=PASSWORD specifies a single line text field into which users may type a password. As the user types, the characters are usually echoed as `<cptr>*</cptr>' to hide the password from prying eyes. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Users should be able to type more characters than this limit with the password scrolling through the field to keep the input cursor in view. </dl> <p> Example: <cptr keep=off type=source width=column> Name: <INPUT NAME=login><> Password: <INPUT TYPE=PASSWORD SIZE=10 MAXLENGTH=8 NAME=passwd> </cptr> <h3 id=i.radio>TYPE=RADIO <p> An &input; element with TYPE=RADIO specifies a boolean choice. A set of &input; elements in a &form; element with the same &a.name; attribute value represents a 1-of-many choice. <p> The other attribute values are as follows: <dl> <dt>CHECKED <dd>Used to mark the corresponding input item as initially selected. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies the value to be returned if this input item is selected. </dl> <p> At all times one and only one of the radio buttons in a set is checked. Initially, if none of the &input; elements in a set of radio buttons specifies CHECKED, then the user agent shall mark the first radio button of the set as checked. <p> Example: <cptr keep=off type=source width=column> Which is your favourite flavour? <INPUT TYPE=RADIO NAME=flavour VALUE=van>Vanilla <BR> <INPUT TYPE=RADIO NAME=flavour VALUE=str>Strawberry <BR> <INPUT TYPE=RADIO NAME=flavour VALUE=pec CHECKED>Pecan <BR> </cptr> <h3 id=i.reset>TYPE=RESET <p> An &input; element with TYPE=RESET specifies an input option, usually represented by a button, that instructs the user agent to reset the form's fields to their initial states. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALUE <dd>Indicates a label for the input field. This value is usually displayed on a button. </dl> <p> Example: <cptr keep=off type=source width=column> When you are finished, you may submit this request: <INPUT TYPE=SUBMIT><BR> You may clear the form and start over at any time: <INPUT TYPE=RESET VALUE="Reset"> </cptr> <h3 id=i.submit>TYPE=SUBMIT <p> An &input; element with TYPE=SUBMIT represents an input option, typically a button, that instructs the user agent to submit the form. <p>The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>Indicates that this element contributes a form field whose value is given by the &a.value; attribute. If the &a.name; attribute is not present, this element does not contribute a form field. <dt>VALUE <dd>Indicates a label for the form field. This value is usually displayed on a button. </dl> <p> Example: <cptr keep=off type=source width=column> You may submit this request internally: <INPUT TYPE=SUBMIT NAME=request VALUE=internal> or externally: <INPUT TYPE=SUBMIT NAME=request VALUE=external> </cptr> <h3 id=i.text>TYPE=TEXT <p> An &input; element with TYPE=TEXT specifies a single line text field into which users may type a string. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Prefered size of the visible input line in average character widths. Users should be able to type more than this limit with the text scrolling through the field to keep the input cursor in view. <dt>VALUE <dd>This attribute is required and specifies an initial value for the text string initially shown in the field. </dl> <p> Example: <cptr keep=off type=source width=column> <INPUT TYPE=TEXT SIZE=40 NAME=user VALUE="Intentionally left blank"> </cptr> <h2 id=input.ex>Example <p> For examples of the use of the &input; element, please see subclauses <hdref refid=i.checkbox page=yes>, <hdref refid=i.file page=yes>, <hdref refid=i.hidden page=yes>, <hdref refid=i.image page=yes>, <hdref refid=i.password page=yes>, <hdref refid=i.radio page=yes>, <hdref refid=i.reset page=yes>, <hdref refid=i.submit page=yes> and <hdref refid=i.text page=yes>. <h1 id=isindex>The ISINDEX element — Keyword choice <h2 id=isindex.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ISINDEX - O EMPTY> <!ATTLIST ISINDEX CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED PROMPT CDATA #IMPLIED %isindex.addon; > </cptr> <h2 id=isindex.dsc>Description <p> The &isindex; element implies a set of hyperlinks. A user agent enables users to choose from this set by providing keywords. <note> User agents compute target URLs by appending the character `<cptr>?</cptr>' and the selected keywords to the base URL. Keywords are escaped according to RFC 1738 <bibref refid=url> and joined together by the character `<cptr>+</cptr>'. </note> <p> &form; elements with &a.method;<cptr>=get</cptr> also represent sets of hyperlinks. See clause <hdref refid=form page=yes> for details. <p> The start tag is required, but the end tag shall be omitted. <h2 id=isindex.att>Attributes <p> The &isindex; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>PROMPT <dd>A text which encourages a user to make a choice. </dl> <h2 id=isindex.ex>Example <p> If a document &head; element contains: <cptr keep=off type=source width=column> <BASE HREF="http://host/index"> <ISINDEX> </cptr> and a user selects the keywords <cptr>apple</cptr> and <cptr>berry</cptr>, then the user agent will access the resource <cptr keep=off type=source width=column> http://host/index?apple+berry </cptr> <h1 id=iso-html>The ISO-HTML element — Document instance <h2 id=iso-html.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ISO-HTML - - (HEAD, BODY) > <!ATTLIST ISO-HTML CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VERSION CDATA #FIXED "&version;" %iso-html.addon; > </cptr> <h2 id=iso-html.dsc>Description <p> The &iso-html; element contains a document instance marked up in the language specified by this &text;. <p> Both start and end tags are required. <h2 id=iso-html.att>Attributes <p> The &iso-html; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VERSION <dd>The &a.version; attribute of the &iso-html; document element may be omitted, but if specified shall have the value <cptr>&version;</cptr>. </dl> <h2 id=iso-html.ex>Example <cptr keep=off type=source width=column> <ISO-HTML> <HEAD> <TITLE>Scientific Units</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> <H1>Scientific units and their conversions<H1> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <h1 id=kbd>The KBD element — Keyboard input <h2 id=kbd.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | KBD | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=kbd.dsc>Description <p> The &kbd; element contains text to be typed by a user. <p> Both start and end tags are required. <h2 id=kbd.att>Attributes <p> The &kbd; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=kbd.ex>Example <cptr keep=off type=source width=column> At the prompt, you might try typing: <KBD>rm -rf *</KBD> </cptr> <h1 id=li>The LI element — List item <h2 id=li.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT LI - O %section.content; > <!ATTLIST LI ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VALUE NUMBER #IMPLIED %li.addon; > </cptr> <h2 id=li.dsc>Description <p> The &li; element contains a list item in an ordered list <hdref refid=ol page=yes> or an unordered list <hdref refid=ul page=yes>. <p> The start tag is required, but the end tag may be omitted. <h2 id=li.att>Attributes <p> The &li; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALUE <dd>This attribute is optional and shall only be used in ordered lists. It provides an integer number for the ordered list item and re-sequences the numbering of the items in the ordered list. </dl> <h2 id=li.ex>Example <p> See <hdref refid=ol page=yes> and <hdref refid=ul page=yes> for examples of the use of the &li; element. <h1 id=link>The LINK element — Interdocument relations <h2 id=link.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT LINK - O EMPTY > <!ATTLIST LINK CHARSET NAME #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED HREF %URL; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED REL CDATA #IMPLIED REV CDATA #IMPLIED TITLE CDATA #IMPLIED %link.addon; > </cptr> <h2 id=link.dsc>Description <p> The &link; element provides a means for specifying relationships between this hypertext document and other documents and resources. <p> Any number of &link; elements may appear in the head of a document and are typically used to indicate authorship, related indexes and glossaries, older and more recent versions, document hierarchy and associated resources such as style sheets. <p> The start tag is required, but the end tag shall be omitted. <h2 id=link.att>Attributes <p> The &link; element has the following attributes: <dl> <dt>CHARSET <dd>Please see the &a.charset; attribute of the &a; element <hdref refid=a page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>HREF <dd>Please see the &a.href; attribute of the &a; element. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>REL <dd>Please see the &a.rel; attribute of the &a; element. <dt>REV <dd>Please see the &a.rev; attribute of the &a; element. <dt>TITLE <dd>Please see the &a.title; attribute of the &a; element. </dl> <h2 id=link.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &link; element. <h1 id=map>The MAP element — Client-side image map <h2 id=map.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT MAP - - (AREA)+ > <!ATTLIST MAP NAME CDATA #REQUIRED %map.addon; > </cptr> <h2 id=map.dsc>Description <p> The &e.map; element provides a mechanism for constructing client-side image maps. It acts as a container for a set of &area; elements which define individual regions in an image <hdref refid=area page=yes>. <p> Within a &e.map; careful attention should be given to the ordering of the &area; elements: If two or more regions overlap, then the first &area; element listed will take precedence over subsequent &area; elements, etc. Thus &area; elements with no associated action (which have the &a.nohref; attribute), should be specified in the &e.map; before &area;s defining hotzones (which have the &a.href; attribute). <p> Both start and end tags are required. <h2 id=map.att>Attributes <p> The &e.map; element has the following attribute: <dl> <dt>NAME <dd>This required attribute specifies a name for the map which may be referenced by the &a.usemap; attribute of an &img; element using a URL fragment identifier <hdref refid=img page=yes>. <note> The value of the &a.name; attribute is case sensitive. </note> </dl> <h2 id=map.ex>Example <cptr keep=off type=source width=column> <IMG SRC="USA.jpeg" USEMAP="#States"> ... <MAP NAME="States"> ... <AREA HREF="Florida.html" ALT="Florida" SHAPE=poly COORDS="..."> ... <AREA HREF="Texas.html" ALT="Texas" SHAPE=poly COORDS="..."> ... </MAP> </cptr> <h1 id=meta>The META element <h2 id=meta.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT META - O EMPTY > <!ATTLIST META CONTENT CDATA #REQUIRED HTTP-EQUIV NAME #IMPLIED NAME NAME #IMPLIED %meta.addon; > </cptr> <h2 id=meta.dsc>Description <p> The &meta; element is an extensible container for use in identifying specialized document meta-information. Meta-information has two main functions: <ul> <li>To provide a means to verify that the data set exists, and determine how it might be obtained or accessed. <li>To document the content, quality and features of a data set, indicating its fitness for use. </ul> Each &meta; element specifies a name/value pair. If multiple &meta; elements are provided with the same name, their combined contents — concatenated as a comma-separated list — is the value that will be associated with that name. <nl> <li>The &meta; element should not be used where a specific element, such as &title;, would be more appropriate. Rather than specifying a &meta; element with a URL as the value of the &a.content; attribute, use a &link; element instead. <li> HTTP servers may read the content of a document's &head; element to generate header fields corresponding to those elements defining a value for the attribute &a.http-equiv; <li> The method by which the server extracts document meta-information is unspecified and not mandatory. The &meta; element only provides an extensible mechanism for identifying and embedding such document meta-information — how it may be used is up to the individual server implementation and ISO-HTML user agent. </nl> <p> An HTTP server shall not use the &meta; element to form an HTTP response header unless the &a.http-equiv; attribute is present. <p> An HTTP server may disregard any &meta; elements that specify information controlled by the HTTP server, for example `Server', `Date', and `Last-modified'. <p> The start tag is required, but the end tag shall be omitted. <h2 id=meta.att>Attributes <p> The attributes of the &meta; element are: <dl> <dt>CONTENT <dd>This attribute is required and specifies the value part of a name/value pair. <dt>HTTP-EQUIV <dd>Binds the element to an HTTP header field. An HTTP server may use this information to process the document. In particular, it may include a header field in the responses to requests for this document: the header name is taken from the &a.http-equiv; attribute value, and the header value is taken from the value of the &a.content; attribute. HTTP header names are not case sensitive. <dt>NAME <dd>Specifies the name part of the name/value pair. If not present, then the &a.http-equiv; attribute is used to specify the name. </dl> <p> Either the &a.http-equiv; or the &a.name; attribute shall be present. <h2 id=meta.ex>Example <p> If a document contains: <cptr keep=off type=source width=column> <META HTTP-EQUIV="Expires" CONTENT="Sat, 1 Jan 2000 00:00:01 UTC"> <META HTTP-EQUIV="Keywords" CONTENT="Fred"> <META HTTP-EQUIV="Reply-to" CONTENT="Flinstones@bedrock.old"> <META HTTP-EQUIV="Keywords" CONTENT="Barney"> </cptr> then the server may include the following header fields: <cptr keep=off type=source width=column> Expires: Sat, 1 Jan 2000 00:00:01 UTC Keywords: Fred, Barney Reply-to: Flintstones@bedrock.old </cptr> as part of an HTTP response to a `get' or `post' request for that document. <h1 id=nop>The NOP element — Interface default value <h2 id=nop.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT NOP - O EMPTY > </cptr> <h2 id=nop.dsc>Description <p> The &nop; element has no content and no meaning. It shall be ignored by user agents, and shall not be used in conforming documents. <note> The &nop; element is used in the DTD to ensure syntactically correct content models for unused interfaces. </note> <h2 id=nop.att>Attributes <p> The &nop; element has no attributes. <h1 id=object>The OBJECT element — Simple agent <h2 id=object.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT OBJECT - - (PARAM | %section.content;)* > <!ATTLIST OBJECT ALIGN (texttop,middle,textmiddle,baseline, textbottom,left,center,centre,right) #IMPLIED CLASS NAMES #IMPLIED CLASSID %URL; #IMPLIED CODEBASE %URL; #IMPLIED CODETYPE CDATA #IMPLIED DATA %URL; #IMPLIED DECLARE (declare) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME %URL; #IMPLIED STANDBY CDATA #IMPLIED TYPE CDATA #IMPLIED USEMAP %URL; #IMPLIED %object.addon; > </cptr> <note>&NB; The &a.name; attribute is declared as having type <cptr>%URL;</cptr>. Is this correct? [Dix points]. </note> <h2 id=object.dsc>Description <p> The &object; element provides a technique for specifying richer behaviour for the document in the form of a simple agent. The scheme uses only elements and attributes and is simple to implement. <note> This technique deliberately avoids using sophisticated SGML techniques in order to facilitate its use by current user agents. Nothing in this &text; prevents the use of those SGML techniques by more sophisticated user agents. </note> <p> The specification of the &object; element supports the general insertion of media other than text into documents, but excludes the agent architecture, application programming interface (API) and inter-agent communication. <p> The agent may access data specified: <ul compact> <li>In the document text flow, <li>As a set of named properties, <li>In a separate file which is the target of a link specified within the document, <li>Using techniques outside the scope of this &text;. </ul> <p> A conforming ISO-HTML browser is not required to support the agent behaviour, and if it does not it shall render the <cptr>%section.content;</cptr> part of the content. Browsers which render the agent behaviour shall not render the <cptr>%section.content;</cptr>. If a browser chooses to render the <cptr>%section.content;</cptr>, then it may also display the content of the &a.alt; attribute to indicate that this choice has been made. <p> It is recommended that authors provide a suitable substitute for the agent behaviour in the <cptr>%section.content;</cptr> part of the content of the element. <p> The agents are normally loaded relative to the document URL (or &base; element if it is defined), and the &a.codebase; attribute may be used to change this default behaviour. If the &a.codebase; attribute is defined, then it specifies a different location for the agent resources. The value can be an absolute or relative URL. An absolute URL is used as is without modification and is not effected by a document's &base; element. When the &a.codebase; attribute is relative, then it is relative to the document URL (or &base; tag if defined). <p> Both start and end tags are required. <h2 id=object.att>Attributes <p> The &object; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>CLASSID <dd>Provides a URL which identifies an implementation for the agent. In some systems the value is a class identifier. The values are not defined by this &text;. If this attribute is omitted, a default value may be deduced from the &a.data; attribute value. <note> Since the class identifiers in some object systems may be cumbersome, the &a.classid; attribute may use a short URL to specify a class identifier indirectly. </note> <dt>CODEBASE <dd>Some URL schemes used to identify implementations require an additional URL to locate the implementation. This attribute provides the additional URL. The values are not defined by this &text;. <dt>CODETYPE <dd>The Internet media type of the code referenced by the &a.classid; attribute. <dt>DATA <dd>Provides a URL for the agent's data, eg. a JPEG file. The &a.type; attribute may provide the Internet Media Type for the data. In the absence of the &a.classid; attribute, the Internet Media Type of the data gives the default value to the &a.classid; attribute. <dt>DECLARE <dd>The &a.declare; attribute calls for late binding of the agent. The attribute is usually used to indicate that <ul compact> <li>¶m; elements are to be "object valued". <li>Hyperlinks point to objects which cannot otherwise be addressed with a single URL. </ul> Each such binding typically results in a separate instantiation of the agent. <cptr><OBJECT DECLARE></cptr> is treated as a declaration for making an instance of the agent. <p> If the declared agent is not supported, the user agent should use the content of the <cptr><OBJECT DECLARE></cptr> element which may contain another &object; element. The &a.type; attribute may be used to specify the Internet Media Type for the outer agent as a hint for this situation. <p> Anchors may exploit nested declared agents to provide alternative media for a given resource. <note>&NB; This explanation of the &a.declare; attribute is based on text that came from the W3C. It needs further work and some good examples. </note> <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. The &a.id; attribute is also available to assist inter agent communication. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>If the &a.name; attribute is present and if the &object; element is contained in a &form; element, then this agent participates in the form submission process. If &a.name; is present and the &a.declare; attribute is absent, the user agent should include the value of the &a.name; attribute and the data obtained from the agent along with information derived from other form fields, see <hdref refid=form.submit page=yes>. <note> The mechanism used to obtain an agent's data is specific to each agent. </note> <dt>STANDBY <dd>The attribute value provides a message to be displayed while the agent is being prepared. <dt>TYPE <dd>The Internet Media Type of the data referenced through use of the &a.data; attribute. <dt>USEMAP <dd>Specifies a URL fragment identifier for a client-side image map defined with the &e.map; element, see <hdref refid=map page=yes>. In general, the use of this attribute is only appropriate for static images. </dl> <h2 id=object.ex>Examples <note>&NB; Good examples are urgently needed. </note> <h1 id=ol>The OL element — Ordered list <h2 id=ol.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (OL|UL) - - (LI)+ > <!ATTLIST OL ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED START NUMBER #IMPLIED %ol.addon; > </cptr> <h2 id=ol.dsc>Description <p> The &ol; element represents a non empty ordered list of items, presented in sequence or order of importance. It is usually rendered as a numbered list. <p> The content is a sequence of &li; elements <hdref refid=li page=yes>. <p> Both start and end tags are required. <h2 id=ol.att>Attributes <p> The attributes of the &ol; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &li; elements. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>START <dd>Specifies an initial value for the item sequence number. If the &a.start; attribute is omitted, the default start value is 1. If specified, the value shall be an integer. </dl> <h2 id=ol.ex>Example <cptr keep=off type=source width=column> Easter Calendar <OL START=1997> <LI>30th March <LI>12th April <LI>4th April <LI>23rd April </OL> </cptr> <h1 id=option>The OPTION element — User choice <h2 id=option.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT OPTION - O (#PCDATA) > <!ATTLIST OPTION CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED SELECTED (selected) #IMPLIED VALUE CDATA #IMPLIED %option.addon; > </cptr> <h2 id=option.dsc>Description <p> The &option; element can only occur within a &select; element and represents one of the possible choices that is offered. The content of the &option; element is presented to users as a representation of the option. Its content is used as the returned value if the &a.value; attribute is not present. <p> The start tag is required, but the end tag may be omitted. <h2 id=option.att>Attributes <p> The attributes of the &option; element are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>SELECTED <dd>Indicates that the option is initially selected. <dt>VALUE <dd>Indicates the value to be returned if this option is chosen. If omitted, the returned value defaults to the content of the &option; element. <note> The name associated with this value is given by the &a.name; attribute of the parent &select; element. </note> </dl> <h2 id=option.ex>Example <p> See <hdref refid=select page=yes> for an example of the use of the &option; element. <h1 id=p>The P element — Paragraph <h2 id=p.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT P - O (%text;)+ > <!ATTLIST P ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %p.addon; > </cptr> <h2 id=p.dsc>Description <p> The contents of the &p; element form a paragraph. <note> Authors often start paragraphs on new lines to facilitate document maintenance. </note> <p> The start tag is required, but the end tag may be omitted. <h2 id=p.att>Attributes <p> The &p; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=p.ex>Example <cptr keep=off type=source width=column> <P> It was a dark and stormy night ... </cptr> <h1 id=param>The PARAM element — Agent interface parameter <h2 id=param.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT PARAM - O EMPTY> <!ATTLIST PARAM NAME NAME #REQUIRED TYPE CDATA #IMPLIED VALUE CDATA #IMPLIED VALUETYPE (data|ref|object) data %param.addon; > </cptr> <h2 id=param.dsc>Description <p> The ¶m; element appears only in &object; element content and specifies the name and value of a parameter to be passed to the agent <hdref refid=object page=yes>. The ¶m; element has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=param.att>Attributes <p> The ¶m; element has the following attributes: <dl> <dt>NAME <dd>This attribute is required and specifies the name of the parameter. <note> Case sensitivity is implementation dependent. </note> <dt>TYPE <dd>This attribute is used only when the &a.valuetype; attribute has the value "ref" and specifies the Internet Media Type of the referenced value. <dt>VALUE <dd>Specifies the value of the parameter. For example: <cptr>VALUE="expensive"</cptr>. <dt>VALUETYPE <dd>Specifies how the &a.value; attribute is to be interpreted. Possible values are as follows: <dl> <dt>VALUETYPE=data <dd>The value "expensive" is to be passed directly to the agent as a string. In the absence of a &a.valuetype; attribute, "data" is the default behaviour for the &a.valuetype; attribute. <dt>VALUETYPE=object <dd>The value "expensive" is treated as the URL of an object or agent in the same document. This is used primarily for object or agent valued properties where the value of a property is a pointer (or reference) to a running object or to an agent. <dt>VALUETYPE=ref <dd>The value "expensive" is a URL. </dl> <note> The attribute value is obtained from the attribute value literal after replacing any entity references or character references within the literal and then normalizing the result by throwing out any entity ends and record starts and replacing any record end or separator characters with a space [see Goldfarb p.331]. </note> </dl> <h2 id=param.ex>Example <p> See <hdref refid=object page=yes> for an example of the use of the ¶m; element. <h1 id=pre>The PRE element — Pre-formatted text <h2 id=pre.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT PRE - - (%text;)+ -(BIG|IMG|MAP|SMALL|SUB|SUP|%body.extend;) > <!ATTLIST PRE CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED WIDTH NUMBER #IMPLIED %pre.addon; > </cptr> <h2 id=pre.dsc>Description <p> The &e.pre; element contains a non-empty block of text that is to be formatted as is, with its contents rendered in a mono-spaced font. <p> Within pre-formatted text, line breaks are rendered faithfully as line breaks. A user agent may use a constant indent when rendering pre-formatted text. <p> Anchor elements and phrase markup may be embedded in pre-formatted text. <note> Constraints on the processing of &e.pre; element content may limit the ability of a user agent to faithfully render phrase markup. </note> <p> Both start and end tags are required. <h2 id=pre.att>Attributes <p> The &e.pre; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>WIDTH <dd>Specifies the maximum number of characters in a line and allows a user agent to select a suitable font and indentation. The value of this attribute shall be a strictly positive integer. </dl> <h2 id=pre.ex>Example <cptr keep=off type=source width=column> <PRE WIDTH=16> Tic Tac Toe X | | O ----------- O | X | ----------- | | X </PRE> </cptr> <h1 id=q>The Q element — Quote <h2 id=q.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | Q | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=q.dsc>Description <p> The &q; element contains a quotation. Rendering is usually language and platform dependent. <p> Both start and end tags are required. <h2 id=q.att>Attributes <p> The &q; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=q.ex>Example <p> <cptr keep=off type=source width=column> A <Q LANG=de>quotation in German</Q> and a <Q LANG=fr>quotation in French</Q>. </cptr> might be rendered as: <p> A ,,quotation in German'' and a << quotation in French >>. <h1 id=samp>The SAMP element — Sample output <h2 id=samp.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | SAMP | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=samp.dsc>Description <p> The &samp; element contains sample output from a program or a script. <p> Both start and end tags are required. <h2 id=samp.att>Attributes <p> The &samp; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=samp.ex>Example <cptr keep=off type=source width=column> The installation begins: <samp>Insert diskette 1</samp> </cptr> <h1 id=select>The SELECT element — Selection in forms <h2 id=select.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT SELECT - - (OPTION)+ > <!ATTLIST SELECT CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED MULTIPLE (multiple) #IMPLIED NAME CDATA #REQUIRED SIZE NUMBER #IMPLIED %select.addon; > </cptr> <h2 id=select.dsc>Description <p> The &select; element defines one-from-many and many-from-many menus. The items from which the choice is made are defined by a sequence of &option; elements. <hdref refid=option page=yes> <p> At least one &option; shall be specified in a &select; element. <p> One-from-many menus are usually rendered as drop-down menus while many-from-many menus are usually rendered as list menus. <p> The initial state has the first option selected, unless a &a.selected; attribute is present on at least one of the &option; elements <hdref refid=option page=yes>. <p> Both start and end tags are required. <h2 id=select.att>Attributes <p> The &select; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MULTIPLE <dd>The presence of this attribute signifies that the users can make multiple selections. If this attribute is omitted, at most one selection is allowed. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>NAME <dd>Specifies the name used to identify the menu choice when the form is submitted to a server. Each selected option results in a name/pair value being included in the form's contents. This attribute is required. <dt>SIZE <dd>Suggests the number of visible choices. </dl> <h2 id=select.ex>Example <cptr keep=off type=source width=column> <SELECT NAME="flavour"> <OPTION>Vanilla <OPTION SELECTED>Chocolate <OPTION VALUE="LemonLime">Lemon and Lime <OPTION>Strawberry </SELECT> </cptr> <h1 id=small>The SMALL element — Small character style <h2 id=small.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SMALL | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=small.dsc>Description <p> The contents of a &small; element should be rendered in a smaller font than would otherwise be used. <p> Both start and end tags are required. <h2 id=small.att>Attributes <p> The &small; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=small.ex>Example <cptr keep=off type=source width=column> <BIG>What the large print giveth,</BIG> <SMALL>the small print taketh away.</SMALL> </cptr> <h1 id=span>The SPAN element — Generic container <h2 id=span.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT SPAN - - (%text;)+ > <!ATTLIST SPAN CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %span.addon > </cptr> <h2 id=span.dsc>Description <p> The &span; element provides a generic container to carry language and bidirectional attributes in cases where no other element is appropriate. <p> Both start and end tags are required. <h2 id=span.att>Attributes <p> The optional attributes of the &span; element are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=span.ex>Example <cptr keep=off type=source width=column> The licence plate <SPAN DIR=ltr>CIA 1</SPAN> will appear correctly even if cut and pasted into the source of a report normally rendered in a right-to-left fashion. </cptr> <h1 id=strike>The STRIKE element — Strike-out character style <h2 id=strike.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | STRIKE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=strike.dsc>Description <p> The contents of a &strike; element should be rendered in a "strike-through" text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=strike.att>Attributes <p> The &strike; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=strike.ex>Example <cptr keep=off type=source width=column> The original description should <STRIKE>NOT</STRIKE> have been omitted. </cptr> <h1 id=strong>The STRONG element — Strong emphasis <h2 id=strong.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | STRONG | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=strong.dsc>Description <p> The contents of a &strong; element are strongly emphasised and are usually rendered in a bold font. <p> Both start and end tags are required. <h2 id=strong.att>Attributes <p> The &strong; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=strong.ex>Example <cptr><EM>High Voltage <STRONG>Danger</STRONG></EM></cptr> <h1 id=sub>The SUB element — Subscript character style <h2 id=sub.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SUB | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=sub.dsc>Description <p> A &e.sub; element contains subscripted text. Its contents should be positioned below the base line and rendered in a smaller size than the preceding text. <p> Both start and end tags are required. <h2 id=sub.att>Attributes <p> The &e.sub; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=sub.ex>Example <cptr>H<SUB>2</SUB>O</cptr> <h1 id=sup>The SUP element — Superscript character style <h2 id=sup.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SUP | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=sup.dsc>Description <p> A &e.sup; element contains superscripted text. Its contents should be positioned above the base line and rendered in a smaller size than the preceding text. <p> Both start and end tags are required. <h2 id=sup.att>Attributes <p> The &e.sup; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=sup.ex>Example <cptr>E = mc<SUP>2</SUP></cptr> <h1 id=table>The TABLE element — Tables <h2 id=table.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TABLE - - (CAPTION?, TR+) > <!ATTLIST TABLE ALIGN (left,center,centre,right) left BORDER %pixels; #IMPLIED CELLPADDING %pixels; #IMPLIED CELLSPACING %pixels; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED DUMMY (border) #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED WIDTH %length; #IMPLIED %table.addon; > </cptr> <h2 id=table.dsc>Description <p> The &table; element provides a widely deployed subset of the specification given in RFC 1942 <bibref refid=tables>. It is used to <ul compact> <li>Markup tabular material. <li>Describe tabular layout. <note> Authors should be aware that this latter use causes problems when rendering to speech or text-only agents. </note> </ul> <p> A table contains an optional caption <hdref refid=caption page=yes> followed by a non-empty sequence of table rows <hdref refid=tr page=yes>. <p> Both start and end tags are required. <h2 id=table.att>Attributes <p> The attributes of the &table; element are as follows: <note> The pixel values refer to screen pixels, and should be multiplied by an appropriate factor when rendering to high resolution output devices such as laser printers. For instance if a user agent has a display with 75 pixels per inch and is rendering to a laser printer with 600 dots per inch, then the pixel values given in the attributes should be multiplied by a factor of 8 prior to output. </note> <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes>. <dt>BORDER <dd>Specifies the width of the outer border around a table in screen pixels, eg. &a.border;=4. The border may be suppressed by setting &a.border;=0. If both this attribute and the attribute &a.dummy; are omitted, the border should be suppressed. <dt>CELLPADDING <dd>Specifies the padding in screen pixels between a cell's contents and the border around the cell. <dt>CELLSPACING <dd>Each cell in a table has its own border which is separated from the borders around neighbouring cells. This separation can be set using the &a.cellspacing; attribute, eg. &a.cellspacing;=10 for a 10 pixel separation. This attribute also determines the separation between the table border and the borders of the outermost cells. <note> In traditional desktop publishing software, adjacent table cells share a common border. </note> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>DUMMY <dd>Allows an author to specify the attribute value BORDER as if it were an attribute name with no associated value. Such a specification requests that the border be drawn using the user agent's default width, usually one pixel. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>WIDTH <dd>Sets the table width. If an integer is specified, the unit is the screen pixel, eg. &a.width;=212, but if a percentage is specified by concatenating a "%" character as a suffix to the value, the table width is defined as a percentage of the space between the current left and right margins, eg. &a.width="80%". <p> If this attribute is omitted, the table width is automatically determined from the table contents. </dl> <h2 id=table.ex>Example <p> Tables take the general form: <cptr keep=off type=source width=column> <TABLE BORDER=3 WIDTH="64%" CELLSPACING=2 CELLPADDING=2> <CAPTION>Complexity of sorting algorithms</CAPTION> <TR><TH> Algorithm <TH> Complexity <TR><TD> Random shuffle <TD> O(exp(N)) <TR><TD> Bubble sort <TD> O(NxN) <TR><TD> Quicksort <TD> O(NlogN) <TR><TD> Heap sort <TD> O(NlogN) </TABLE> </cptr> <h1 id=td>The TD element — Table data cell <h2 id=td.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT (TH|TD) - O %table.content; > <!ATTLIST ( ... | TD ) ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COLSPAN NUMBER 1 DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NOWRAP (nowrap) #IMPLIED ROWSPAN NUMBER 1 VALIGN %valign; #IMPLIED %td.addon; > </cptr> <h2 id=td.dsc>Description <p> The &td; element is found only in &tr; table rows <hdref refid=tr page=yes> and contains a data cell in a table. <p> The start tag is required, but the end tag may be omitted. <h2 id=td.att>Attributes <p> The attributes of the &td; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLSPAN <dd>Specifies the number of columns spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NOWRAP <dd>The presence of this attribute disables automatic word wrap within the contents of the cell. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <note> This is equivalent to using the   entity for non-breaking spaces within the content of the cell. </note> <dt>ROWSPAN <dd>Specifies the number of rows spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents which overrides any value specified by the containing &tr; element <hdref refid=tr page=yes>. The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle", unless overridden by the containing &tr; element. The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <note> A table cell may contain a nested table. </note> <h2 id=td.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &td; element. <h1 id=textarea>The TEXTAREA element — Multi-line text field <h2 id=textarea.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TEXTAREA - - (#PCDATA) > <!ATTLIST TEXTAREA CLASS NAMES #IMPLIED COLS NUMBER #REQUIRED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME CDATA #REQUIRED ROWS NUMBER #REQUIRED %textarea.addon; > </cptr> <h2 id=textarea.dsc>Description <p> The &textarea; element is used in forms to represent the initial value of a multi-line text field which may be modified by a user. Its contents are restricted to text and character entities. <nl> <li> It is recommended that user agents canonicalize line endings to CRLF when submitting a field's contents to a server. <li> This &text; does not specify the coded character set to be used for submitting such text to a server. </nl> <p> The field is usually rendered in a mono-spaced font. <p> Both start and end tags are required. <h2 id=textarea.att>Attributes <p> The attributes of the &textarea; element are: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLS <dd>This required attribute specifies the prefered width of the field in visible mono-spaced characters. Users should be able to enter longer lines than this, so user agents should provide a means for scrolling through the text field when its contents extend beyond the visible area. User agents may wrap visible text lines to keep long lines visible without the need for scrolling. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This required attribute specifies the name that is used to identify the field when the form is submitted to a server. <dt>ROWS <dd>This required attribute specifies the number of visible text lines. Users should be able to enter more lines that this, so user agents should provide some means for scrolling through the text field when its contents extend beyond the visible area. </dl> <h2 id=textarea.ex>Example <cptr keep=off type=source width=column> <TEXTAREA NAME=comments ROWS=3 COLS=32> I found that your product was excellent! </TEXTAREA> </cptr> <h1 id=th>The TH element — Table header <h2 id=th.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT (TH|TD) - O %table.content; > <!ATTLIST ( TH | ... ) ALIGN %justify;#IMPLIED CLASS NAMES #IMPLIED COLSPAN NUMBER 1 DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NOWRAP (nowrap) #IMPLIED ROWSPAN NUMBER 1 VALIGN %valign; #IMPLIED %th.addon; > </cptr> <h2 id=th.dsc>Description <p> The &th; element is found only in &tr; table rows <hdref refid=tr page=yes> and contains a header cell in a table. &th; elements are distinguished from &td; elements so that a user agent may provide different rendering styles for cell header and cell data. <p> &th; is otherwise identical to &td; except that if the &a.align; attribute is not specified, the default alignment is "centre" unless overridden by the enclosing &tr; element. <p> The start tag is required, but the end tag may be omitted. <h2 id=th.att>Attributes <p> The &th; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes> for a general description. This use of the &a.align; attribute differs only in that the default value "centre" is to be used if none is supplied. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLSPAN <dd>Specifies the number of columns spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NOWRAP <dd>The presence of this attribute disables automatic word wrap within the contents of the cell. <note> This is equivalent to using the   entity for non-breaking spaces within the content of the cell. </note> <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>ROWSPAN <dd>Specifies the number of rows spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents which overrides any value specified by the containing &tr; element. <hdref refid=tr page=yes> The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle", unless overridden by the containing &tr; element. The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <h2 id=th.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &th; element. <h1 id=title>The TITLE element — Document title <h2 id=title.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TITLE - - (#PCDATA) -(LINK | META | %head.extend;) > <!ATTLIST TITLE CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %title.addon; > </cptr> <note>&NB; Shouldn't the content model be replaced by RCDATA? </note> <h2 id=title.dsc>Description <p> The &title; element identifies the contents of a document in a global context. A short title such as "Introduction" may be meaningless out of context, and hence a title such as "Introduction to medieval bee-keeping" might be more appropriate. <note> The length of a title is not limited; however long titles may be truncated in some applications. To minimize this possibility titles should contain fewer than 64 characters. </note> <p> An ISO-HTML document shall contain one &title; element. <p> A user agent may display the title of a document in a history list or as a label for the window displaying the document's content. <p> Both start and end tags are required. <h2 id=title.att>Attributes <p> The &title; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=title.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &title; element. <h1 id=tr>The TR element — Table row <h2 id=tr.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT TR - O (TH|TD)+ > <!ATTLIST TR ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VALIGN %valign; #IMPLIED %tr.addon; > </cptr> <h2 id=tr.dsc>Description <p> The &tr; is found only in a &table; element <hdref refid=table page=yes> and specifies a non-empty row in a table. It acts as a container for &th; table header <hdref refid=th page=yes> and &td; table data <hdref refid=td page=yes> elements. <p> The start tag is required, but the end tag may be omitted. <h2 id=tr.att>Attributes <p> The &tr; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents. The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle". The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <h2 id=tr.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &tr; element. <h1 id=tt>The TT element — Mono-spaced character style <h2 id=tt.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | TT | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=tt.dsc>Description <p> The contents of a &tt; element should be rendered in a mono-spaced font. <p> Both start and end tags are required. <h2 id=tt.att>Attributes <p> The &tt; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=tt.ex>Example <cptr keep=off type=source width=column> The telex began: <TT LANG=fr>CONFIDENTIEL</TT> </cptr> <h1 id=u>The U element — Underlined character style <h2 id=u.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | U" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=u.dsc>Description <p> The contents of an &u; element should be rendered in an underlined text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=u.att>Attributes <p> The &u; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=u.ex>Example <cptr keep=off type=source width=column> Browsers frequently <U>underline</U> the text associated with a link. </cptr> <h1 id=ul>The UL element — Unordered list <h2 id=ul.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (OL|UL) - - (LI)+ > <!ATTLIST UL ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %ul.addon; > </cptr> <h2 id=ul.dsc>Description <p> The &ul; element represents a non empty unordered list of items usually rendered as a bulleted sequence. <p> The content is a sequence of &li; elements <hdref refid=li page=yes>. <p> Both start and end tags are required. <h2 id=ul.att>Attributes <p> The attributes of the &ul; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &li; elements. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=ul.ex>Example <note>&NB; Full details to be provided by Italian NB. </note> <cptr keep=off type=source width=column> <P LANG=it>Gelati <UL> <LI>Vanilla <LI>Chocolate <LI>Strawberry </UL> </cptr> <h1 id=var>The VAR element — Generic variable <h2 id=var.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | VAR" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=var.dsc>Description <p> The &var; element contains a placeholder variable, usually rendered in italic. <p> Both start and end tags are required. <h2 id=var.att>Attributes <p> The &var; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=var.ex>Example <cptr keep=off type=source width=column> Type <CODE>rm <VAR>file</VAR></CODE> to delete <VAR>file</VAR> from directory. </cptr> </body>  <annexn id=formal cols=1> <h0t>Formal definitions <note> These formal definitions are not intended to be self documenting. The elements and attributes are described in the body of this &text;. The SGML techniques are described in <bibref refid=goldfarb>. </note> <h1 id=dcl>The SGML declaration <cptr keep=off type=source width=page> iso-html.dcl </cptr> <h1 id=ent>The entity set <cptr keep=off type=source width=page> iso-html.ent </cptr> <h1 id=dtd>The elements and attributes <cptr keep=off type=source width=page> iso-html.dtd </cptr>  <annexi id=samples cols=1 page=no> <h0t>Sample marked up documents <note>&NB; Sample documents will not be provided until any corrections resulting from the CD balloting process have been folded into the text. </note> <annexi id=bibl cols=1 page=no> <h0t>Bibliography <bl> <bib id=ethno> <name>Grimes<cname>Barbara F. <pubtitle>Ethnologue, Languages of the World <pubedit>12th <publoc>Dallas <publish>Summer Institute of Linguistics <pubyear>1992 <pubibs>ISBN xxxxxxxxxxx <bib id=file> <name>Nebel<cname>Ernesto <name>Masinter<cname>Larry <pubtitle>Form-based File Upload in HTML <publoc>URL:ftp://ds.internic.net/rfc/rfc1867.txt <publish>Internet Engineering Task Force <pubyear>Novenber 1994 <pubref>RFC 1867 <bib id=goldfarb> <name>Goldfarb<cname>Charles F. <pubtitle>The SGML Handbook <pubedit>1st <publoc>Oxford <publish>Oxford University Press <pubyear>1990 <pubibs>ISBN 0-19-853737-9 <bib id=html2.0> <name>Berners-Lee<cname>Tim <name>Connolly><cname>Daniel <pubtitle>Hypertext Markup Language — 2.0 <publoc>URL:ftp://ds.internic.net/rfc/rfc1866.txt <publish>Internet Engineering Task Force <pubyear>1995 <pubref>RFC 1866 <bib id=html3.2> <name>Raggett<cname>Dave <pubtitle>HTML 3.2 Reference Specification REC-html32 <publoc>URL:http://www.w3.org/pub/WWW/TR <publish>World-Wide Web Consortium <pubyear>1997 <bib id=http> <name>Fielding<cname>Roy T. <name>Gettys<cname>Jim <name>Mogul<cname>Jeffrey C. <name>Nielsen<cname>Henrik Frystyk <name>Berners-Lee<cname>Tim <pubtitle>Hypertext Transfer Protocol — HTTP/1.1 <publoc>URL:ftp://ds.internic.net/rfc/rfc2068.txt <publish>Internet Engineering Task Force <pubyear>1997 <pubref>RFC 2068 <bib id=mime> <name>Borenstein<cname>N. <name>Freed<cname>N. <pubtitle>MIME (Multipurpose Internet Mail Extensions) Part One: Mechanisms for Specifying and Describing the Format of Internet Message Bodies.  <publoc>URL:ftp://ds.internic.net/rfc/rfc1521.txt <publish>Internet Engineering Task Force <pubyear>September 1993 <pubref>RFC 1521 <bib id=w3c.object> <name>Ragget<cname>Dave <name>Kindel<cname>Charlie <name>Montulli<cname>Lou <name>Sink<cname>Eric <name>Gramlich<cname>Wayne <name>Hirschman<cname>Jonathan <name>Berners-Lee<cname>Tim <name>Connolly<cname>Dan <pubtitle>Inserting objects into HTML, work in progress <publoc>URL:http://www.w3.org/pub/WWW/TR/WD-object.html <publish>World Wide Web Consortium <pubyear>1996 <pubref>WD-object-960422 <bib id=relurl> <name>Fielding<cname>Roy T. <pubtitle>Relative Uniform Resource Locators <publoc>URL:ftp://ds.internic.net/rfc/rfc1808.txt <publish>Internet Engineering Task Force <pubyear>June 1995 <pubref>RFC 1808 <bib id=rfci18n> <name>Yergeau<cname>François <name>Nicol<cname>Gavin <name>Adams<cname>Glenn <name>Dürst<cname>Martin <pubtitle>Internationalization of the Hypertext Markup Language <publoc>URL:ftp://ds.internic.net/rfc/rfc2070.txt <publish>Internet Engineering Task Force <pubyear>January 1997 <pubref>RFC 2070 <bib id=rfc1700> <name>Reynolds<cname>J <name>Postel<cname>Jon <pubtitle>Assigned Numbers <publoc>URL:ftp://ds.internic.net/rfc/rfc1700.txt <publish>Internet Engineering Task Force <pubyear>October 1994 <pubref>RFC 1700 <bib id=rfc1766> <name>Alverstrand<cname>H <pubtitle>Tags for the Identification of languages <publoc>URL:ftp://ds.internic.net/rfc/rfc1766.txt <publish>Internet Engineering Task Force <pubyear>March 1995 <pubref>RFC 1766 <bib id=tables> <name>Ragget<cname>Dave <pubtitle>HTML Tables <publoc>URL:ftp://ds.internic.net/rfc/rfc1942.txt <publish>Internet Engineering Task Force <pubyear>May 1996 <pubref>RFC 1942 <bib id=uri> <name>Berners-Lee<cname>Tim <pubtitle>Universal Resource Identifiers in WWW: A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web <publoc>URL:ftp://ds.internic.net/rfc/rfc1630.txt <publish>Internet Engineering Task Force <pubyear>1994 <pubref>RFC 1630 <bib id=url> <name>Berners-Lee<cname>Tim <name>Masinter<cname>Larry <name>McCahill<cname>M <pubtitle>Uniform Resource Locators (URL) <publoc>URL:ftp://ds.internic.net/rfc/rfc1738.txt <publish>Internet Engineering Task Force <pubyear>1994 <pubref>RFC 1738 </bl> <backm> <index> </standard>

From: ftp://ftp.cs.tcd.ie/isohtml/iso-html.sgml Date: 970326 Editors to NB experts:"> <A>" > <ADDRESS>" > <APPLET>" > <AREA>" > <B>" > <BASE>" > <BASEFONT>" > <BDO>" > <BLOCKQUOTE>" > <BODY>" > <BIG>" > <BR>" > <B1>" > <B2>" > <B3>" > <B4>" > <B5>" > <B6>" > <CAPTION>" > <CENTER>" > <CITE>" > <CODE>" > <DD>" > <DFN>" > <DIR>" > <DIV>" > <DL>" > <DT>" > <EM>" > <FONT>" > <FORM>" > <HEAD>" > <HR>" > <HTML>" > <H1>" > <H2>" > <H3>" > <H4>" > <H5>" > <H6>" > <I>" > <IMG>" > <INPUT>" > <ISINDEX>" > <ISO-HTML>" > <KBD>" > <LI>" > <LINK>" > <LISTING>" > <MAP>" -- Avoid clash with ISOamsa --> <MENU>" > <META>" > <NOP>" > <NOTE>" > <OBJECT>" > <OL>" > <OPTION>" > <P>" > <PARAM>" > <PLAINTEXT>" > <PRE>" -- Avoid clash in ISOamsr --> <Q>" > <SAMP>" > <SCRIPT>" > <SELECT>" > <SMALL>" > <SPAN>" > <STRIKE>" > <STRONG>" > <STYLE>" > <SUB>" -- Avoid clash with ISOtech --> <SUP>" -- Avoid clash with ISOtech --> <TABLE>" > <TEXTAREA>" > <TEXTFLOW>" > <TD>" > <TH>" > <TITLE>" > <TR>" > <TT>" > <U>" > <UL>" > <VAR>" > <XMP>" > ACCEPT" > ACTION" > ALIGN" > ALT" > BORDER" > CELLSPACING" > CHARSET" > CLASS" > CLASSID" > CODEBASE" > COMPACT" > CONTENT" > COORDS" > DATA" > DECLARE" > DIR" > DUMMY" > ENCTYPE" > HEIGHT" > HREF" > HSPACE" > HTTP-EQUIV" > ID" > ISMAP" > LANG" > METHOD" > NAME" > NOHREF" > NOP" > REL" > REV" > SELECTED" > SHAPE" > SRC" > START" > TITLE" > TYPE" > USEMAP" > VALUE" > VALUETYPE" > VERSION" > VSPACE" > WIDTH" > body.extend" > head.extend" > X.addon" > a.addon" > address.addon" > area.addon" > base.addon" > bdo.addon" > big.addon" > blockquote.addon" > body.addon" > bold.addon" > br.addon" > b1.addon" > b2.addon" > b3.addon" > b4.addon" > b5.addon" > b6.addon" > caption.addon" > dd.addon" > dfn.addon" > div.addon" > dl.addon" > dt.addon" > form.addon" > head.addon" > hr.addon" > h1.addon" > h2.addon" > h3.addon" > h4.addon" > h5.addon" > h6.addon" > img.addon" > input.addon" > isindex.addon" > iso-html.addon" > italic.addon" > li.addon" > link.addon" > literal.addon" > map.addon" > meta.addon" > ol.addon" > option.addon" > p.addon" > param.addon" > pre.addon" > select.addon" > small.addon" > span.addon" > strike.addon" > sub.addon" > sup.addon" > table.addon" > textarea.addon" > th.addon" > title.addon" > tr.addon" > u.addon" > ul.addon" > ]> This &text; has not yet been approved by SC18/WG8. JTC1/SC18/WG815xxx 15xxx e 1997-05-01 E Information Technology — HyperText Markup Language (ISO-HTML) <stitle>Technologies de l'information — Langage de balisage d'hypertexte (ISO-HTML) </titlep> <toc level=1> <foreword> <stdsrc>  <jtcsc>1 18 </stdsrc> <p> This &text; was prepared by the working group WG8 of SC18. The New Work Item Proposal was approved at the SC18 level in November 1996. <p> The New Work Item Proposal was approved at the JTC1 level in whenever 199x and assigned project number is 02.xx. <p> SC18/WG8 works on this project in liaison with the Internet Engineering Task Force and the World Wide Web Consortium. The text of this &text; follows closely the &dave; specification edited by David Raggett <cptr>URL:http://www.w3.org/pub/WWW/TR</cptr>, and the W3C Working Draft dated 1996-04-22 describing work in progress on "Inserting objects into HTML" <cptr>URL:http://www.w3.org/pub/WWW/TR/WD-object.html</cptr>. The World Wide Web Consortium's Recommendation is based on specifications prepared by the Internet Engineering Task Force: <ul> <li>IETF Proposed Standard RFC 1866 "HyperText Markup Language — 2.0" edited by Tim Berners-Lee and Daniel W. Connolly, September 22nd, 1995. <cptr>URL:ftp://ds.internic.net/rfc/rfc1866.txt</cptr> <li>IETF RFC 1867 "Form-based File Upload in HTML" edited by E. Nebel and L. Masinter, November 1995. <cptr>URL:ftp://ds.internic.net/rfc/rfc1867.txt</cptr> <li>IETF RFC 1942 "HTML Tables" edited by Dave Raggett, May 1996. <cptr>URL:ftp://ds.internic.net/rfc/rfc1942.txt</cptr> <li>IETF Internet Draft "Internationalization of the HyperText Markup Language" edited by François Yergeau, Gavin T. Nicol, Glenn Adams and Martin J. Dürst, January, 1997. <cptr>URL:ftp://ds.internic.net/rfc/rfc2070.txt</cptr> </ul> <replace>all previous working documents <repcom>which are now obsolete. </foreword> <intro id=intro> <p> The HyperText Markup Language is an SGML application conforming to International Standard ISO 8879 — Standard Generalized Markup Language. It provides a simple way of structuring hypertext documents which refer to one another and which collectively create an enormous "web" which continues to grow and evolve as many hypertext authors add and modify documents. <p> The web has expanded and browser developers have added additional features to the markup language such as new tags and new semantics for the tags. As a result, many documents have been created which can only be rendered faithfully on a limited number of browsers. Normal web practice is to hide any syntactic problems detected by the browsers and thus the reader is not always aware that the page being browsed is not faithful to the original authored document. <p> This &text; has been developed in an effort to ensure that it will remain possible for an author to produce simple hypertext for the web and be confident that a conforming browser will be able to render the document faithfully. This specification represents a core of the language to be supported by all conforming browsers and provides techniques for extending the core that are SGML conformant and represent good SGML practice. <p> The language defined by this &text; differs slightly from the &dave; specification. It omits all deprecated features of the language, and some other features whose role is purely cosmetic. This is done in preparation for the expected future introduction of style sheets. Certain optional facilities such as markup omission of the document element and the major elements have been removed to produce more robust texts in keeping with recognized good SGML practice. This does not reduce in any way the expressive power of the language. A minimal ISO-HTML document has the form: <cptr keep=off type=source width=column> <!DOCTYPE ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language//EN" [<!ENTITY % ISOhtme PUBLIC "ISO 15xxx//ENTITIES CD//EN" > <!ENTITY % ISOlat1 PUBLIC "ISO 8879:1986//ENTITIES Added Latin 1//EN" > %ISOlat1;%ISOhtme; ]> <ISO-HTML> <HEAD> <TITLE>Les unités de base</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <p> This &text; follows the current World Wide Web convention of presenting element and attribute names in upper case, although there is no formal requirement for the practice. <p> In order to support world wide use of the markup language, the internationalization facilities specified by the IETF have been included in this &text; in anticipation of their expected inclusion in the W3C specifications. <p> The elements of the ISO-HTML language are specified in this &text; in alphabetical order but may also be reviewed in the following order which reflects more closely the structure of the DTD: <sl compact> <li>The ISO-HTML element <hdref refid=iso-html page=yes> <sl compact> <li>The HEAD element <hdref refid=head page=yes> <sl compact> <li>The BASE element <hdref refid=base page=yes> <li>The ISINDEX element <hdref refid=isindex page=yes> <li>The LINK element <hdref refid=link page=yes> <li>The META element <hdref refid=meta page=yes> <li>The TITLE element <hdref refid=title page=yes> </sl> <li>The BODY element <hdref refid=body page=yes> <sl compact> <li>The ADDRESS element <hdref refid=address page=yes> <li>The block elements parameter entity <cptr>%block;</cptr> <sl compact> <li>The BLOCKQUOTE element <hdref refid=blockquote page=yes> <li>The DIV element <hdref refid=div page=yes> <li>The DL element <hdref refid=dl page=yes> <sl compact> <li>The DT element <hdref refid=dt page=yes> <li>The DD element <hdref refid=dd page=yes> </sl> <li>The FORM element <hdref refid=form page=yes> <li>The HR element <hdref refid=hr page=yes> <li>The ISINDEX element <hdref refid=isindex page=yes> <li>The OL element <hdref refid=ol page=yes> <sl compact> <li>The LI element <hdref refid=li page=yes> </sl> <li>The P element <hdref refid=p> page=yes> <li>The PRE element <hdref refid=pre page=yes> <li>The TABLE element <hdref refid=table page=yes> <sl compact> <li>The CAPTION element <hdref refid=caption page=yes> <li>The TR element <hdref refid=tr page=yes> <li>The TH element <hdref refid=th page=yes> <li>The TD element <hdref refid=td page=yes> </sl> <li>The UL element <hdref refid=ul page=yes> </sl> <li>The text elements parameter entity <cptr>%text;</cptr> <sl compact> <li>The physical style elements parameter entity <cptr>%physical.styles;</cptr> <sl compact> <li>The B element <hdref refid=b page=yes> <li>The BIG element <hdref refid=big page=yes> <li>The I element <hdref refid=i page=yes> <li>The SMALL element <hdref refid=small page=yes> <li>The STRIKE element <hdref refid=strike page=yes> <li>The SUB element <hdref refid=sub page=yes> <li>The SUP element <hdref refid=sup page=yes> <li>The TT element <hdref refid=tt page=yes> <li>The U element <hdref refid=u page=yes> </sl> <li>The logical style elements parameter entity <cptr>%logical.styles;</cptr> <sl compact> <li>The CITE element <hdref refid=cite page=yes> <li>The CODE element <hdref refid=code page=yes> <li>The DFN element <hdref refid=dfn page=yes> <li>The EM element <hdref refid=em page=yes> <li>The KBD element <hdref refid=kbd page=yes> <li>The Q element <hdref refid=q page=yes> <li>The SAMP element <hdref refid=samp page=yes> <li>The STRONG element <hdref refid=strong page=yes> <li>The VAR element <hdref refid=var page=yes> </sl> <li>The form elements, parameter entity <cptr>%form;</cptr> <sl compact> <li>The INPUT element <hdref refid=input page=yes> <li>The SELECT element <hdref refid=select page=yes> <sl compact> <li>The OPTION element <hdref refid=option page=yes> </sl> <li>The TEXTAREA element <hdref refid=textarea page=yes> </sl> <li>The special elements, parameter entity <cptr>%special;</cptr> <sl compact> <li>The A element <hdref refid=a page=yes> <li>The BDO element <hdref refid=bdo page=yes> <li>The BR element <hdref refid=br page=yes> <li>The IMG element <hdref refid=img page=yes> <li>The OBJECT element <hdref refid=object page=yes> <sl compact> <li>The PARAM element <hdref refid=param page=yes> </sl> <li>The MAP element <hdref refid=map page=yes> <sl compact> <li>The AREA element <hdref refid=area page=yes> </sl> <li>The SPAN element <hdref refid=span page=yes> </sl> </sl> <li>The sectioning elements <sl compact> <li>The H1 element <hdref refid=h1 page=yes> <li>The B1 element <hdref refid=b1 page=yes> <li>The H2 element <hdref refid=h2 page=yes> <li>The B2 element <hdref refid=b2 page=yes> <li>The H3 element <hdref refid=h3 page=yes> <li>The B3 element <hdref refid=b3 page=yes> <li>The H4 element <hdref refid=h4 page=yes> <li>The B4 element <hdref refid=b4 page=yes> <li>The H5 element <hdref refid=h5 page=yes> <li>The B5 element <hdref refid=b5 page=yes> <li>The H6 element <hdref refid=h6 page=yes> <li>The B6 element <hdref refid=b6 page=yes> </sl> </sl> </sl> </sl> <h2 id=markup>This document <p> This &text; represents the editors' best efforts to capture the continuing development of the language. The editors' have copied text from <ul> <li>the IETF RFC 1866 <bibref refid=html2.0>, <li>the &dave; specification <bibref refid=html3.2>, <li>the W3C Working Draft dated 1996-04-22 describing work in progress on <cit>Inserting objects into HTML</cit> <bibref refid=w3c.object>, <li>the SGML Handbook <bibref refid=goldfarb> where needed to ensure consistency with those specifications, <li>the IETF HTML internationalisation specification <bibref refid=rfci18n>. </ul> <h3 id=source>Source markup <p> This document was prepared using ISO 8879 based technology. Details of the DTD used to structure this document may be found in <cit>ISO/IEC TR 9573-11 Information processing — SGML support facilities — Techniques for using SGML — Part 11: Application at ISO Central Secretariat for International Standards and Reports</cit>. <body cols=2> <scope id=scope> <p> The scope of this &text; is a conforming application of SGML, ISO 8879:1986, which provides a simple and stable markup language for documents to be published on the World Wide Web. This &text; also provides techniques whereby application, regional or industry-specific extensions may be added in a way which follows good SGML practice. <h2 id=specif>Specificity of the scope <p> Since it is expected that a wide range of products, user applications, recommendations and other standards may use this standard, the scope focuses on the general structuring aspects and provides only sufficient semantics to ensure that the structures are rendered in a familiar way. <p> The scope excludes any standardization of models, services, systems, protocols or applications which are likely to make use of the ISO-HTML language. This specification does not define the "look and feel" of any conforming product. </scope> <conf id=conf> <h2 id=docconf>Conforming ISO-HTML documents <p> A document which conforms to this &text; shall <ol> <li>be a conforming SGML document, <note>&NB; This requirement is itself required by ISO 8879:1986 subclause 15.2.2 [Goldfarb p.480]. </note> <li>conform to the requirements of this &text;. </ol> <h2 id=sysconf>Conforming ISO-HTML systems <p> A conforming ISO-HTML system is a conforming SGML system which is able to process any conforming ISO-HTML document. <note> This requires the system to be capable of processing any conforming ISO-HTML document that is not inconsistent with its "system declaration". In order to promote the maximum capability for document interchange among systems, every conforming system must be able to support the reference concrete syntax and the ISO-HTML capacity set [Goldfarb p.481]. </note> <h3 id=sysdoc>Documentation of conforming ISO-HTML systems <p> Systems conforming to this &text; shall display the ISO-HTML identification text prominently: <ol compact> <li>In a prominent location in the front matter of all publications (normally the title page and the cover page), <li>On all identifying display screens of programs, <li>In all promotional and training material. </ol> and in the national language of the documentation. <p> The ISO-HTML system identification text is: <lq>ISO-HTML, an SGML system conforming to International Standard ISO 8879 — Standard Generalised Markup Language.</lq> <note> These requirements are intended to help users apply knowledge gained on one SGML system to the use of other systems, not to inhibit friendly documentation and human-computer interfaces [Goldfarb p.486]. </note> <note>&NB; The editors have deliberately omitted any reference to the national language to be used for the identification text. The issue is too complex, and not worth the trouble [Goldfarb p.486 subclause 15.5.1 line 6]. </note> <p> The documentation shall distinguish SGML constructs from ISO-HTML conventions and system functions, and shall identify the SGML constructs as being part of the Standard Generalized Markup Language. <nl> <li>The objective of this requirement is for the user to be aware of which constructs are common to all SGML systems, and which are unique to ISO-HTML. This will reduce the experienced user's learning time for a new system or application. <li>&NB; This requirement is itself required by ISO 8879:1986 subclause 15.5.2 [Goldfarb p.486 line 23]. </nl> <p> The documentation shall cite ISO 8879:1986 as a reference for supported SGML constructs that are not specifically documented for the system. <p> For example, if, for simplicity's sake, only a subset of some function is presented in the documentation (such as omitting some of the options of the entity declaration), it shall be stated clearly that other options exist and can be found in the SGML International Standard [Goldfarb p.487]. <note>&NB; Do we need to include text to cover the requirement of ISO 8879:1986 subclause 15.5.3 "Terminology"? [Goldfarb p.487]. </note> <h3 id=sysvalid>Validating ISO-HTML systems <p> In addition to being a conforming ISO-HTML system, a system is also a validating ISO-HTML system if <ol> <li>It is a validating SGML parser as defined by ISO 8879:1986 subclause 15.4; and <li>It finds and reports an ISO-HTML error if one exits; and <li>Does not report an ISO-HTML error where none exists. </ol> <note>&NB; The &dave; specification does not say what the reportable HTML errors are. </note> <h2 id=confi18n>Character set conformance <note>&NB; We have taken the same approach here as was taken by ISO 2022, and this subclause is based on ISO 2022 clause 3. </note> <p> The SGML declaration provided with this &text; calls for the use of the Basic Multilingual Plane of the ISO/IEC 10646 Universal Multiple-Octet Coded Character Set (UCS). ISO/IEC 10646 addresses whole classes of provisions and it is not intended by this &text; that they are all implemented in any user agent. As a result it is only practicable to envisage limited conformance to ISO/IEC 10646 as defined in this subclause. <p> Under limited conformance, the following is required: <ol> <li>When the characters described by ISO/IEC 10646 are used, they shall be implemented by the control functions, and with the meanings and coded representation specified in ISO/IEC 10646. <li>When two systems with different levels of implementation of the ISO/IEC 10646 coded character set are required to communicate with one another, they shall do so using using that part of the coded character set they have in common. <nl> <li>This &text; does not discuss the operation of the HTTP protocol <bibref refid=http>. </nl> <li>If a server is unable to express a document using the limited character set supported by the user agent, it should instead deliver a document in the limited character set explaining the impossibility. <li>Code positions that are either reserved for registration or reserved for future standardization shall not be used. <li>No registered escape sequence shall be used with a meaning different from that defined by ISO/IEC 10646. </ol> <p> The UTF-1 transformation format of ISO/IEC 10646, registered by IANA as ISO-10646-UTF-1, has been removed from the ISO/IEC 10646 and should not be used. </conf>  <refs apply=all> <refref>ISO 639:1988 <refref>ISO 646:1991 <refref>ISO 3166:1993 <refref>ISO 8859-1:1987 <refref>ISO 8879:1986 <refref>ISO 10646-1:1993 </refs> <defs apply=all id=defs> <defref>ISO 8879:1986 <dl> <dt id=d.justifying>Justifying element <dd>An element which carries an &a.align; attribute used to specify horizontal justification of its content relative to the current left and right margins. The &a.align; attribute takes one of the following values: <ul compact> <li>LEFT: text is left justified, ragged right. <li>CENTER: text is centered, ragged left and ragged right. <li>CENTRE: a permitted alternative specification of CENTER. <li>RIGHT: text is right justified, ragged left. </ul> <p> The justifying elements specified by this &text; are &justifying;. <dt id=d.boxing>Boxing element <dd>An element which carries an &a.align; attribute used to specify the position of a box around the visible rendered area. <p> The boxing elements specified by this &text; are &boxing;. <dt id=d.browser>Browser <dd>A User Agent whose main function is to present documents to the user. <dt id=d.character>Character <dd source="RFC1866">An atom of information, for example a letter or a digit. Graphic characters have associated glyphs, whereas control characters have associated processing semantics. <p> The multiple definitions and techniques for the representation of characters may be the source of confusion. The following table shows some of the ideas involved: <fig id=chars place=fixed width=page> <figbody> <cptr keep=off type=source width=page> The set of characters in this column forms the character repertoire |-------------------->| | This function called| |"coded character set"| | by RFC 1866 | ,---|---------------------|---------------------------------, | Code | Bit | Character |ISOlat1 |Numeric |Glyph| |posit- |pattern| name |entity |character| | | ion |(hexa) | |ref. |reference| | |-------+-------+------------------+--------+---------+-----| | 0 | 00 |Unused | | � | | | ... | | | | | | | 121 | 79 |SMALL LETTER y | | y | y | | 122 | 7A |SMALL LETTER z | | z | z | | 123 | 7B |LEFT CURLY BRACKET| | { | { | | ... | | | | | | | 199 | C7 |CAPITAL C, CEDILLA|Ç| Ç | | | 200 | C8 |CAPITAL E, GRAVE |È| È | | | 201 | C9 |CAPITAL E, ACUTE |É| É | | | ... | | | | | | | 255 | FF |SMALL y, UMLAUT |ÿ | ÿ | | '--|-------|-------------|----------------------------------' | | | | |------------>| | | This function is called the "character encoding | | scheme" by RFC 1866. | |<------------| | | This function is called "character set" by SGML | | | Called the "code set" by SGML Called the "character number" by SGML </cptr> <figcap>Illustration of some character representation definitions <figdesc>This illustration is based on the character set defined by ISO 8859-1:1987 "8-bit single-byte coded graphic character sets", Part 1: Latin alphabet No. 1. </fig> <dt id=d.ces>Character encoding scheme <dd source="RFC1866">A function whose domain is the set of sequences of octets, and whose range is the set of sequences of characters from a character repertoire; that is, a sequence of octets and a character encoding scheme determines a sequence of characters. <dt id=d.charrep>Character repertoire <dd source="RFC1866">A finite set of characters; eg. the range of a coded character set. <dt id=d.codepos>Code position <dd source="RFC1866">An integer; a coded character set and a code position from its domain determine a character. <dt id=d.ccs>Coded character set <dd source="RFC1866">A function whose domain is a subset of the integers and whose range is a character repertoire. That is, for some set of integers (usually of the form {0, 1, 2, ..., N}), a coded character set and an integer in that set determine a character. Conversely, a character and a coded character set determine the character's code position (or, in rare cases, a few code positions). <dt id=d.CRLF>CRLF <dd source="RFC1521">The sequence of the two ISO 646:1983 characters CR (13) and LF (10) which, taken together, in this order, denote a line break. <dt id=d.fds>Form data set <dd source="RFC 1866">A sequence of name/value pairs; the names are given by an ISO-HTML document and the values are given by the user. <dt id=d.fragid>Fragment identifier <dd source="RFC1866">The portion of an &a.href; attribute following the `<cptr>#</cptr>' character which may modify the presentation of the destination of a hyperlink. <dt id=d.HTMLbrowser>ISO-HTML browser <dd>Browser which presents ISO-HTML documents. <dt id=d.document>ISO-HTML document <dd>A document structured in accordance with this &text;. <dt id=d.hlink>Hyperlink <dd>A relationship between two anchors, called the source and the target. The link goes from the source to the target. The source is also known as the tail, and the target is also known as the destination or head. </dl> </defs> <symabb id=abbrev> <p> The following symbols and abbreviations are used in this &text;. <syml> <sym>ISO-HTML <desc>Pertaining to this standard. <sym>HTML 3.2 <desc>A specification for the Hypertext Markup Language developed by the World Wide Web Consortium. <sym>HTTP <desc>IETF RFC 2068 HyperText Transfer Protocol. <sym>HyTime <desc>One or more architectural forms as provided by <refref>ISO 10744:-</refref>. <sym>IANA <desc>Internet Assigned Numbers Authority. IANA is the central coordinator for the assignment of unique parameter values for Internet protocols. The IANA is chartered by the Internet Society (ISOC) and the Federal Network Council (FNC) to act as the clearinghouse to assign and coordinate the use of numerous Internet protocol parameters <bibref refid=rfc1700>. <sym>IETF <desc>Internet Engineering Task Force. <sym>RFC <desc>Request for Comments. An Internet Engineering Task Force specification. <sym>SGML <desc>Notation provided through use of <refref>ISO 8879:1986</refref>. <sym>URI <desc>Universal Resource Identifier as defined by <bibref refid=uri>. <sym>URL <desc>Uniform Resource Locator as defined by <bibref refid=url>. <sym>WWW <desc>World Wide Web <sym>W3C <desc>World Wide Web Consortium, founded in 1994 to develop common standards for the evolution of the World Wide Web. It is an industry consortium, hosted by the Massachusetts Institute of Technology Laboratory for Computer Science (MIT/LCS) in the United States, the Institut National de Recherche en Informatique et en Automatique (INRIA) in Europe and the Keio University Shonan Fujisawa Campus in Asia <cptr>http://www.w3.org/</cptr>. </syml> </symabb> <h1 id=requ>Requirements <p> This &text; has been designed to satisfy the following requirements, as documented in ISO/IEC JTC1/SC18/WG8 N1852 dated 24 June 1996: <ul> <li>Provide a minimum presentation architecture for SGML applications. <li>Clarify relationships between ISO-HTML and SGML details, such as SGML declaration and minimization. <li>Allow ISO-HTML to be used in environments where ISO standards are required. <li>Allow ISO-HTML to be used as a base architecture for other SGML applications. <li>Documents conforming to this &text; should be viewable by browsers that conform to the &dave; (or an appropriate later version of that specification). <li>Define useful subsets of the &dave; (or an appropriate later version of that specification). </ul>  <h1 id=i18n>Internationalization <p> The early applications of HTML on the World-Wide Web were seriously restricted by their reliance on the ISO 8859-1:1987 coded character set, which is appropriate only for Western European languages. This &text; provides facilities for a fuller internationalization, based on RFC 2070 <bibref refid=rfci18n>. The following subclauses specify the general internationalization features. <h2 id=a.justify>The ALIGN attribute when used for justification <p> The internationalization facilities use the &a.align; attribute to specify the horizontal alignment to be applied to content within the scope of the justifying elements: &justifying;. <p> The &a.align; justification attribute takes the following values: <dl> <dt>ALIGN=left <dd>Render "flush left" or left justified. The right edge is ragged. <dt>ALIGN=center <dd>Render "centered". The left and right edges are ragged. <dt>ALIGN=centre <dd>A permitted alternative specification of ALIGN=center. <dt>ALIGN=right <dd>Render "flush right" or right justified. The left edge is ragged. <dt>ALIGN=justify <dd>Render fully justified on the left and right. There are no ragged edges. </dl> <p> In the case of nested justification specifications, the innermost containing specification shall apply. If a justifying &a.align; attribute is omitted, the value of the next outer justifying specification is used. If the &a.align; attribute is omitted from all containing justifying elements, the value "left" shall be used. <note> There are elements defined by this &text; which have an &a.align; attribute for historical reasons, but are not justifying elements. E.g. the &caption; element and the boxing elements &boxing;. </note> <h3 id=align.ex>Examples of justifying behaviour <p> <ol> <li>In this example all the containing justification specifications are present, and the innermost applies. <cptr keep=off type=source width=column> <DIV ALIGN=left> <OL ALIGN=justify> <LI ALIGN=right>Important sentence ... </OL> </DIV> </cptr> The <q>Important sentence</q> is right aligned. <li>In this example only some of the containing justification specifications are present, and the innermost specification applies. <cptr keep=off type=source width=column> <DIV ALIGN=left> <OL> <LI ALIGN=justify> <P>Important sentence ... </OL> </DIV> </cptr> The <q>Important sentence</q> is fully justified. <li>In this example none of the containing justification specifications are present, hence the default value applies. <cptr keep=off type=source width=column> <DIV> <P>Important sentence ... </DIV> </cptr> The <q>Important sentence</q> is left justified. </ol> <h2 id=a.class>The CLASS attribute <p> It is expected that many ISO-HTML documents will be created by automatic processes from SGML documents structured according to richer document type definitions, eg. the DTD for ISO documents specified in ISO TR 9573. Conversion to the ISO-HTML structure may lead to the loss of some information about document content. For example the notes in an ISO document are tagged using the ¬e; element, rather than the &p; element. A rendering system might typically apply the ISO note style to the content of a ¬e; element, making it distinct in the output presentation. There may be a need to retain this type of information during the conversion to ISO-HTML so that browsers with the capability may provide a distinctive rendering. <p> The &a.class; attribute provides a means of attaching a sequence of SGML names to the element indicating the class to which the element belongs: typically the source element name if the document is the result of an automatic conversion from another document type. <note> This &text; does not define values for the &a.class; attribute. </note>  <h3 id=class.ex>An example <p> A conversion process might convert <cptr><note>This text does not define values for the class attribute.</note></cptr> to <cptr><P CLASS=note>This text does not define values for the class attribute.</P></cptr> <h2 id=a.dir>The DIR attribute <p> The &a.dir; attribute indicates the writing direction of text. With block-type elements the &a.dir; attribute indicates the base writing direction of the text in the block. If the attribute is omitted, the writing directional is inherited from the parent element if present. With inline elements, it indicates the element start a new writing direction embedding level. Directional-embedding is used to handle nested changes in writing direction. A common need for embedding characters is to handle text that has been pasted from one bidirectional context to another. <h3 id=dir.ex>An example <p> Here is an example of a case where direction-embedding is needed: <p> The following latin (shown in upper case) and arabic (shown in lower case) letters in backing store with the specified writing direction embeddings (LRE is shorthand for <SPAN DIR=ltr>, RLE for <SPAN DIR=rtl> and PDF for </SPAN>): <cptr keep=off type=source width=column> LRE A B RLE a b LRE C D PDF c d PDF E F PDF </cptr> result in the following rendering with square brackets showing the directional transitions: <cptr keep=off type=source width=column> [ A B [ d c [ C D ] b a ] E F ] </cptr> <note> If the &span; elements with &a.dir; attributes were not used, this example would result in the following rendering, assuming an inherited direction of "ltr". <cptr keep=off type=source width=column> [ A B [ b a [ C D ] d c ] E F ] </cptr> Notice that <cptr>b a</cptr> is on the left and <cptr>d c</cptr> is on the right unlike the example where writing direction embedding levels are used. Without writing direction embedding characters there would be at most two levels: a base directional level and a single counterflow level. </note> <h2 id=a.id>The ID attribute <p> When documents with the same information content in different languages are created, there may be a need to identify elements in the documents as being the "same" element, differing only in the language employed. <p> The &a.id; attribute provides a means of identifying the "same" element in different documents. Within each document, the elements' &a.id; attribute is set to the same value. <h2 id=a.lang>The LANG attribute <p> The &a.lang; attribute identifies a natural language spoken, sung, written or otherwise used by human beings for communication of information between people. Computer languages are explicitly excluded. The value of the &a.lang; attribute is referred to as the "language tag". <p> The syntax and registry of language tags used by this &text; is the same as that defined by RFC 1766 <bibref refid=rfc1766>, where a language tag is composed of one or more parts: a primary language tag and a possible empty series of subtags. For example, using RFC 822 EBNF: <cptr keep=off type=source width=column> language-tag = primary-tag *( "-" subtag ) primary-tag = 1*8ALPHA subtag = 1*8ALPHA </cptr> Whitespace is not allowed within the tag and all tags are case insensitive. The namespace of language tags is administered by the IANA. Example tags include: <cptr>en</cptr>, <cptr>en-US</cptr>, <cptr>en-cockney</cptr>, <cptr>i-cherokee</cptr> and <cptr>x-pig-latin</cptr>. <p> Two-letter primary-tags are reserved for ISO 639 language abbreviations. This &text; does not specify three-letter primary-tags, however their description may be found in the "Ethnologue" <bibref refid=ethno>. Any two-letter initial sub-tag is an ISO 3166 country name. <p> In the context of this &text;, a language tag is not to be interpreted as a single token, per RFC 1766, but as a hierarchy. For example, a user agent that adjusts its rendering according to language should recognize that it has a match when a language tag in a style sheet entry matches the initial portion of the language tag of an element. An exact match should be preferred. This interpretation allows an element marked up as, for instance, <cptr>en-US</cptr> to trigger styles corresponding to, in order of preference, US-english (<cptr>en-US</cptr>) or `international' English (<cptr>en</cptr>). <note> Using the language tag as a hierarchy does not imply that all languages with a common prefix will be understood by those fluent in one or more of the languages; it simply allows the user to request commonality when desired. </note> <note> It is intended that any new element introduced in later versions or extensions to ISO-HTML will admit the &a.lang; attribute, unless there is a compelling reason not to do so. </note> <p> The rendering of elements is meant to be controlled (in part) by the &a.lang; attribute. Specific user preferences set within a browser should override the &a.lang; attribute which in turn overrides the value specified by the &a.lang; attribute of any enclosing element. If none of these are set, a suitable default, perhaps controlled by the user's locale, should be used to control rendering. <h2 id=unknown>User agent unable to handle characters <p> If a user agent is unable to handle a character due, for example, to a lack of a resource such as a font, this should be indicated to the end user, clearly but unobtrusively. <nl> <li>Since some documents may contain many characters that cannot be rendered, showing an alert for each one is excessive. <li>If a numeric representation of a missing character is given, its hexadecimal, not decimal, form is to be preferred since this is the form used in character set standards. </nl> <h2 id=bytesex>Byte order <p> When an ISO-HTML text is transmitted directly in a multibyte representation, this &text; recommends: <ol> <li>That it be transmitted in big-endian byte order — high order byte first. <li>That the document always begin with a ZERO-WIDTH NON-BREAKING SPACE character (hexadecimal FEFF) which, when byte-reversed becomes number FFFE, a character guaranteed never to be assigned. Thus a user agent receiving an FFFE as the first two octets of a text would know that bytes have to be reversed for the remainder of the text. </ol> <h2 id=namedch>Use of named character references <p> It is often convenient for an author to use named character references to specify characters which are not available on the author's keyboard. Wherever named characters are used for characters specified by ISO 10646, the name used shall be that specified in ISO TR 9573. <note> ISO 9573 defines names for all the characters specified by ISO 10646. The intention of this requirement is to facilitate interchange. </note> <p> Example: <cptr><EM>Liberté Égalité Fraternité!</EM></cptr> <h1 id=useDTD>The DTD <p> The DTD provided by this &text; has a public document type definition with the invocation: <cptr keep=off type=source width=column> <!ENTITY % ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language//EN"> %ISO-HTML; </cptr> <h2 id=nosubset>Alternate public document type definition <p> Many popular user agents do not support the use this &text; makes of the document type declaration subset. To facilitate the use of ISO-HTML an alternate public document type definition is provided to specify a DTD which is totally self contained and contains all of the features of ISO-HTML without using the document type declaration subset. <cptr keep=off type=source width=column> <!ENTITY % ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language Core//EN"> %ISO-HTML; </cptr> <p> With the use of this public document type definition the minimum document becomes: <cptr keep=off type=source width=column> <!DOCTYPE ISO-HTML PUBLIC "ISO 15xxx:199x //DTD HyperText Markup Language Core//EN"> <ISO-HTML> <HEAD> <TITLE>Les unités de base</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <h1 id=copyr>Copyright protection <p> This document makes several references to industry and proprietary standards, products and publications. Such references are not normative, and do not imply endorsement by the ISO, IEC, or their national member bodies or affiliates. Any brand names or trademarks mentioned are the property of their respective owners. <p> The formal SGML definitions are part of the text of this &text; and are protected by copyright held by the IETF and the W3C. <h1 id=linking>Hyperlinks <note>&NB; This clause is based entirely on clause 7 in RFC 1866 HTML 2.0 <bibref refid=html2.0>. </note> <p> Hyperlinks provide the mechanism which ties the web together and gives it an overall structure. They are the threads of gossamer glueing documents together. They have two ends, known as the <hp1>source anchor</hp1> and the <hp1>target anchor</hp1> which are both identified by their address: an absolute Uniform Resource Identifier (URI) <bibref refid=uri>, optionally followed by a `<cptr>#</cptr>' character and a further sequence of characters called a fragment identifier. For example: <cptr keep=off type=source width=column> http://www.xyz.org/hypertext/WWW/TheProject.html http://www.xyz.org/hypertext/WWW/TheProject.html#Overview </cptr> <note>&NB; In RFC 1866, source anchors are called <hp1>tail</hp1> anchors, and target anchors are called <hp1>head</hp1> anchors as viewed from the current position in the hypertext document by a user agent. That is, prior to selecting a hypertext link its destination is seen as a target, but after selection and rendering the same position subsequently appears as a source. </note>  <p> In an anchor address, the URL refers to a resource; it may be used in a variety of information retrieval protocols to obtain an entity that represents the resource, such as an ISO-HTML document. The fragment identifier, if present, marks some position in the resource. <p> Each of the following constructs provides the target anchor of a hyperlink or set of hyperlinks: <ul compact> <li>The &a; element with the &a.href; attribute present. <li>The &img; element. <li>The &link; element. <li>The &isindex; element. <li>The &form; element with &a.method;<cptr>=get</cptr>. <li>The &input; element with the &a.src; attribute present. </ul> These markup constructs refer to source anchors by URL, in either absolute or relative form, or to a fragment identifier, or to both. <p> In the case of a relative URL, the actual address referenced by the user agent is constructed by combining the specified relative address with an absolute base URL as in RFC 1808 <bibref refid=relurl>. The base address is taken from the document's &base; element, if present; otherwise it is determined as in RFC 1808 <bibref refid=relurl>. <h2 id=access>Accessing resources <p> Once the address specified by the target anchor is determined, the user agent may obtain a representation of the resource (now the source). <p> For example, if the base address of a document is <cptr>http://host/x/y.html</cptr> and the document contains: <cptr keep=off type=source width=column> <IMG SRC="../icons/abc.jpeg"> </cptr> then the user agent uses the URL <cptr>http://host/icons/abc.jpeg</cptr> to access the resource. <note> The user agent constructs the destination URL by discarding the file identifier portion of the base address (if present). </note> <h2 id=activ>Activation of hyperlinks <p> An ISO-HTML user agent allows the user to navigate the content of a document and request activation of hyperlinks denoted by &a; elements. To activate a link, the user agent obtains a representation of the resource identified by the address specified in the source anchor. If the representation is another ISO-HTML document, navigation may begin again with this new document. <h2 id=simulimg>Simultaneous presentation of image resources <p> An ISO-HTML user agent may activate hyperlinks indicated by &img; elements concurrently with processing a document; that is, image hyperlinks may be processed without explicit request by the user. Image resources should be embedded in the presentation at the point of the source anchor, that is the &img; element. The same also holds for &input; elements within forms. <h2 id=fragid>Fragment identifiers <p> Any characters following a `<cptr>#</cptr>' character in a hypertext address constitute a fragment identifier. In particular, an address of the form <cptr>#fragment</cptr> refers to a target anchor in the document. <p> The meaning of fragment identifiers depends on the media type of the representation of the target anchor's resource. For <cptr>text/html</cptr> representations, it refers to the &a; element whose &a.name; attribute value is the same as the fragment identifier. Character matching is case sensitive. The target document shall have exactly one such element. The user agent should render the source anchor element by scrolling to and/or highlighting the text at the destination position. <p> For example, if the base URL is <cptr>http://host/x/</cptr> and the user activates the link denoted by the following markup: <cptr keep=off type=source width=column> <P>See: <A> HREF="app1.html#bananas">Appendix 1</A> for more details on bananas. </cptr> then the user agent accesses the resource identified by <cptr>http://host/x/app1.html</cptr> and assuming the resource is represented using the <cptr>text/html</cptr> media type, the user agent shall locate the &a; element whose &a.name; attribute has the value <cptr>bananas</cptr> and begin navigation at that point. <h1 id=addon>The extend and addon interfaces <p> This &text; does not define facilities for specifying style in the body of a conforming document. However, authors of simple documents may wish to make use of such a facility and suppliers of user agents may wish to add the facility to the ISO-HTML document architecture. This is possible, and conveniently done using one of the <hp1>extend</hp1> or <hp1>addon</hp1> interfaces. Many extensions and addons may be defined in the future: this clause defines the general <hp1>extend</hp1> and <hp1>addon</hp1> interface mechanism and provides an example of its use. <p> There are three types of interfaces available to document designers: <dl> <dt>New element interface <dd>The &head.extend; and &body.extend; interfaces which allow the addition of new elements and their attributes. <dt>Additional attribute interface <dd>The &a.addons; interfaces which allow the addition of further attributes to existing elements. <dt>Notation interface <dd>The use of <hp1>notations</hp1> as defined by ISO 8879:1986. </dl> <note> This &text; does not define any uses of the <hp1>extend</hp1> or <hp1>addon</hp1> interfaces, although an example is given to assist readers in their understanding of the SGML mechanisms used. </note> <note>&NB; This mention of the use of SGML notations does not extend the proposed ISO-HTML standard. The full power of SGML is available to all conforming applications [see Goldfarb 15.2.2 p.480]. </note> <h2 id=addon.conf>Conformance of extensions <p> No <hp1>extend</hp1> or <hp1>addon</hp1> shall cause a document to be non-conforming if that document is conforming in the absence of the <hp1>extend</hp1> or <hp1>addon</hp1>. <note> This &text; does not address the question of interdependencies between possibly incompatible extensions defined by other specifications. </note> <h2 id=extend>The new element interface <p> The &head.extend; and &body.extend; interfaces provide a means of extending the architecture of a document to include additional elements and attributes. The &head.extend; interface allows additions within the head of a document, at any point at which the &meta; element may be used. The &body.extend; interface allows additions within the body of a document at any point where an &a; element might be used. The technique used is to extend the content models which contain the &meta; and &a; elements. <p> By default the interfaces contain a &nop; element which has no semantic value, but the interface may be redefined to include an additional DTD fragment in the document architecture. <p> The procedure for adding new elements is as follows: <ol> <li>Place the definitions of the new elements and their attributes in a DTD fragment and assign a public text identifier. The definitions in this DTD fragment may make use of the entities defined in the ISO-HTML entity set. <note> The fragment may be in a single file, or in two or more files, for example one file for parameter entity definitions and another for element definitions. </note> <li>Place a new definition for the &head.extend; or &body.extend; parameter entity in the document type declaration subset. The definition should provide a content model for the new elements. Typically it is of the form <cptr>ELEMENT1 | ELEMENT2 ...</cptr>. The definition will override the default definition &nop; provided by the &text; DTD entity set [see Goldfarb p.404 line 32]. More than one extension may be defined in the &head.extend; or &body.extend; parameter entities with declarations of the form <cptr>EXTEND1.EL1 | EXTEND1.EL2 | ... | EXTEND2.EL1 ...</cptr>. <li>Place a parameter entity definition (or definitions) for the DTD fragment(s) containing definitions of the new elements and attributes in the document type declaration subset. <li>Invoke the &head.extend; or &body.extend; parameter entity. <li>Invoke the parameter entities for the DTD fragments at the end of the document type declaration subset. </ol> <h2 id=a.addon>The additional attribute interface <p> The &a.addons; interfaces provide a means of extending the set of attributes of an existing element. By default, the interfaces contain an empty string, ie. no additional attributes, but by redefining the interface, new attributes may be added to extend the document architecture. <p> The procedure for adding new attributes to an existing element X is as follows: <ol> <li>Place the definitions of the new attributes for the X element in an &X.addon; parameter entity in a DTD fragment and assign it a public text identifier. The definitions in this DTD fragment may make use of the entities defined in the ISO-HTML entity set. <note> This DTD fragment may be shared with the one defined in <hdref refid=extend page=yes>. </note> <p> This definition will override the default definition provided by the &text; DTD entity set [see Goldfarb p.404 line 32]. <li>Place a parameter entity definition for the DTD fragment in the document type declaration subset. <li>Invoke the parameter entities for the DTD fragments at the end of the document type declaration subset. </ol> <h2 id=subset>Use of the document type declaration subset <p> The extension techniques provided by this &text; make use of the document type declaration subset which is a part of all SGML documents. However many historic World Wide Web browsers do not process the subset, so it is recommended that for wide public use, the designer of an extension consider repackaging the extension with the parameter entity set and the document type definition provided by this &text; as a single entity (file) which can be referenced by a suitable public text identifier in the DOCTYPE declaration. <p> Such a repackaging will reduce the risk of error by users of the extension, and will allow the authority defining the package to provide better documentation and quality assurance, particularly when the package contains several extensions. <h2 id=nop.addon>The NOP attribute <p> A small number of the elements defined by this specification have a single &a.nop; attribute assigned to them. <p> This attribute is defined in the DTD to ensure that the DTD remains syntactically correct when the <hp1>addon</hp1> interface is not used. This attribute has no meaning and is not intended for use by authors or authoring systems. <h2 id=addonex>Example of addon definitions <p> <note>&NB; See the file <cptr>sample2.iso-html</cptr> for an example of a complete document type declaration subset. <p> These files will be cleaned up and placed in an informative annex. </note> <h1 id=a.box>Elements rendered as rectangles <p> The &boxing; elements defined by this &text; are usually rendered as rectangles with boxes around the visible area. Each of the elements which creates such a box is provided with an &a.align; attribute to specify how the rendered item is positioned with respect to the baseline of the text flow in which it occurs. A user agent may take into account only that text which has been placed on the line prior to the placement of the item in deciding the alignment of the enclosing box. <p> Each of the elements described in this clause use some of the attribute values from the following list: <dl> <dt>ALIGN=top <dd>Aligns the top of the rectangle with the top of the text rendered in the current line. <dt>ALIGN=texttop <dd>A permitted alternative specification for &a.align;=top. <dt>ALIGN=middle <dd>Aligns the middle of the rectangle with the current baseline. <dt>ALIGN=textmiddle <dd>Aligns the middle of the rectangle with a horizontal line drawn through the middle of the text rendered in the current line. <dt>ALIGN=bottom <dd>Aligns the bottom of the rectangle with the current baseline. <dt>ALIGN=baseline <dd>A permitted alternative specification for &a.align;=bottom. <dt>ALIGN=textbottom <dd>Aligns the bottom of the rectangle with a horizontal line drawn through the bottom of the text rendered in the current line. <dt>ALIGN=left <dd>Places the rectangle to the right of the current left margin, temporarily changing this margin. Subsequent text may be flowed along the rectangle's righthand side. <dt>ALIGN=center <dd>Places the rectangle after the end of the current line, centered between the left and right margins. Subsequent text starts at the beginning of the next line. <dt>ALIGN=centre <dd>A permitted alternative specification for &a.align;=center. <dt>ALIGN=right <dd>Places the rectangle to the left of the current right margin, temporarily changing this margin so that subsequent text is flowed along the rectangle's lefthand side. </dl> <note> The &caption; element and the justifying elements <hdref refid=a.justify page=yes> also have an &a.align; attribute, but in these cases its meaning is different. </note> <h1 id=a>The A element — Source and target anchors <h2 id=a.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT A - - (%text;)* -(A) > <!ATTLIST A CHARSET NAME #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED HREF %URL; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME CDATA #IMPLIED REL CDATA #IMPLIED REV CDATA #IMPLIED TITLE CDATA #IMPLIED %a.addon; > </cptr> <note>&NB; The attributes &a.rel; and &a.rev; are declared by &dave; to have type CDATA but &dan; declares them as NAMES. Which do we choose? </note> <h2 id=a.dsc>Description <p> The &a; element defines hyperlink source and target anchors (see clause <hdref refid=linking page=yes>). A source anchor requires the &a.href; attribute which provides the address of the target, while a target anchor requires the &a.name; attribute which identifies it as a location that may be used as a target in source anchors. At least one of the &a.href and &a.name attributes shall be present. <p> Good practice dictates that anchors should not be attached to markup, thus <cptr><H1><A NAME="xxx">Heading</A></H1></cptr> is preferred to <cptr><A NAME="xxx"><H1>Heading</H1></a></cptr>. <note> In the example given, the deprecated usage is also forbidden by the DTD, since it would not respect the required correct nesting of sections. </note> <p> Both start and end tags are required. <h2 id=a.att>Attributes <p> The attributes of the &a; element are: <dl> <dt>CHARSET <dd>A hint to the user agent concerning the character encoding scheme used by the resource pointed to by the link. It should be the appropriate value of the MIME charset parameter for that resource <bibref refid=mime>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>HREF <dd>Gives the URL of the source anchor of a hyperlink. For example: <cptr keep=off type=source width=column> The way to <A HREF="hands-on.html">happiness</A>. </cptr> <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>Gives the name of the anchor, and makes it available as a target for a hyperlink. The value shall be unique for the scope of the current document. For example: <cptr keep=off type=source width=column> <H2><A NAME=mit>Hacker's Paradise</A></H2> </cptr> <dt>REL <dd>The &a.rel; attribute gives the relationship(s) described by the hyperlink. The value is a whitespace separated list of relationship names. The semantics of link relationships are not specified in this document. <dt>REV <dd>Same as the &a.rel; attribute, but the semantics of the relationship are in the reverse direction. A link from A to B with <cptr>REL="X"</cptr> expresses the same relationship as a link from B to A with <cptr>REV="X"</cptr>. An anchor may have both &a.rel; and &a.rev; attributes. <dt>TITLE <dd>Suggests a title for the destination resource — advisory only. The &a.title; attribute may be used: <ul> <li>for display prior to accessing the destination resource, for example, as a margin note or in a small box while the mouse is over the anchor, or while the document is being loaded; <li>for resources that do not include a title, such as graphics, plain text and Gopher menus, for use as a window title. </ul> </dl> <h2 id=a.ex>Example <cptr keep=off type=source width=column> ... Please see <A NAME="intro" HREF="#details">here</A> for further details, and then <A HREF="http://www.acme.firm">go here</A> to apply for the job. ... <H3><A NAME="details">Job description</A></H3> ... <P> <A HREF="#intro">Return to introduction.</A> </cptr> <h1 id=address>The ADDRESS element — Author's address <h2 id=address.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % justify "(left|center|centre|right|justify)" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ADDRESS - - (%text;|P)+ > <!ATTLIST ADDRESS ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %address.addon; > </cptr> <h2 id=address.dsc>Description <p> The &address; element contains such information as address, signature and authorship, often at the beginning or end of the body of a document. <p> The &address element requires start and end tags. User agents should render the content with paragraph-breaks before and after. In the absence of a style-sheet, the &address; element should be rendered in an italic typeface and may be indented. <h2 id=address.att>Attributes <p> The &address; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=address.ex>Example <p> <cptr keep=off type=source width=column> <ADDRESS> Madame l'Éditrice en Chef<BR> Le Clarion du Pays<BR> 1, rue Rotative<BR> Ruritanie<BR> Téléphone: +800 l2 34 56 789 </ADDRESS> </cptr> <h1 id=area>The AREA element — Region in image map <h2 id=area.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % shape "(rect|circle|poly)" > <!ELEMENT AREA - O EMPTY > <!ATTLIST AREA ALT CDATA #REQUIRED COORDS CDATA #IMPLIED HREF %URL; #IMPLIED NOHREF (nohref) #IMPLIED SHAPE %shape; rect %area.addon; > </cptr> <h2 id=area.dsc>Description <p> The &area; element describes a region in an image map. It has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=area.att>Attributes <p> The &area; element has the following set of attributes: <dl> <dt>ALT <dd>Specifies a text label which may be displayed in a status line (when a pointing device is moved over the corresponding region in an image), or used for constructing a textual menu for non-graphical user agents. Authors are strongly recommended to provide meaningful &a.alt; attributes to support interoperability with speech-based or text-only agents. <dt>COORDS <dd>Specifies the size and position of a region in an image. See the &a.shape; attribute for further details. <dt>HREF <dd>Specifies the target hypertext link (behaviour) associated with a region. <dt>NOHREF <dd>Specifies that there is no target hypertext link (behaviour) associated with a region. This allows authors to create holes in regions. <dt>SHAPE <dd>Specifies the shape of a region and takes one of the following values: <dl> <dt>SHAPE=rect <dd>The region is rectangular and the &a.coords; attribute specifies the position of its top left and bottom right corners: &a.coords;="<hp1>left-x, top-y, right-x, bottom-y</hp1>", measured (in pixels) from the top left corner of the image. <dt>SHAPE=circle <dd>The region is circular and the &a.coords; attribute specifies the position of its centre and the length of its radius: &a.coords;="<hp1>centre-x,centre-y,radius</hp1>". <dt>SHAPE=poly <dd>The region is a polygon and the &a.coords; attribute specifies the position of its corners listed in clockwise order: &a.coords;="<hp1>x1,y1, x2,y2, ...</hp1>". </dl> If an <hp1>x</hp1> or <hp1>y</hp1> value is given with a percent sign as suffix, the value will be interpreted as a percentage of the image's width or height, respectively. For example, <cptr>SHAPE=rect COORDS="0,0, 50%,100%"</cptr>. </dl> <note>One of &a.href; or &a.nohref; shall be specified.</note> <h2 id=area.ex>Example <cptr keep=off type=source width=column> <AREA NOHREF SHAPE=circle COORDS="50%,50%,3" ALT="Bulls Eye"> </cptr> <p> See also the example of a &e.map; element <hdref refid=map page=yes>. <h1 id=b>The B element — Bold character style <h2 id=b.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "B | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=b.dsc>Description <p> The contents of a &b; element should be rendered in a bold text style. <p> Both start and end tags are required. <h2 id=b.att>Attributes <p> The &b; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=b.ex>Example <p> <cptr keep=off type=source width=column> Thank you for <B>NOT SMOKING</B> </cptr> <h1 id=base>The BASE element — Base URL specification <h2 id=base.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % URL "CDATA" > <!ELEMENT BASE - O EMPTY > <!ATTLIST BASE HREF %URL; #REQUIRED %base.addon; > </cptr> <h2 id=base.dsc>Description <p> The &base; element provides a base URL for dereferencing relative URLs, using the rules given by the URL specification <bibref refid=url>. <p> In the absence of a &base; element the document URL should be used. <note> This is not necessarily the same as the URL used to request the document, as the base URL may be overridden by an HTTP header accompanying a document. </note> <note>&NB; This clause appears to contain a normative reference to RFC 1738. How should this be handled? </note> <p> The start tag is required, but the end tag shall be omitted. <h2 id=base.att>Attributes <p> The &base; element has the following required attribute: <dl> <dt>HREF <dd>The value is a Uniform Resource Locator as defined by IETF RFC 1808 and IETF RFC 1738. </dl> <h2 id=base.ex>Example <p> Given the &base; specification <cptr keep=off type=source width=column> <BASE HREF="http://www.acme.com/intro.html"> </cptr> the image refernced by <cptr keep=off type=source width=column> <IMG SRC="icons/logo.jpeg"> </cptr> is dereferenced to <cptr keep=off type=source width=column> http://www.acme.com/icons/logo.jpeg </cptr> <h1 id=bdo>The BDO element — Bidirectional override <h2 id=bdo.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BDO - - (%text;)+ > <!ATTLIST BDO DIR %dirn; #REQUIRED LANG NAME #IMPLIED %bdo.addon; > </cptr> <h2 id=bdo.dsc>Description <p> The &bdo; element contains unusual pieces of text in which directionality cannot be resolved from the context in an unambiguous fashion. For example in part numbers, formulas, telephone numbers, punctuation and other texts whose directionality cannot be determined by context alone. <p> Both start and end tags are required. <h2 id=bdo.att>Attributes <p> The &bdo; element has the following attributes: <dl> <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=bdo.ex>Example <p> <cptr keep=off type=source width=column> <p>A palindrome appears the same whether rendered left-to-write: <BDO DIR=ltr>evil rats on no star live</BDO> or right-to-left: <BDO DIR=rtl>evil rats on no star live</BDO>. </cptr> <h1 id=big>The BIG element — Big character style <h2 id=big.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | BIG | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=big.dsc>Description <p> The contents of a &big; element should be rendered in a larger font than would otherwise be used. <p> Both start and end tags are required. <h2 id=big.att>Attributes <p> The &big; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=big.ex>Example <cptr keep=off type=source width=column> <BIG>What the large print giveth,</BIG> <SMALL>the small print taketh away.</SMALL> </cptr> <h1 id=blockquote>The BLOCKQUOTE element — Block quotation <h2 id=blockquote.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BLOCKQUOTE - - %section.content; > <!ATTLIST BLOCKQUOTE ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %blockquote.addon; > </cptr> <h2 id=blockquote.dsc>Description <p> The &blockquote; element contains a quotation. A typical rendering might be with a slight left and right indent, and/or an italic font. The &blockquote; usually provides space above and below the quoted text. <p> Single font renditions may reflect the quotation style on Internet mail by putting a vertical line of graphic characters, such as the greater than symbol (>), in the margin. <p> Both start and end tags are required. <h2 id=blockquote.att>Attributes <p> The &blockquote; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=blockquote.ex>Example <cptr keep=off type=source width=column> <BLOCKQUOTE>Afin de réaliser les objectifs visés à l'article 129B, la Communauté : <p> met en oeuvre toute action qui peut s'avérer nécessaire pour assurer l'interoperabilité des réseaux, en particulier dans le domaine de l'harmonisation des normes techniques ; </BLOCKQUOTE> Traité sur l'Union Européenne, Article 129 C. </cptr> <h1 id=body>The BODY element — Document body <h2 id=body.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT BODY - - (%section.content;, (H1,B1)* ) > <!ATTLIST BODY CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %body.addon; > </cptr> <h2 id=body.dsc>Description <p> The &body; element contains the textual flow of the document. <p> Both start and end tags are required. <h2 id=body.att>Attributes <p> The &body; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=body.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &body; element. <h1 id=br>The BR element — Line break <h2 id=br.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT BR - O EMPTY > <!ATTLIST BR NOP (nop) #IMPLIED %br.addon; > </cptr> <h2 id=br.dsc>Description <p> The &br; element specifies a line break between words. There is no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=br.att>Attributes <p> This &text; defines no attributes for the &br; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h2 id=br.ex>Example <cptr keep=off type=source width=column> <ADDRESS> ISO/IEC Copyright Office<BR> P.O. Box 56<BR> 1211 Geneva 20<BR> Switzerland </ADDRESS> </cptr> <h1 id=b1>The B1 element — Major section body <h2 id=b1.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B1 O O (%section.content;, (H2,B2)* ) > <!ATTLIST B1 NOP (nop) #IMPLIED %b1.addon; > </cptr> <h2 id=b1.dsc>Description <p> The &b1; element provides a formal definition for the body of a major section. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b1.att>Attributes <p> This &text; defines no attributes for the &b1; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b2>The B2 element — Section body <h2 id=b2.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B2 O O (%section.content;, (H3,B3)* ) > <!ATTLIST B2 NOP (nop) #IMPLIED %b2.addon; > </cptr> <h2 id=b2.dsc>Description <p> The &b2; element provides a formal definition for the body of a section. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b2.att>Attributes <p> This &text; defines no attributes for the &b2; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b3>The B3 element — Subsection body <h2 id=b3.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B3 O O (%section.content;, (H4,B4)* ) > <!ATTLIST B3 NOP (nop) #IMPLIED %b3.addon; > </cptr> <h2 id=b3.dsc>Description <p> The &b3; element provides a formal definition for the body of a subsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b3.att>Attributes <p> This &text; defines no attributes for the &b3; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b4>The B4 element — Subsubsection body <h2 id=b4.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B4 O O (%section.content;, (H5,B5)* ) > <!ATTLIST B4 NOP (nop) #IMPLIED %b4.addon; > </cptr> <h2 id=b4.dsc>Description <p> The &b4; element provides a formal definition for the body of a subsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b4.att>Attributes <p> This &text; defines no attributes for the &b4; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b5>The B5 element — Subsubsubsection body <h2 id=b5.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B5 O O (%section.content;, (H6,B6)* ) > <!ATTLIST B5 NOP (nop) #IMPLIED %b5.addon; > </cptr> <h2 id=b5.dsc>Description <p> The &b5; element provides a formal definition for the body of a subsubsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b5.att>Attributes <p> This &text; defines no attributes for the &b5; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=b6>The B6 element — Minor subsubsubsection body <h2 id=b6.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT B6 O O (%section.content;) > <!ATTLIST B6 NOP (nop) #IMPLIED %b6.addon; > </cptr> <h2 id=b6.dsc>Description <p> The &b6; element provides a formal definition for the body of a minor subsubsubsection. This is a part of the mechanism by which the DTD specifies the required nesting of sections, and thus a validating system will detect nesting errors. <p> Since there are many historic HTML browsers which do not handle the section body elements, it is recommended that authors and authoring systems make use of the permitted omission of both start and end tags. <h2 id=b6.att>Attributes <p> This &text; defines no attributes for the &b6; element that are available to the author of ISO-HTML documents, although the following attribute is present: <dl> <dt>NOP <dd>Please see <hdref refid=nop.addon page=yes>. </dl> <h1 id=caption>The CAPTION element — Table caption <h2 id=caption.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT CAPTION - - (%text;)+ > <!ATTLIST CAPTION ALIGN (top|bottom) #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %caption.addon; > </cptr> <h2 id=caption.dsc>Description <p> The &caption; element is used only in tables, where it contains a caption for a table. <p> Both start and end tags are required. <h2 id=caption.att>Attributes <p> The &caption; element has the following attributes: <dl> <dt>ALIGN <dd>The &a.align; attribute has value <cptr>top</cptr> or <cptr>bottom</cptr> and indicates the preferred placement of the caption with respect to the table. If the &a.align; attribute is omitted, browsers usually place the caption at the top of the table. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <note> &caption; is not a justifying element. Authors requiring horizontal text alignment should consider including a justifying element such as &p; within the content of the &caption;. </note> <h2 id=caption.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &caption; element. <h1 id=cite>The CITE element — Citation <h2 id=cite.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "CITE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=cite.dsc>Description <p> The &cite; element contains a citation or a reference to other sources. <p> Both start and end tags are required. <h2 id=cite.att>Attributes <p> The &cite; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=cite.ex>Example <cptr><CITE LANG="en-US">The buck stops here.</CITE></cptr> <h1 id=code>The CODE element — Program code <h2 id=code.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | CODE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=code.dsc>Description <p> The &code; element contains extracts from program code. <p> Both start and end tags are required. <h2 id=code.att>Attributes <p> The &code; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <note> The &a.lang; attribute is used for specifying human languages, not computer languages. </note> </dl> <h2 id=code.ex>Example <p> The list append function may be coded elegantly in Prolog as follows: <cptr keep=off type=source width=column> <CODE>append([],L,L) .</CODE> <CODE>append([X|L1],L2,[X|L3]) :- append(L1,L2,L3) .</CODE> </cptr> <h1 id=dd>The DD element — Definition data <h2 id=dd.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DD - O %section.content; > <!ATTLIST DD CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dd.addon; > </cptr> <h2 id=dd.dsc>Description <p> The ⅆ element is used only in definition lists <hdref refid=dl page=yes> and contains a definition for a term marked with a &dt; element. The definition is usually formatted as an indented paragraph after the term. <p> The start tag is required, but the end tag may be omitted. <h2 id=dd.att>Attributes <p> The ⅆ element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dd.ex>Example <p> See <hdref refid=dl page=yes> for an example of the use of the ⅆ element. <h1 id=dfn>The DFN element — Defining instance <h2 id=dfn.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | DFN | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=dfn.dsc>Description <p> The contents of the &dfn; element are the defining instance of a term. <p> Both start and end tags are required. <h2 id=dfn.att>Attributes <p> The &dfn; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dfn.ex>Example <p> <cptr keep=off type=source width=column> ICAO <DFN>International Civil Aviation Organization</DFN> responsible ... </cptr> <h1 id=div>The DIV element — Document division <h2 id=div.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT DIV - - %section.content; > <!ATTLIST DIV ALIGN %justify; #IMPLIED %div.addon; > </cptr> <h2 id=div.dsc>Description <p> The ÷ element may be used to structure documents into divisions. This &text; makes no recommendation for the presentation of divisions. <p> Both start and end tags are required. <h2 id=div.att>Attributes <p> The ÷ element has the following attribute: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. </dl> <h2 id=div.ex>Example <cptr keep=off type=source width=column> <BODY> ... <DIV> English Dictionary Terms <DL> ... </DL> </DIV> <DIV> Japanese Dictionary Terms <DL> ... </DL> </DIV> ... </BODY> </cptr> <h1 id=dl>The DL element — Definition list <h2 id=dl.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DL - - (DT|DD)+ > <!ELEMENT DT - O (%text;)+ > <!ELEMENT DD - O %section.content; > <!ATTLIST DL CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dl.addon; > </cptr> <h2 id=dl.dsc>Description <p> The &dl; element provides structure for a definition list: a list of terms and corresponding definitions. <p> The content is a non empty sequence of intermixed &dt; and ⅆ elements, usually in pairs. Multiple &dt; may be paired with a single ⅆ element and vice versa. <p> Definition lists are usually formatted with the term flushed to the margin and the definition, formatted paragraph style, indented to follow the term. If the &dt; term does not fit in the &dt; column (often one third of the display area), it may be extended across the page with the ⅆ content moved onto the next line, or it may be wrapped onto successive lines in the &dt; column. <p> Both start and end tags are required. <h2 id=dl.att>Attributes <p> The &dl; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &dt;, ⅆ groupings. The &a.compact; attribute may also imply a reduction of the width of the &dt; column. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dl.ex>Example <cptr keep=off type=source width=column> <DL compact> <DT>Center <DT>Centre <DD>A point equidistant from all points on the surface of a sphere <DD>In some field sports the player who holds the middle position on the field, court, or forward line </DL> </cptr> <h1 id=dt>The DT element — Definition term <h2 id=dt.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT DT - O (%text;)+ > <!ATTLIST DT CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %dt.addon; > </cptr> <h2 id=dt.dsc>Description <p> The &dt; element is used only in definition lists <hdref refid=dl page=yes> and contains a term to be defined. The term is typically formatted flush to the margin. If the &dt; term does not fit in the &dt; column (often one third of the display area), it may be extended across the page with the ⅆ content moved onto the next line, or it may be wrapped onto successive lines in the &dt; column. <p> The start tag is required, but the end tag may be omitted. <h2 id=dt.att>Attributes <p> The &dt; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=dt.ex>Example <p> See <hdref refid=dl page=yes> for an example of the use of the &dt; element. <h1 id=em>The EM element — Emphasized text <h2 id=em.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | EM | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=em.dsc>Description <p> The contents of the &em; element should be emphasized and are usually rendered in italics. <p> Both start and end tags are required. <h2 id=em.att>Attributes <p> The &em; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=em.ex>Example <cptr>Trespassers <EM>will</EM> be prosecuted!</cptr> <h1 id=form>The FORM element — Forms <h2 id=form.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % form.fields "INPUT | SELECT | TEXTAREA" > <!ENTITY % form.text "%text; | %form.fields;" > <!ENTITY % form.content "(%block; | %form.text;)+" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT FORM - - %form.content; -(FORM) > <!ATTLIST FORM ACTION %URL; #REQUIRED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ENCTYPE %Content-Type; "application/x-www-form-urlencoded" ID ID #IMPLIED LANG NAME #IMPLIED METHOD %HTTP-Method; get %form.addon; > </cptr> <h2 id=form.dsc>Description <p> The &form; element provides a template for a form data set and an associated method and action URL. A form data set is a sequence of name/value pair fields. The names are specified in the &a.name; attributes of form input elements, and the values are provided with initial values by various forms of markup, and may subsequently be edited by the user. The resulting form data set is used to access an information service as defined by the &a.action; and &a.method; attributes. <p> Form elements can be mixed in with document structuring elements. For example, a &form; element may contain lists which contain &input; elements. This gives considerable flexibility in designing the layout of forms. &form; elements do not contain other &form; elements. <p> Usually &form; elements contain a sequence of &input; elements along with additional document structure. <p> Both start and end tags are required. <h3 id=form.submit>Form submission <p> A user agent usually begins processing by presenting a document with the fields in their initial state. The user is allowed to modify the content of fields, constrained by the field type. When the user indicates that the form should be submitted (using for example a submit button), the form data set is processed as defined by its &a.method;, &a.action; URL and &a.enctype; attributes. <p> Where there is only one single-line text input field in a form, the user agent should accept "Enter" in that field as a request to submit the form. <h2 id=form.att>Attributes <p> The attributes of the &form; element are: <dl> <dt>ACTION <dd>Specifies the action URL for the form. The action URL of a form defaults to the base URL of the document <hdref refid=linking page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ENCTYPE <dd>Specifies the media type used to encode the name/value pairs for transport, in case the protocol itself does not impose a format. The default encoding for all forms is <cptr>application/x-www-form-urlencoded</cptr> A form data set is represented in this media type as follows: <ol> <li>The form field names and values are escaped; spaces are replaced by `<cptr>+</cptr>', and then reserved characters are escaped as per RFC 1738 <bibref refid=url>; that is non-alphanumeric characters are replaced by <cptr>%HH</cptr>, a percent sign and two hexadecimal digits representing the ASCII code of the character. Line breaks, as in multi-line text field values, are represented as CRLF pairs, i.e. %0D%0A. <note>&NB; How do we handle the reference to "ASCII"? </note> <li>The fields are listed in the order they appear in the document with the name separated from the value by `<cptr>=</cptr>' and the pairs separated from each other by `<cptr>;</cptr>'. Fields with null values may be omitted. In particular, unselected radio buttons and checkboxes should not appear in the encoded data, but hidden fields with &a.value; attributes present should. <note>&NB; RFC1866 allows the use of `<cptr>&</cptr>' instead of `<cptr>;</cptr>', but notes that this leads to ambiguities, since `<cptr>&</cptr>' is an SGML entity reference delimiter, and encourages the use of `<cptr>;</cptr>'. </note> </ol> <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>METHOD <dd>Selects a method for accessing the action URL. The set of applicable methods is a function of the scheme of the action URL specified for the form <hdref refid=form.submit page=yes>. The &a.method; attribute has one of the following values: <dl> <dt>METHOD=get — Forms with no server side effects <dd>If the processing of a form is idempotent (ie. it has no lasting observable effect on the state of the world), then the form method should be <cptr>get</cptr>. Many database searches have no visible side-effects and make ideal applications for query forms. <p> To process a form whose action URL is an HTTP URL and whose method is <cptr>get</cptr>, the user agent starts with the action URL and appends a `<cptr>?</cptr>' and the form data set, in <cptr>application/x-www-form-urlencoded</cptr> format. The user agent then traverses the link to this URL as if it were an anchor. <note> The URL encoding may result in very long identifiers, which cause some historical HTTP server implementations to exhibit defective behaviour. As a result, some ISO-HTML forms may be written using <cptr>METHOD=post</cptr> even though the form submission has no server side effects. </note> <dt>METHOD=post — Forms with server side effects <dd>If the service associated with the processing of a form has side effects (for example modification of a data base or subscription to a service), the method should be `post'. <p> To process a form whose action URL is an HTTP URL and whose method is <cptr>post</cptr>, the user agent conducts an HTTP `post' transaction using the action URL, and a message body of type <cptr>application/x-www-form-urlencoded</cptr> format. The user agent should display the response from the HTTP `post' interaction just as it would display the response from an HTTP `get'. </dl> </dl> <h2 id=form.ex>Example <cptr keep=off type=source width=column> <H1>Acme Company Inc</H1> <H2>Smallsville Agency</H2> <FORM METHOD=post ACTION="http://www.acme.firm/job"> <B>Job Application Form</B> <BR> Your Name: <INPUT TYPE=text NAME=name VALUE="Yosemite Sam"> <BR> Do you have a valid driver licence <BR> <INPUT TYPE=radio NAME=driver VALUE=yes>Yes <INPUT TYPE=radio NAME=driver VALUE=no CHECKED>No <BR> When you are finished, you may submit this application: <INPUT TYPE=submit VALUE="Apply"> <BR> You may clear the application form and start over at any time: <INPUT TYPE=reset VALUE="New Form"> <BR> <EM>Acme is an equal opportunity employer<EM> </FORM> </cptr> <h1 id=head>The HEAD element — Document header <h2 id=head.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT HEAD - - (TITLE & ISINDEX? & BASE?) +(LINK | META | %head.extend;) > <!ATTLIST HEAD CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %head.addon; > </cptr> <h2 id=head.dsc>Description <p> The &head; element defines the head of an ISO-HTML document, and contains an unordered collection of general information about the document. <p> Both start and end tags are required. <p> The content model of the &head; element also provides the &head.extend; interface. See <hdref refid=extend page=yes>. <h2 id=head.att>Attributes <p> The &head; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=head.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &head; element. <h1 id=hr>The HR element — Horizontal rule <h2 id=hr.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT HR - O EMPTY > <!ATTLIST HR ALIGN %justify; #IMPLIED DIR %dirn; #IMPLIED %hr.addon; > </cptr> <h2 id=hr.dsc>Description <p> The &hr; element specifies a horizontal separator in a document. Visual user agents usually present it as a full width horizontal rule, while speech-based user agents might present it as a pause. <p> The start tag is required, but the end tag shall be omitted. <h2 id=hr.att>Attributes <p> The &hr; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. </dl> <h2 id=hr.ex>Example <cptr keep=off type=source width=column> ... <H5>Monday Activities</H5> <P> ... <HR> <H5>Tuesday Activities</H5> <P> ... <HR> ... </cptr> <h1 id=h1>The H1 element — Major section header <h2 id=h1.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H1 - - (%text;)* > <!ATTLIST H1 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h1.addon; > </cptr> <h2 id=h1.dsc>Description <p> The &h1; element specifies the beginning of a major section of a document and contains the title of that major section. <p> Both start and end tags are required. <p> This &text; requires the correct nesting of sections. The &h1; element shall not be followed by an &h3;, &h4;, &h5;, or &h6; element without an intervening &h2; element. <h2 id=h1.att>Attributes <p> The &h1; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h1.ex>Example <cptr keep=off type=source width=column> <H1>Continent</H1> ... <H2>Country</H2> ... <H3>Province</H3> ... <H4>County</H5> ... <H5>City</H4> ... <H3>State</H3> ... <H4>City</H4> ... </cptr> <h1 id=h2>The H2 element — Section header <h2 id=h2.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H2 - - (%text;)* > <!ATTLIST H2 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h2.addon; > </cptr> <h2 id=h2.dsc>Description <p> The &h2; element specifies the beginning of a section of a document and contains the title of the section. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h2; element shall not be followed by an &h4;, &h5;, or &h6; element without an intervening &h3; element. An &h2; element shall be preceded by an &h1; element. <h2 id=h2.att>Attributes <p> The &h2; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h2.ex>Example <p> See <hdref refid=h1 page=yes> for an example of the use of header elements. <h1 id=h3>The H3 element — Subsection header <h2 id=h3.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H3 - - (%text;)* > <!ATTLIST H3 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h3.addon; > </cptr> <h2 id=h3.dsc>Description <p> The &h3; element specifies the beginning of a subsection of a document and contains the title of the subsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h3; element shall not be followed by an &h5; or &h6; element without an intervening &h4; element. An &h3; element shall be preceded by an &h1; element. <h2 id=h3.att>Attributes <p> The &h3; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h3.ex>Example <p> See <hdref refid=h1 page=yes> for an example of the use of header elements. <h1 id=h4>The H4 element — Subsubsection header <h2 id=h4.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H4 - - (%text;)* > <!ATTLIST H4 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h4.addon; > </cptr> <h2 id=h4.dsc>Description <p> The &h4; element specifies the beginning of a subsubsection of a document and contains the title of the subsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. The &h4; element shall not be followed by an &h6; element without an intervening &h5; element. An &h4; element shall be preceded by an &h1; element. <h2 id=h4.att>Attributes <p> The &h4; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h4.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=h5>The H5 element — Subsubsubsection header <h2 id=h5.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H5 - - (%text;)* > <!ATTLIST H5 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h5.addon; > </cptr> <h2 id=h5.dsc>Description <p> The &h5; element specifies the beginning of a subsubsubsection of a document and contains the title of the subsubsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. An &h5; element shall be preceded by an &h1; element. <h2 id=h5.att>Attributes <p> The &h5; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h5.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=h6>The H6 element — Minor subsubsubsection header <h2 id=h6.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT H6 - - (%text;)* > <!ATTLIST H6 ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %h6.addon; > </cptr> <h2 id=h6.dsc>Description <p> The &h6; element specifies the beginning of a minor subsubsubsection of a document and contains the title of the minor subsubsubsection. <p> Both start and end tags are required. <p> This &text; requires correct nesting of sections. An &h6; element shall be preceded by an &h1; element. <h2 id=h6.att>Attributes <p> The &h6; element has the following attributes. <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=h6.ex>Example <p> See <hdref refid=h1 page=yes> for an example of header elements. <h1 id=i>The I element — Italic character style <h2 id=i.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | I | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=i.dsc>Description <p> The contents of an &i; element should be rendered in an italic text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=i.att>Attributes <p> The &i; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=i.ex>Example <cptr>Following her <I LANG=fr>affaire</I> with the well known rugby player, ...</cptr> <h1 id=img>The IMG element — Inline images <h2 id=img.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT IMG - O EMPTY > <!ATTLIST IMG ALIGN (top|middle|bottom|left|right) #IMPLIED ALT CDATA #REQUIRED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED ISMAP (ismap) #IMPLIED LANG NAME #IMPLIED SRC %URL; #REQUIRED USEMAP %URL; #IMPLIED %img.addon; > </cptr> <h2 id=img.dsc>Description <p> The &img; element refers to an image or icon via a hyperlink, see <hdref refid=linking page=yes>. <p> Non-graphical user agents should process the value of the &a.alt; attribute as an alternative to processing the image resource indicated by the &a.src; attribute. <p> The &img; element has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=img.att>Attributes <p> The &img; attributes are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes> <dt>ALT <dd>The &a.alt; attribute is required, and specifies text that may be used in place of the referenced image resource. Use of this attribute is vital for interoperability with speech-based and text only user agents. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ISMAP <dd>If the &a.ismap; attribute is present in an &img; element, the &img; element shall be contained in an &a; element with an &a.href; attribute present. This construct represents a set of hyperlinks. The user can choose from the set by selecting a pixel on the image. The user agent computes the target URL by appending the character `<cptr>?</cptr> and the comma separated x and y coordinates of the selected pixel to the URL given in the &a; element. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>SRC <dd>The &a.src; attribute is required, and specifies the URL of the image resource. <note> In practice, the media types of the image resources are limited to a few raster graphic formats: typically <cptr>image/jpeg</cptr>, <cptr>image/gif</cptr>. In particular, <cptr>text/html</cptr> resources are not intended to be used for images. </note> <dt>USEMAP <dd>Specifies a URL fragment identifier for a client-side image map defined with the &e.map; element, see <hdref refid=map page=yes>. </dl> <h2 id=img.ex>Example <p> If the document contains: <cptr keep=off type=source width=column> ... <ISO-HTML><HEAD><TITLE>ImageMap Example</TITLE> <BASE HREF="http://host/index"></HEAD> <BODY> <P>Select a point of interest in the following image: <A HREF="/cgi-bin/imagemap"> <IMG ismap SRC="atlas.jpeg" ALT="Atlas of central Europe"> </A> ... </cptr> and the user selects the upper-leftmost pixel in the image, the chosen hyperlink has the URL <cptr>http://host/cgi-bin/imagemap?0,0</cptr> <h1 id=input>The INPUT element — User input field <h2 id=input.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % InputType "(checkbox|file|hidden|image| password|radio|reset|submit|text)" > <!ELEMENT INPUT - O EMPTY > <!ATTLIST INPUT CHECKED (checked) #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED MAXLENGTH NUMBER #IMPLIED NAME CDATA #IMPLIED SIZE CDATA #IMPLIED SRC %URL; #IMPLIED TYPE %InputType; text VALUE CDATA #IMPLIED %input.addon; > </cptr> <h2 id=input.dsc>Description <p> The &input; element is used in forms and represents a field for user input. <p> The start tag is required, but the end tag shall be omitted. <h2 id=input.att>Attributes <p> The &a.type; attribute of the &input; element discriminates between several different types of input field. The set of applicable attributes depends on the value of the &a.type; attribute as specified in the following subclauses. By default the value of the &a.type; attribute is TEXT. <h3 id=i.checkbox>TYPE=CHECKBOX <p> An &input; element with TYPE=CHECKBOX specifies a boolean choice. A set of &input; elements in the same &form; element with the same &a.name; attribute value represents an n-of-many choice. <p> The other attribute values are as follows: <dl> <dt>CHECKED <dd>Used to mark the corresponding input item as initially selected. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies the value to be returned if this input item is selected. </dl> <p> Example: <cptr keep=off type=source width=column> Which flavours do you like? <INPUT TYPE=CHECKBOX NAME=flavour VALUE=van CHECKED>Vanilla <BR> <INPUT TYPE=CHECKBOX NAME=flavour VALUE=str>Strawberry <BR> <INPUT TYPE=CHECKBOX NAME=flavour VALUE=pec CHECKED>Pecan <BR> </cptr> Note that more than one flavour may be checked. <h3 id=i.file>TYPE=FILE <p> An &input; element with TYPE=FILE provides a means for users to attach a file to a form's contents. It is typically rendered by a text field and an associated button which when selected invokes a file browser to select a file name. The file name can also be entered directly in the text field. See RFC 1867 for further details <bibref refid=file>. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Prefered size of the visible input line in average character widths. Users should be able to type more characters than this limit with the text scrolling through the field to keep the input cursor in view. </dl> <p> Example: <cptr keep=off type=source width=column> <FORM ACTION="http://server.rec/cgi/handle" ENCTYPE="multipart/form-data" METHOD=post> What is your name? <INPUT TYPE=TEXT NAME=submitter><BR> Which file are you sending? <INPUT TYPE=FILE NAME=pics> </FORM> </cptr> The user types "Joe Sixpack" in the name field, and selects an image file "football.jpeg" for the answer to 'Which file are you sending?'. <h3 id=i.hidden>TYPE=HIDDEN <p> An &input; element with TYPE=HIDDEN declares that fields should not be rendered — they are hidden from the user. The user does not interact with the field; instead, the &a.value; attribute specifies the value of the field. The &a.name; and &a.value; attributes are required, and are returned to the server when the form is submitted. <note> This input element may be used to handle state information in a form. </note> <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies a value for the field. </dl> <p> Example: <cptr keep=off type=source width=column> <INPUT TYPE=HIDDEN NAME=customer VALUE="827364"> </cptr> <h3 id=i.image>TYPE=IMAGE <p> An &input; element with TYPE=IMAGE specifies an image resource to display. If a point in the image is selected, the x and y coordinates of that point are returned to the server. The names of the fields used for returning the x and y pixel coordinates are the specified name of the field with ".x" and ".y" appended. <p> TYPE=IMAGE implies TYPE=SUBMIT processing, that is when a pixel is chosen, the form as a whole is submitted. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SRC <dd>This attribute is required and specifies the source of the image. </dl> <p> Example: <cptr keep=off type=source width=column> Choose a point on the map: <INPUT TYPE=IMAGE SRC="world.jpeg" NAME=place> </cptr> <p> When the user selects a point on the map, say with pixel coordinates (129,436), the input fields become <cptr>place.x=129,place.y=436</cptr>. See <hdref refid=form.submit page=yes>. <h3 id=i.password>TYPE=PASSWORD <p> An &input; element with TYPE=PASSWORD specifies a single line text field into which users may type a password. As the user types, the characters are usually echoed as `<cptr>*</cptr>' to hide the password from prying eyes. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Users should be able to type more characters than this limit with the password scrolling through the field to keep the input cursor in view. </dl> <p> Example: <cptr keep=off type=source width=column> Name: <INPUT NAME=login><> Password: <INPUT TYPE=PASSWORD SIZE=10 MAXLENGTH=8 NAME=passwd> </cptr> <h3 id=i.radio>TYPE=RADIO <p> An &input; element with TYPE=RADIO specifies a boolean choice. A set of &input; elements in a &form; element with the same &a.name; attribute value represents a 1-of-many choice. <p> The other attribute values are as follows: <dl> <dt>CHECKED <dd>Used to mark the corresponding input item as initially selected. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>VALUE <dd>This attribute is required and specifies the value to be returned if this input item is selected. </dl> <p> At all times one and only one of the radio buttons in a set is checked. Initially, if none of the &input; elements in a set of radio buttons specifies CHECKED, then the user agent shall mark the first radio button of the set as checked. <p> Example: <cptr keep=off type=source width=column> Which is your favourite flavour? <INPUT TYPE=RADIO NAME=flavour VALUE=van>Vanilla <BR> <INPUT TYPE=RADIO NAME=flavour VALUE=str>Strawberry <BR> <INPUT TYPE=RADIO NAME=flavour VALUE=pec CHECKED>Pecan <BR> </cptr> <h3 id=i.reset>TYPE=RESET <p> An &input; element with TYPE=RESET specifies an input option, usually represented by a button, that instructs the user agent to reset the form's fields to their initial states. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALUE <dd>Indicates a label for the input field. This value is usually displayed on a button. </dl> <p> Example: <cptr keep=off type=source width=column> When you are finished, you may submit this request: <INPUT TYPE=SUBMIT><BR> You may clear the form and start over at any time: <INPUT TYPE=RESET VALUE="Reset"> </cptr> <h3 id=i.submit>TYPE=SUBMIT <p> An &input; element with TYPE=SUBMIT represents an input option, typically a button, that instructs the user agent to submit the form. <p>The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>Indicates that this element contributes a form field whose value is given by the &a.value; attribute. If the &a.name; attribute is not present, this element does not contribute a form field. <dt>VALUE <dd>Indicates a label for the form field. This value is usually displayed on a button. </dl> <p> Example: <cptr keep=off type=source width=column> You may submit this request internally: <INPUT TYPE=SUBMIT NAME=request VALUE=internal> or externally: <INPUT TYPE=SUBMIT NAME=request VALUE=external> </cptr> <h3 id=i.text>TYPE=TEXT <p> An &input; element with TYPE=TEXT specifies a single line text field into which users may type a string. <p> The other attribute values are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MAXLENGTH <dd>The upper limit on the number of characters in the input field. <dt>NAME <dd>This attribute is required and specifies a name for this &form; field. <dt>SIZE <dd>Prefered size of the visible input line in average character widths. Users should be able to type more than this limit with the text scrolling through the field to keep the input cursor in view. <dt>VALUE <dd>This attribute is required and specifies an initial value for the text string initially shown in the field. </dl> <p> Example: <cptr keep=off type=source width=column> <INPUT TYPE=TEXT SIZE=40 NAME=user VALUE="Intentionally left blank"> </cptr> <h2 id=input.ex>Example <p> For examples of the use of the &input; element, please see subclauses <hdref refid=i.checkbox page=yes>, <hdref refid=i.file page=yes>, <hdref refid=i.hidden page=yes>, <hdref refid=i.image page=yes>, <hdref refid=i.password page=yes>, <hdref refid=i.radio page=yes>, <hdref refid=i.reset page=yes>, <hdref refid=i.submit page=yes> and <hdref refid=i.text page=yes>. <h1 id=isindex>The ISINDEX element — Keyword choice <h2 id=isindex.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ISINDEX - O EMPTY> <!ATTLIST ISINDEX CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED PROMPT CDATA #IMPLIED %isindex.addon; > </cptr> <h2 id=isindex.dsc>Description <p> The &isindex; element implies a set of hyperlinks. A user agent enables users to choose from this set by providing keywords. <note> User agents compute target URLs by appending the character `<cptr>?</cptr>' and the selected keywords to the base URL. Keywords are escaped according to RFC 1738 <bibref refid=url> and joined together by the character `<cptr>+</cptr>'. </note> <p> &form; elements with &a.method;<cptr>=get</cptr> also represent sets of hyperlinks. See clause <hdref refid=form page=yes> for details. <p> The start tag is required, but the end tag shall be omitted. <h2 id=isindex.att>Attributes <p> The &isindex; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>PROMPT <dd>A text which encourages a user to make a choice. </dl> <h2 id=isindex.ex>Example <p> If a document &head; element contains: <cptr keep=off type=source width=column> <BASE HREF="http://host/index"> <ISINDEX> </cptr> and a user selects the keywords <cptr>apple</cptr> and <cptr>berry</cptr>, then the user agent will access the resource <cptr keep=off type=source width=column> http://host/index?apple+berry </cptr> <h1 id=iso-html>The ISO-HTML element — Document instance <h2 id=iso-html.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT ISO-HTML - - (HEAD, BODY) > <!ATTLIST ISO-HTML CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VERSION CDATA #FIXED "&version;" %iso-html.addon; > </cptr> <h2 id=iso-html.dsc>Description <p> The &iso-html; element contains a document instance marked up in the language specified by this &text;. <p> Both start and end tags are required. <h2 id=iso-html.att>Attributes <p> The &iso-html; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VERSION <dd>The &a.version; attribute of the &iso-html; document element may be omitted, but if specified shall have the value <cptr>&version;</cptr>. </dl> <h2 id=iso-html.ex>Example <cptr keep=off type=source width=column> <ISO-HTML> <HEAD> <TITLE>Scientific Units</TITLE> ... <hp1>other head elements</hp1> </HEAD> <BODY> <H1>Scientific units and their conversions<H1> ... <hp1>document body</hp1> </BODY> </ISO-HTML> </cptr> <h1 id=kbd>The KBD element — Keyboard input <h2 id=kbd.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | KBD | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=kbd.dsc>Description <p> The &kbd; element contains text to be typed by a user. <p> Both start and end tags are required. <h2 id=kbd.att>Attributes <p> The &kbd; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=kbd.ex>Example <cptr keep=off type=source width=column> At the prompt, you might try typing: <KBD>rm -rf *</KBD> </cptr> <h1 id=li>The LI element — List item <h2 id=li.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT LI - O %section.content; > <!ATTLIST LI ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VALUE NUMBER #IMPLIED %li.addon; > </cptr> <h2 id=li.dsc>Description <p> The &li; element contains a list item in an ordered list <hdref refid=ol page=yes> or an unordered list <hdref refid=ul page=yes>. <p> The start tag is required, but the end tag may be omitted. <h2 id=li.att>Attributes <p> The &li; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALUE <dd>This attribute is optional and shall only be used in ordered lists. It provides an integer number for the ordered list item and re-sequences the numbering of the items in the ordered list. </dl> <h2 id=li.ex>Example <p> See <hdref refid=ol page=yes> and <hdref refid=ul page=yes> for examples of the use of the &li; element. <h1 id=link>The LINK element — Interdocument relations <h2 id=link.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT LINK - O EMPTY > <!ATTLIST LINK CHARSET NAME #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED HREF %URL; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED REL CDATA #IMPLIED REV CDATA #IMPLIED TITLE CDATA #IMPLIED %link.addon; > </cptr> <h2 id=link.dsc>Description <p> The &link; element provides a means for specifying relationships between this hypertext document and other documents and resources. <p> Any number of &link; elements may appear in the head of a document and are typically used to indicate authorship, related indexes and glossaries, older and more recent versions, document hierarchy and associated resources such as style sheets. <p> The start tag is required, but the end tag shall be omitted. <h2 id=link.att>Attributes <p> The &link; element has the following attributes: <dl> <dt>CHARSET <dd>Please see the &a.charset; attribute of the &a; element <hdref refid=a page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>HREF <dd>Please see the &a.href; attribute of the &a; element. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>REL <dd>Please see the &a.rel; attribute of the &a; element. <dt>REV <dd>Please see the &a.rev; attribute of the &a; element. <dt>TITLE <dd>Please see the &a.title; attribute of the &a; element. </dl> <h2 id=link.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &link; element. <h1 id=map>The MAP element — Client-side image map <h2 id=map.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT MAP - - (AREA)+ > <!ATTLIST MAP NAME CDATA #REQUIRED %map.addon; > </cptr> <h2 id=map.dsc>Description <p> The &e.map; element provides a mechanism for constructing client-side image maps. It acts as a container for a set of &area; elements which define individual regions in an image <hdref refid=area page=yes>. <p> Within a &e.map; careful attention should be given to the ordering of the &area; elements: If two or more regions overlap, then the first &area; element listed will take precedence over subsequent &area; elements, etc. Thus &area; elements with no associated action (which have the &a.nohref; attribute), should be specified in the &e.map; before &area;s defining hotzones (which have the &a.href; attribute). <p> Both start and end tags are required. <h2 id=map.att>Attributes <p> The &e.map; element has the following attribute: <dl> <dt>NAME <dd>This required attribute specifies a name for the map which may be referenced by the &a.usemap; attribute of an &img; element using a URL fragment identifier <hdref refid=img page=yes>. <note> The value of the &a.name; attribute is case sensitive. </note> </dl> <h2 id=map.ex>Example <cptr keep=off type=source width=column> <IMG SRC="USA.jpeg" USEMAP="#States"> ... <MAP NAME="States"> ... <AREA HREF="Florida.html" ALT="Florida" SHAPE=poly COORDS="..."> ... <AREA HREF="Texas.html" ALT="Texas" SHAPE=poly COORDS="..."> ... </MAP> </cptr> <h1 id=meta>The META element <h2 id=meta.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT META - O EMPTY > <!ATTLIST META CONTENT CDATA #REQUIRED HTTP-EQUIV NAME #IMPLIED NAME NAME #IMPLIED %meta.addon; > </cptr> <h2 id=meta.dsc>Description <p> The &meta; element is an extensible container for use in identifying specialized document meta-information. Meta-information has two main functions: <ul> <li>To provide a means to verify that the data set exists, and determine how it might be obtained or accessed. <li>To document the content, quality and features of a data set, indicating its fitness for use. </ul> Each &meta; element specifies a name/value pair. If multiple &meta; elements are provided with the same name, their combined contents — concatenated as a comma-separated list — is the value that will be associated with that name. <nl> <li>The &meta; element should not be used where a specific element, such as &title;, would be more appropriate. Rather than specifying a &meta; element with a URL as the value of the &a.content; attribute, use a &link; element instead. <li> HTTP servers may read the content of a document's &head; element to generate header fields corresponding to those elements defining a value for the attribute &a.http-equiv; <li> The method by which the server extracts document meta-information is unspecified and not mandatory. The &meta; element only provides an extensible mechanism for identifying and embedding such document meta-information — how it may be used is up to the individual server implementation and ISO-HTML user agent. </nl> <p> An HTTP server shall not use the &meta; element to form an HTTP response header unless the &a.http-equiv; attribute is present. <p> An HTTP server may disregard any &meta; elements that specify information controlled by the HTTP server, for example `Server', `Date', and `Last-modified'. <p> The start tag is required, but the end tag shall be omitted. <h2 id=meta.att>Attributes <p> The attributes of the &meta; element are: <dl> <dt>CONTENT <dd>This attribute is required and specifies the value part of a name/value pair. <dt>HTTP-EQUIV <dd>Binds the element to an HTTP header field. An HTTP server may use this information to process the document. In particular, it may include a header field in the responses to requests for this document: the header name is taken from the &a.http-equiv; attribute value, and the header value is taken from the value of the &a.content; attribute. HTTP header names are not case sensitive. <dt>NAME <dd>Specifies the name part of the name/value pair. If not present, then the &a.http-equiv; attribute is used to specify the name. </dl> <p> Either the &a.http-equiv; or the &a.name; attribute shall be present. <h2 id=meta.ex>Example <p> If a document contains: <cptr keep=off type=source width=column> <META HTTP-EQUIV="Expires" CONTENT="Sat, 1 Jan 2000 00:00:01 UTC"> <META HTTP-EQUIV="Keywords" CONTENT="Fred"> <META HTTP-EQUIV="Reply-to" CONTENT="Flinstones@bedrock.old"> <META HTTP-EQUIV="Keywords" CONTENT="Barney"> </cptr> then the server may include the following header fields: <cptr keep=off type=source width=column> Expires: Sat, 1 Jan 2000 00:00:01 UTC Keywords: Fred, Barney Reply-to: Flintstones@bedrock.old </cptr> as part of an HTTP response to a `get' or `post' request for that document. <h1 id=nop>The NOP element — Interface default value <h2 id=nop.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT NOP - O EMPTY > </cptr> <h2 id=nop.dsc>Description <p> The &nop; element has no content and no meaning. It shall be ignored by user agents, and shall not be used in conforming documents. <note> The &nop; element is used in the DTD to ensure syntactically correct content models for unused interfaces. </note> <h2 id=nop.att>Attributes <p> The &nop; element has no attributes. <h1 id=object>The OBJECT element — Simple agent <h2 id=object.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT OBJECT - - (PARAM | %section.content;)* > <!ATTLIST OBJECT ALIGN (texttop,middle,textmiddle,baseline, textbottom,left,center,centre,right) #IMPLIED CLASS NAMES #IMPLIED CLASSID %URL; #IMPLIED CODEBASE %URL; #IMPLIED CODETYPE CDATA #IMPLIED DATA %URL; #IMPLIED DECLARE (declare) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME %URL; #IMPLIED STANDBY CDATA #IMPLIED TYPE CDATA #IMPLIED USEMAP %URL; #IMPLIED %object.addon; > </cptr> <note>&NB; The &a.name; attribute is declared as having type <cptr>%URL;</cptr>. Is this correct? [Dix points]. </note> <h2 id=object.dsc>Description <p> The &object; element provides a technique for specifying richer behaviour for the document in the form of a simple agent. The scheme uses only elements and attributes and is simple to implement. <note> This technique deliberately avoids using sophisticated SGML techniques in order to facilitate its use by current user agents. Nothing in this &text; prevents the use of those SGML techniques by more sophisticated user agents. </note> <p> The specification of the &object; element supports the general insertion of media other than text into documents, but excludes the agent architecture, application programming interface (API) and inter-agent communication. <p> The agent may access data specified: <ul compact> <li>In the document text flow, <li>As a set of named properties, <li>In a separate file which is the target of a link specified within the document, <li>Using techniques outside the scope of this &text;. </ul> <p> A conforming ISO-HTML browser is not required to support the agent behaviour, and if it does not it shall render the <cptr>%section.content;</cptr> part of the content. Browsers which render the agent behaviour shall not render the <cptr>%section.content;</cptr>. If a browser chooses to render the <cptr>%section.content;</cptr>, then it may also display the content of the &a.alt; attribute to indicate that this choice has been made. <p> It is recommended that authors provide a suitable substitute for the agent behaviour in the <cptr>%section.content;</cptr> part of the content of the element. <p> The agents are normally loaded relative to the document URL (or &base; element if it is defined), and the &a.codebase; attribute may be used to change this default behaviour. If the &a.codebase; attribute is defined, then it specifies a different location for the agent resources. The value can be an absolute or relative URL. An absolute URL is used as is without modification and is not effected by a document's &base; element. When the &a.codebase; attribute is relative, then it is relative to the document URL (or &base; tag if defined). <p> Both start and end tags are required. <h2 id=object.att>Attributes <p> The &object; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>CLASSID <dd>Provides a URL which identifies an implementation for the agent. In some systems the value is a class identifier. The values are not defined by this &text;. If this attribute is omitted, a default value may be deduced from the &a.data; attribute value. <note> Since the class identifiers in some object systems may be cumbersome, the &a.classid; attribute may use a short URL to specify a class identifier indirectly. </note> <dt>CODEBASE <dd>Some URL schemes used to identify implementations require an additional URL to locate the implementation. This attribute provides the additional URL. The values are not defined by this &text;. <dt>CODETYPE <dd>The Internet media type of the code referenced by the &a.classid; attribute. <dt>DATA <dd>Provides a URL for the agent's data, eg. a JPEG file. The &a.type; attribute may provide the Internet Media Type for the data. In the absence of the &a.classid; attribute, the Internet Media Type of the data gives the default value to the &a.classid; attribute. <dt>DECLARE <dd>The &a.declare; attribute calls for late binding of the agent. The attribute is usually used to indicate that <ul compact> <li>¶m; elements are to be "object valued". <li>Hyperlinks point to objects which cannot otherwise be addressed with a single URL. </ul> Each such binding typically results in a separate instantiation of the agent. <cptr><OBJECT DECLARE></cptr> is treated as a declaration for making an instance of the agent. <p> If the declared agent is not supported, the user agent should use the content of the <cptr><OBJECT DECLARE></cptr> element which may contain another &object; element. The &a.type; attribute may be used to specify the Internet Media Type for the outer agent as a hint for this situation. <p> Anchors may exploit nested declared agents to provide alternative media for a given resource. <note>&NB; This explanation of the &a.declare; attribute is based on text that came from the W3C. It needs further work and some good examples. </note> <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. The &a.id; attribute is also available to assist inter agent communication. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>If the &a.name; attribute is present and if the &object; element is contained in a &form; element, then this agent participates in the form submission process. If &a.name; is present and the &a.declare; attribute is absent, the user agent should include the value of the &a.name; attribute and the data obtained from the agent along with information derived from other form fields, see <hdref refid=form.submit page=yes>. <note> The mechanism used to obtain an agent's data is specific to each agent. </note> <dt>STANDBY <dd>The attribute value provides a message to be displayed while the agent is being prepared. <dt>TYPE <dd>The Internet Media Type of the data referenced through use of the &a.data; attribute. <dt>USEMAP <dd>Specifies a URL fragment identifier for a client-side image map defined with the &e.map; element, see <hdref refid=map page=yes>. In general, the use of this attribute is only appropriate for static images. </dl> <h2 id=object.ex>Examples <note>&NB; Good examples are urgently needed. </note> <h1 id=ol>The OL element — Ordered list <h2 id=ol.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (OL|UL) - - (LI)+ > <!ATTLIST OL ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED START NUMBER #IMPLIED %ol.addon; > </cptr> <h2 id=ol.dsc>Description <p> The &ol; element represents a non empty ordered list of items, presented in sequence or order of importance. It is usually rendered as a numbered list. <p> The content is a sequence of &li; elements <hdref refid=li page=yes>. <p> Both start and end tags are required. <h2 id=ol.att>Attributes <p> The attributes of the &ol; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &li; elements. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>START <dd>Specifies an initial value for the item sequence number. If the &a.start; attribute is omitted, the default start value is 1. If specified, the value shall be an integer. </dl> <h2 id=ol.ex>Example <cptr keep=off type=source width=column> Easter Calendar <OL START=1997> <LI>30th March <LI>12th April <LI>4th April <LI>23rd April </OL> </cptr> <h1 id=option>The OPTION element — User choice <h2 id=option.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT OPTION - O (#PCDATA) > <!ATTLIST OPTION CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED SELECTED (selected) #IMPLIED VALUE CDATA #IMPLIED %option.addon; > </cptr> <h2 id=option.dsc>Description <p> The &option; element can only occur within a &select; element and represents one of the possible choices that is offered. The content of the &option; element is presented to users as a representation of the option. Its content is used as the returned value if the &a.value; attribute is not present. <p> The start tag is required, but the end tag may be omitted. <h2 id=option.att>Attributes <p> The attributes of the &option; element are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>SELECTED <dd>Indicates that the option is initially selected. <dt>VALUE <dd>Indicates the value to be returned if this option is chosen. If omitted, the returned value defaults to the content of the &option; element. <note> The name associated with this value is given by the &a.name; attribute of the parent &select; element. </note> </dl> <h2 id=option.ex>Example <p> See <hdref refid=select page=yes> for an example of the use of the &option; element. <h1 id=p>The P element — Paragraph <h2 id=p.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT P - O (%text;)+ > <!ATTLIST P ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %p.addon; > </cptr> <h2 id=p.dsc>Description <p> The contents of the &p; element form a paragraph. <note> Authors often start paragraphs on new lines to facilitate document maintenance. </note> <p> The start tag is required, but the end tag may be omitted. <h2 id=p.att>Attributes <p> The &p; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=p.ex>Example <cptr keep=off type=source width=column> <P> It was a dark and stormy night ... </cptr> <h1 id=param>The PARAM element — Agent interface parameter <h2 id=param.dtd>Formal definition <cptr keep=off type=source width=column> <!ELEMENT PARAM - O EMPTY> <!ATTLIST PARAM NAME NAME #REQUIRED TYPE CDATA #IMPLIED VALUE CDATA #IMPLIED VALUETYPE (data|ref|object) data %param.addon; > </cptr> <h2 id=param.dsc>Description <p> The ¶m; element appears only in &object; element content and specifies the name and value of a parameter to be passed to the agent <hdref refid=object page=yes>. The ¶m; element has no content. <p> The start tag is required, but the end tag shall be omitted. <h2 id=param.att>Attributes <p> The ¶m; element has the following attributes: <dl> <dt>NAME <dd>This attribute is required and specifies the name of the parameter. <note> Case sensitivity is implementation dependent. </note> <dt>TYPE <dd>This attribute is used only when the &a.valuetype; attribute has the value "ref" and specifies the Internet Media Type of the referenced value. <dt>VALUE <dd>Specifies the value of the parameter. For example: <cptr>VALUE="expensive"</cptr>. <dt>VALUETYPE <dd>Specifies how the &a.value; attribute is to be interpreted. Possible values are as follows: <dl> <dt>VALUETYPE=data <dd>The value "expensive" is to be passed directly to the agent as a string. In the absence of a &a.valuetype; attribute, "data" is the default behaviour for the &a.valuetype; attribute. <dt>VALUETYPE=object <dd>The value "expensive" is treated as the URL of an object or agent in the same document. This is used primarily for object or agent valued properties where the value of a property is a pointer (or reference) to a running object or to an agent. <dt>VALUETYPE=ref <dd>The value "expensive" is a URL. </dl> <note> The attribute value is obtained from the attribute value literal after replacing any entity references or character references within the literal and then normalizing the result by throwing out any entity ends and record starts and replacing any record end or separator characters with a space [see Goldfarb p.331]. </note> </dl> <h2 id=param.ex>Example <p> See <hdref refid=object page=yes> for an example of the use of the ¶m; element. <h1 id=pre>The PRE element — Pre-formatted text <h2 id=pre.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT PRE - - (%text;)+ -(BIG|IMG|MAP|SMALL|SUB|SUP|%body.extend;) > <!ATTLIST PRE CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED WIDTH NUMBER #IMPLIED %pre.addon; > </cptr> <h2 id=pre.dsc>Description <p> The &e.pre; element contains a non-empty block of text that is to be formatted as is, with its contents rendered in a mono-spaced font. <p> Within pre-formatted text, line breaks are rendered faithfully as line breaks. A user agent may use a constant indent when rendering pre-formatted text. <p> Anchor elements and phrase markup may be embedded in pre-formatted text. <note> Constraints on the processing of &e.pre; element content may limit the ability of a user agent to faithfully render phrase markup. </note> <p> Both start and end tags are required. <h2 id=pre.att>Attributes <p> The &e.pre; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>WIDTH <dd>Specifies the maximum number of characters in a line and allows a user agent to select a suitable font and indentation. The value of this attribute shall be a strictly positive integer. </dl> <h2 id=pre.ex>Example <cptr keep=off type=source width=column> <PRE WIDTH=16> Tic Tac Toe X | | O ----------- O | X | ----------- | | X </PRE> </cptr> <h1 id=q>The Q element — Quote <h2 id=q.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | Q | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=q.dsc>Description <p> The &q; element contains a quotation. Rendering is usually language and platform dependent. <p> Both start and end tags are required. <h2 id=q.att>Attributes <p> The &q; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=q.ex>Example <p> <cptr keep=off type=source width=column> A <Q LANG=de>quotation in German</Q> and a <Q LANG=fr>quotation in French</Q>. </cptr> might be rendered as: <p> A ,,quotation in German'' and a << quotation in French >>. <h1 id=samp>The SAMP element — Sample output <h2 id=samp.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | SAMP | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=samp.dsc>Description <p> The &samp; element contains sample output from a program or a script. <p> Both start and end tags are required. <h2 id=samp.att>Attributes <p> The &samp; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=samp.ex>Example <cptr keep=off type=source width=column> The installation begins: <samp>Insert diskette 1</samp> </cptr> <h1 id=select>The SELECT element — Selection in forms <h2 id=select.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT SELECT - - (OPTION)+ > <!ATTLIST SELECT CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED MULTIPLE (multiple) #IMPLIED NAME CDATA #REQUIRED SIZE NUMBER #IMPLIED %select.addon; > </cptr> <h2 id=select.dsc>Description <p> The &select; element defines one-from-many and many-from-many menus. The items from which the choice is made are defined by a sequence of &option; elements. <hdref refid=option page=yes> <p> At least one &option; shall be specified in a &select; element. <p> One-from-many menus are usually rendered as drop-down menus while many-from-many menus are usually rendered as list menus. <p> The initial state has the first option selected, unless a &a.selected; attribute is present on at least one of the &option; elements <hdref refid=option page=yes>. <p> Both start and end tags are required. <h2 id=select.att>Attributes <p> The &select; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>MULTIPLE <dd>The presence of this attribute signifies that the users can make multiple selections. If this attribute is omitted, at most one selection is allowed. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>NAME <dd>Specifies the name used to identify the menu choice when the form is submitted to a server. Each selected option results in a name/pair value being included in the form's contents. This attribute is required. <dt>SIZE <dd>Suggests the number of visible choices. </dl> <h2 id=select.ex>Example <cptr keep=off type=source width=column> <SELECT NAME="flavour"> <OPTION>Vanilla <OPTION SELECTED>Chocolate <OPTION VALUE="LemonLime">Lemon and Lime <OPTION>Strawberry </SELECT> </cptr> <h1 id=small>The SMALL element — Small character style <h2 id=small.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SMALL | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=small.dsc>Description <p> The contents of a &small; element should be rendered in a smaller font than would otherwise be used. <p> Both start and end tags are required. <h2 id=small.att>Attributes <p> The &small; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=small.ex>Example <cptr keep=off type=source width=column> <BIG>What the large print giveth,</BIG> <SMALL>the small print taketh away.</SMALL> </cptr> <h1 id=span>The SPAN element — Generic container <h2 id=span.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT SPAN - - (%text;)+ > <!ATTLIST SPAN CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %span.addon > </cptr> <h2 id=span.dsc>Description <p> The &span; element provides a generic container to carry language and bidirectional attributes in cases where no other element is appropriate. <p> Both start and end tags are required. <h2 id=span.att>Attributes <p> The optional attributes of the &span; element are as follows: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=span.ex>Example <cptr keep=off type=source width=column> The licence plate <SPAN DIR=ltr>CIA 1</SPAN> will appear correctly even if cut and pasted into the source of a report normally rendered in a right-to-left fashion. </cptr> <h1 id=strike>The STRIKE element — Strike-out character style <h2 id=strike.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | STRIKE | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=strike.dsc>Description <p> The contents of a &strike; element should be rendered in a "strike-through" text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=strike.att>Attributes <p> The &strike; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=strike.ex>Example <cptr keep=off type=source width=column> The original description should <STRIKE>NOT</STRIKE> have been omitted. </cptr> <h1 id=strong>The STRONG element — Strong emphasis <h2 id=strong.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | STRONG | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=strong.dsc>Description <p> The contents of a &strong; element are strongly emphasised and are usually rendered in a bold font. <p> Both start and end tags are required. <h2 id=strong.att>Attributes <p> The &strong; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=strong.ex>Example <cptr><EM>High Voltage <STRONG>Danger</STRONG></EM></cptr> <h1 id=sub>The SUB element — Subscript character style <h2 id=sub.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SUB | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=sub.dsc>Description <p> A &e.sub; element contains subscripted text. Its contents should be positioned below the base line and rendered in a smaller size than the preceding text. <p> Both start and end tags are required. <h2 id=sub.att>Attributes <p> The &e.sub; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=sub.ex>Example <cptr>H<SUB>2</SUB>O</cptr> <h1 id=sup>The SUP element — Superscript character style <h2 id=sup.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | SUP | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=sup.dsc>Description <p> A &e.sup; element contains superscripted text. Its contents should be positioned above the base line and rendered in a smaller size than the preceding text. <p> Both start and end tags are required. <h2 id=sup.att>Attributes <p> The &e.sup; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=sup.ex>Example <cptr>E = mc<SUP>2</SUP></cptr> <h1 id=table>The TABLE element — Tables <h2 id=table.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TABLE - - (CAPTION?, TR+) > <!ATTLIST TABLE ALIGN (left,center,centre,right) left BORDER %pixels; #IMPLIED CELLPADDING %pixels; #IMPLIED CELLSPACING %pixels; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED DUMMY (border) #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED WIDTH %length; #IMPLIED %table.addon; > </cptr> <h2 id=table.dsc>Description <p> The &table; element provides a widely deployed subset of the specification given in RFC 1942 <bibref refid=tables>. It is used to <ul compact> <li>Markup tabular material. <li>Describe tabular layout. <note> Authors should be aware that this latter use causes problems when rendering to speech or text-only agents. </note> </ul> <p> A table contains an optional caption <hdref refid=caption page=yes> followed by a non-empty sequence of table rows <hdref refid=tr page=yes>. <p> Both start and end tags are required. <h2 id=table.att>Attributes <p> The attributes of the &table; element are as follows: <note> The pixel values refer to screen pixels, and should be multiplied by an appropriate factor when rendering to high resolution output devices such as laser printers. For instance if a user agent has a display with 75 pixels per inch and is rendering to a laser printer with 600 dots per inch, then the pixel values given in the attributes should be multiplied by a factor of 8 prior to output. </note> <dl> <dt>ALIGN <dd>Please see <hdref refid=a.box page=yes>. <dt>BORDER <dd>Specifies the width of the outer border around a table in screen pixels, eg. &a.border;=4. The border may be suppressed by setting &a.border;=0. If both this attribute and the attribute &a.dummy; are omitted, the border should be suppressed. <dt>CELLPADDING <dd>Specifies the padding in screen pixels between a cell's contents and the border around the cell. <dt>CELLSPACING <dd>Each cell in a table has its own border which is separated from the borders around neighbouring cells. This separation can be set using the &a.cellspacing; attribute, eg. &a.cellspacing;=10 for a 10 pixel separation. This attribute also determines the separation between the table border and the borders of the outermost cells. <note> In traditional desktop publishing software, adjacent table cells share a common border. </note> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>DUMMY <dd>Allows an author to specify the attribute value BORDER as if it were an attribute name with no associated value. Such a specification requests that the border be drawn using the user agent's default width, usually one pixel. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>WIDTH <dd>Sets the table width. If an integer is specified, the unit is the screen pixel, eg. &a.width;=212, but if a percentage is specified by concatenating a "%" character as a suffix to the value, the table width is defined as a percentage of the space between the current left and right margins, eg. &a.width="80%". <p> If this attribute is omitted, the table width is automatically determined from the table contents. </dl> <h2 id=table.ex>Example <p> Tables take the general form: <cptr keep=off type=source width=column> <TABLE BORDER=3 WIDTH="64%" CELLSPACING=2 CELLPADDING=2> <CAPTION>Complexity of sorting algorithms</CAPTION> <TR><TH> Algorithm <TH> Complexity <TR><TD> Random shuffle <TD> O(exp(N)) <TR><TD> Bubble sort <TD> O(NxN) <TR><TD> Quicksort <TD> O(NlogN) <TR><TD> Heap sort <TD> O(NlogN) </TABLE> </cptr> <h1 id=td>The TD element — Table data cell <h2 id=td.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT (TH|TD) - O %table.content; > <!ATTLIST ( ... | TD ) ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COLSPAN NUMBER 1 DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NOWRAP (nowrap) #IMPLIED ROWSPAN NUMBER 1 VALIGN %valign; #IMPLIED %td.addon; > </cptr> <h2 id=td.dsc>Description <p> The &td; element is found only in &tr; table rows <hdref refid=tr page=yes> and contains a data cell in a table. <p> The start tag is required, but the end tag may be omitted. <h2 id=td.att>Attributes <p> The attributes of the &td; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLSPAN <dd>Specifies the number of columns spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NOWRAP <dd>The presence of this attribute disables automatic word wrap within the contents of the cell. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <note> This is equivalent to using the   entity for non-breaking spaces within the content of the cell. </note> <dt>ROWSPAN <dd>Specifies the number of rows spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents which overrides any value specified by the containing &tr; element <hdref refid=tr page=yes>. The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle", unless overridden by the containing &tr; element. The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <note> A table cell may contain a nested table. </note> <h2 id=td.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &td; element. <h1 id=textarea>The TEXTAREA element — Multi-line text field <h2 id=textarea.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TEXTAREA - - (#PCDATA) > <!ATTLIST TEXTAREA CLASS NAMES #IMPLIED COLS NUMBER #REQUIRED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NAME CDATA #REQUIRED ROWS NUMBER #REQUIRED %textarea.addon; > </cptr> <h2 id=textarea.dsc>Description <p> The &textarea; element is used in forms to represent the initial value of a multi-line text field which may be modified by a user. Its contents are restricted to text and character entities. <nl> <li> It is recommended that user agents canonicalize line endings to CRLF when submitting a field's contents to a server. <li> This &text; does not specify the coded character set to be used for submitting such text to a server. </nl> <p> The field is usually rendered in a mono-spaced font. <p> Both start and end tags are required. <h2 id=textarea.att>Attributes <p> The attributes of the &textarea; element are: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLS <dd>This required attribute specifies the prefered width of the field in visible mono-spaced characters. Users should be able to enter longer lines than this, so user agents should provide a means for scrolling through the text field when its contents extend beyond the visible area. User agents may wrap visible text lines to keep long lines visible without the need for scrolling. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NAME <dd>This required attribute specifies the name that is used to identify the field when the form is submitted to a server. <dt>ROWS <dd>This required attribute specifies the number of visible text lines. Users should be able to enter more lines that this, so user agents should provide some means for scrolling through the text field when its contents extend beyond the visible area. </dl> <h2 id=textarea.ex>Example <cptr keep=off type=source width=column> <TEXTAREA NAME=comments ROWS=3 COLS=32> I found that your product was excellent! </TEXTAREA> </cptr> <h1 id=th>The TH element — Table header <h2 id=th.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT (TH|TD) - O %table.content; > <!ATTLIST ( TH | ... ) ALIGN %justify;#IMPLIED CLASS NAMES #IMPLIED COLSPAN NUMBER 1 DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED NOWRAP (nowrap) #IMPLIED ROWSPAN NUMBER 1 VALIGN %valign; #IMPLIED %th.addon; > </cptr> <h2 id=th.dsc>Description <p> The &th; element is found only in &tr; table rows <hdref refid=tr page=yes> and contains a header cell in a table. &th; elements are distinguished from &td; elements so that a user agent may provide different rendering styles for cell header and cell data. <p> &th; is otherwise identical to &td; except that if the &a.align; attribute is not specified, the default alignment is "centre" unless overridden by the enclosing &tr; element. <p> The start tag is required, but the end tag may be omitted. <h2 id=th.att>Attributes <p> The &th; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes> for a general description. This use of the &a.align; attribute differs only in that the default value "centre" is to be used if none is supplied. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COLSPAN <dd>Specifies the number of columns spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>NOWRAP <dd>The presence of this attribute disables automatic word wrap within the contents of the cell. <note> This is equivalent to using the   entity for non-breaking spaces within the content of the cell. </note> <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>ROWSPAN <dd>Specifies the number of rows spanned by this cell. The value is a strictly positive integer, and if not specified, defaults to 1. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents which overrides any value specified by the containing &tr; element. <hdref refid=tr page=yes> The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle", unless overridden by the containing &tr; element. The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <h2 id=th.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &th; element. <h1 id=title>The TITLE element — Document title <h2 id=title.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT TITLE - - (#PCDATA) -(LINK | META | %head.extend;) > <!ATTLIST TITLE CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %title.addon; > </cptr> <note>&NB; Shouldn't the content model be replaced by RCDATA? </note> <h2 id=title.dsc>Description <p> The &title; element identifies the contents of a document in a global context. A short title such as "Introduction" may be meaningless out of context, and hence a title such as "Introduction to medieval bee-keeping" might be more appropriate. <note> The length of a title is not limited; however long titles may be truncated in some applications. To minimize this possibility titles should contain fewer than 64 characters. </note> <p> An ISO-HTML document shall contain one &title; element. <p> A user agent may display the title of a document in a history list or as a label for the window displaying the document's content. <p> Both start and end tags are required. <h2 id=title.att>Attributes <p> The &title; element has the following attributes: <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=title.ex>Example <p> See <hdref refid=iso-html page=yes> for an example of the use of the &title; element. <h1 id=tr>The TR element — Table row <h2 id=tr.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ENTITY % valign "(top|middle|bottom|baseline)" > <!ELEMENT TR - O (TH|TD)+ > <!ATTLIST TR ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED VALIGN %valign; #IMPLIED %tr.addon; > </cptr> <h2 id=tr.dsc>Description <p> The &tr; is found only in a &table; element <hdref refid=table page=yes> and specifies a non-empty row in a table. It acts as a container for &th; table header <hdref refid=th page=yes> and &td; table data <hdref refid=td page=yes> elements. <p> The start tag is required, but the end tag may be omitted. <h2 id=tr.att>Attributes <p> The &tr; element has the following attributes: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. <dt>VALIGN <dd>Specifies a default vertical alignment for cell contents. The permitted values are "top", "middle", "bottom" and "baseline". If not specified, the default vertical alignment is "middle". The meanings of the permitted values are the same as those described for the &a.align; attribute in <hdref refid=a.box page=yes>. </dl> <h2 id=tr.ex>Example <p> See <hdref refid=table page=yes> for an example of the use of the &tr; element. <h1 id=tt>The TT element — Mono-spaced character style <h2 id=tt.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | TT | ..." > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=tt.dsc>Description <p> The contents of a &tt; element should be rendered in a mono-spaced font. <p> Both start and end tags are required. <h2 id=tt.att>Attributes <p> The &tt; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=tt.ex>Example <cptr keep=off type=source width=column> The telex began: <TT LANG=fr>CONFIDENTIEL</TT> </cptr> <h1 id=u>The U element — Underlined character style <h2 id=u.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % physical.styles "... | U" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (%physical.styles;|...) - - (%text;)+ > <!ATTLIST (%physical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %physical.style.addon; > </cptr> <h2 id=u.dsc>Description <p> The contents of an &u; element should be rendered in an underlined text style if available, otherwise an alternative representation may be used. <p> Both start and end tags are required. <h2 id=u.att>Attributes <p> The &u; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=u.ex>Example <cptr keep=off type=source width=column> Browsers frequently <U>underline</U> the text associated with a link. </cptr> <h1 id=ul>The UL element — Unordered list <h2 id=ul.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (OL|UL) - - (LI)+ > <!ATTLIST UL ALIGN %justify; #IMPLIED CLASS NAMES #IMPLIED COMPACT (compact) #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %ul.addon; > </cptr> <h2 id=ul.dsc>Description <p> The &ul; element represents a non empty unordered list of items usually rendered as a bulleted sequence. <p> The content is a sequence of &li; elements <hdref refid=li page=yes>. <p> Both start and end tags are required. <h2 id=ul.att>Attributes <p> The attributes of the &ul; element are as follows: <dl> <dt>ALIGN <dd>Please see <hdref refid=a.justify page=yes>. <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>COMPACT <dd>The optional &a.compact; attribute suggests that a compact rendering be used, perhaps because the list items are small, or the entire list is large. Unless the &a.compact; attribute is present, a user agent may leave white space between successive &li; elements. <p> It is recommended that authors and authoring systems make use of the permitted omission of the attribute name, see ISO 8879:1986 subclause 7.9.1.2 [Goldfarb p.329 and p.70 line 20]. This will facilitate the use of historic browsers. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=ul.ex>Example <note>&NB; Full details to be provided by Italian NB. </note> <cptr keep=off type=source width=column> <P LANG=it>Gelati <UL> <LI>Vanilla <LI>Chocolate <LI>Strawberry </UL> </cptr> <h1 id=var>The VAR element — Generic variable <h2 id=var.dtd>Formal definition <cptr keep=off type=source width=column> <!ENTITY % logical.styles "... | VAR" > <!ENTITY % dirn "(ltr|rtl)" > <!ELEMENT (...|%logical.styles;) - - (%text;)+ > <!ATTLIST (%logical.styles;) CLASS NAMES #IMPLIED DIR %dirn; #IMPLIED ID ID #IMPLIED LANG NAME #IMPLIED %logical.style.addon; > </cptr> <h2 id=var.dsc>Description <p> The &var; element contains a placeholder variable, usually rendered in italic. <p> Both start and end tags are required. <h2 id=var.att>Attributes <p> The &var; element has the following attributes. <dl> <dt>CLASS <dd>Please see <hdref refid=a.class page=yes>. <dt>DIR <dd>Please see <hdref refid=a.dir page=yes>. <dt>ID <dd>Please see <hdref refid=a.id page=yes>. <dt>LANG <dd>Please see <hdref refid=a.lang page=yes>. </dl> <h2 id=var.ex>Example <cptr keep=off type=source width=column> Type <CODE>rm <VAR>file</VAR></CODE> to delete <VAR>file</VAR> from directory. </cptr> </body>  <annexn id=formal cols=1> <h0t>Formal definitions <note> These formal definitions are not intended to be self documenting. The elements and attributes are described in the body of this &text;. The SGML techniques are described in <bibref refid=goldfarb>. </note> <h1 id=dcl>The SGML declaration <cptr keep=off type=source width=page> iso-html.dcl </cptr> <h1 id=ent>The entity set <cptr keep=off type=source width=page> iso-html.ent </cptr> <h1 id=dtd>The elements and attributes <cptr keep=off type=source width=page> iso-html.dtd </cptr>  <annexi id=samples cols=1 page=no> <h0t>Sample marked up documents <note>&NB; Sample documents will not be provided until any corrections resulting from the CD balloting process have been folded into the text. </note> <annexi id=bibl cols=1 page=no> <h0t>Bibliography <bl> <bib id=ethno> <name>Grimes<cname>Barbara F. <pubtitle>Ethnologue, Languages of the World <pubedit>12th <publoc>Dallas <publish>Summer Institute of Linguistics <pubyear>1992 <pubibs>ISBN xxxxxxxxxxx <bib id=file> <name>Nebel<cname>Ernesto <name>Masinter<cname>Larry <pubtitle>Form-based File Upload in HTML <publoc>URL:ftp://ds.internic.net/rfc/rfc1867.txt <publish>Internet Engineering Task Force <pubyear>Novenber 1994 <pubref>RFC 1867 <bib id=goldfarb> <name>Goldfarb<cname>Charles F. <pubtitle>The SGML Handbook <pubedit>1st <publoc>Oxford <publish>Oxford University Press <pubyear>1990 <pubibs>ISBN 0-19-853737-9 <bib id=html2.0> <name>Berners-Lee<cname>Tim <name>Connolly><cname>Daniel <pubtitle>Hypertext Markup Language — 2.0 <publoc>URL:ftp://ds.internic.net/rfc/rfc1866.txt <publish>Internet Engineering Task Force <pubyear>1995 <pubref>RFC 1866 <bib id=html3.2> <name>Raggett<cname>Dave <pubtitle>HTML 3.2 Reference Specification REC-html32 <publoc>URL:http://www.w3.org/pub/WWW/TR <publish>World-Wide Web Consortium <pubyear>1997 <bib id=http> <name>Fielding<cname>Roy T. <name>Gettys<cname>Jim <name>Mogul<cname>Jeffrey C. <name>Nielsen<cname>Henrik Frystyk <name>Berners-Lee<cname>Tim <pubtitle>Hypertext Transfer Protocol — HTTP/1.1 <publoc>URL:ftp://ds.internic.net/rfc/rfc2068.txt <publish>Internet Engineering Task Force <pubyear>1997 <pubref>RFC 2068 <bib id=mime> <name>Borenstein<cname>N. <name>Freed<cname>N. <pubtitle>MIME (Multipurpose Internet Mail Extensions) Part One: Mechanisms for Specifying and Describing the Format of Internet Message Bodies.  <publoc>URL:ftp://ds.internic.net/rfc/rfc1521.txt <publish>Internet Engineering Task Force <pubyear>September 1993 <pubref>RFC 1521 <bib id=w3c.object> <name>Ragget<cname>Dave <name>Kindel<cname>Charlie <name>Montulli<cname>Lou <name>Sink<cname>Eric <name>Gramlich<cname>Wayne <name>Hirschman<cname>Jonathan <name>Berners-Lee<cname>Tim <name>Connolly<cname>Dan <pubtitle>Inserting objects into HTML, work in progress <publoc>URL:http://www.w3.org/pub/WWW/TR/WD-object.html <publish>World Wide Web Consortium <pubyear>1996 <pubref>WD-object-960422 <bib id=relurl> <name>Fielding<cname>Roy T. <pubtitle>Relative Uniform Resource Locators <publoc>URL:ftp://ds.internic.net/rfc/rfc1808.txt <publish>Internet Engineering Task Force <pubyear>June 1995 <pubref>RFC 1808 <bib id=rfci18n> <name>Yergeau<cname>François <name>Nicol<cname>Gavin <name>Adams<cname>Glenn <name>Dürst<cname>Martin <pubtitle>Internationalization of the Hypertext Markup Language <publoc>URL:ftp://ds.internic.net/rfc/rfc2070.txt <publish>Internet Engineering Task Force <pubyear>January 1997 <pubref>RFC 2070 <bib id=rfc1700> <name>Reynolds<cname>J <name>Postel<cname>Jon <pubtitle>Assigned Numbers <publoc>URL:ftp://ds.internic.net/rfc/rfc1700.txt <publish>Internet Engineering Task Force <pubyear>October 1994 <pubref>RFC 1700 <bib id=rfc1766> <name>Alverstrand<cname>H <pubtitle>Tags for the Identification of languages <publoc>URL:ftp://ds.internic.net/rfc/rfc1766.txt <publish>Internet Engineering Task Force <pubyear>March 1995 <pubref>RFC 1766 <bib id=tables> <name>Ragget<cname>Dave <pubtitle>HTML Tables <publoc>URL:ftp://ds.internic.net/rfc/rfc1942.txt <publish>Internet Engineering Task Force <pubyear>May 1996 <pubref>RFC 1942 <bib id=uri> <name>Berners-Lee<cname>Tim <pubtitle>Universal Resource Identifiers in WWW: A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web <publoc>URL:ftp://ds.internic.net/rfc/rfc1630.txt <publish>Internet Engineering Task Force <pubyear>1994 <pubref>RFC 1630 <bib id=url> <name>Berners-Lee<cname>Tim <name>Masinter<cname>Larry <name>McCahill<cname>M <pubtitle>Uniform Resource Locators (URL) <publoc>URL:ftp://ds.internic.net/rfc/rfc1738.txt <publish>Internet Engineering Task Force <pubyear>1994 <pubref>RFC 1738 </bl> <backm> <index> </standard>