XML Base Architectures in SP

Date: Sun, 9 Apr 2000 15:53:35 -0500
From: "Steven R. Newcomb" <srn@techno.com>
Subject: XML Base Architectures in SP

[...]

You can now use SP to validate the conformance of XML documents to base architectures (meta-DTDs). TechnoTeacher has created a version of SP with full industrial-strength support for the alternative PI-based "Base Architecture Declaration" syntax. The enhancement builds on pioneering work done by Luis Martinez while he was working at TechnoTeacher, and it has recently been brought up to industrial strength by Peter Newcomb. Because of urgent need in certain industrial quarters (mortgage, healthcare, etc.), we've placed binaries of this version of SP at our FTP site:

ftp://ftp.techno.com/TechnoTeacher/SPt

There are builds for Linux, Solaris, and Windows. There is an announcement at . We're not making a lot of noise about this because we should have completed Luis's pioneering work long ago; it's just that Peter didn't have time to do it. Anyway, we now have industrial-strength validating parser support for inheritable architectures (meta-DTDs, base architectures, architectural forms, etc.) in XML, using SP. Hooray!

Unfortunately, there are as yet no really good examples and no installation instructions at the TechnoTeacher FTP site. We're assuming that anyone who is interested in inheritable architectures is sufficiently clueful that he or she can install executables and libraries on his or her own machine. In [the file] demo1.zip (and in the exactly equivalent demo1.tar.gz), there is a very simple example. (For those who are receiving copies of this note, the documentation for SP is available at http://www.jclark.com. The OpenSP website is http://peano.mathematik.uni-freiburg.de/OpenJade, but, as I write this, I can't get through to it.) In demo1, there is also some very ill-organized explanatory material about inheritable architectures that I have written for various purposes at various times over the past couple of years. Good luck with it. All those examples should be tested, now that we have a working test bench. They are probably full of dumb errors.

Our intent is to fold our SP enhancements into OpenSP, which is where they belong. It's a big job to do that, though, and Peter is one seriously overloaded software guy. I'm telling you this so you won't worry about our intentions. We have NO intention of creating anything that would compete with OpenSP. We fully intend to contribute our SP enhancements to OpenSP. Please excuse any delay in our making that contribution on the basis that we're simply unable to do that work immediately.

I spoke about inheritable architectures at the Healthcare Information Management Systems Society (HIMSS) meeting here in Dallas this morning, at the invitation of Consolidated Business Forms/VMP/Electronic Laser Forms. I mentioned the pioneering architectural work done by the Kona group of HL7. The inevitable question, "But isn't this all taken care of by XML Namespaces?" was asked at the end, as if on cue. I enumerated the ways in which XML Namespaces are bogus, concluding with: "When information interchange fails, XML Namespaces don't provide a way to point the finger of blame at the software that caused the failure." The questioner responded, "But then it's really an industry problem." I agreed with him, saying that the healthcare industry must take control of its own information architectures. It cannot expect system vendors, including consortia of systems vendors like the W3C, to act in the best interests of the healthcare industry voluntarily. The inheritable information architectures paradigm is the only way that the healthcare industry can hope to impose the rigor necessary to allow the finger of blame to be pointed, while at the same time accommodating its own ongoing evolution, complexity, and diversity. (It was fun to sing HIMSS on a Sunday morning.)

-Steve

Steven R. Newcomb, President, TechnoTeacher, Inc.
srn@techno.com  http://www.techno.com  ftp.techno.com

voice: +1 972 517 7954
fax    +1 972 517 4571

Suite 211
7101 Chase Oaks Boulevard 
Plano, Texas 75025 USA

[From "HyTime News and Events": "INHERITABLE INFORMATION ARCHITECTURES: Amendment 1 to ISO/IEC 10744:1997: http://www.ornl.gov/sgml/wg8/document/1985.htm (ISO/IEC SC34 document N1985). This amendment provides an XML-friendly PI-based syntax for declaring the use of SGML and XML inheritable architectures (this is the element subtyping paradigm that is also variously known as 'meta-DTDs', 'SGML architectures', 'architectural forms' or 'AFs', 'base architectures', 'client architectures', etc.):

     <?xml version="1.0"?>
     <?IS10744 arch
     name="somearch"
     public-id="+//IDN me.com//NOTATION Some Architecture//EN"
     dtd-public-id="+//IDN me.com//DTD Some Architecture//EN"
     doc-elem-form="somedoc"
     renamer-att="somedocnames"
     options="option1 option2"
     ?>
     <mydoc somearch="somedoc"/>

This new syntax is supported by a enhancements (created by Luis Martinez and Peter Newcomb of TechnoTeacher, Inc.) to the "SP" SGML/XML parser. Compiled versions for Linux, Solaris, and Windows are available at ftp://ftp.techno.com/TechnoTeacher/SPt. (It may interest some of you to know that this is the version of SP that is used in GroveMinder.) The SP parser itself was created by James Clark (www.jclark.com). The PI-based base architecture declaration syntax is also supported by David Megginson's XAF package..."]

Date: Tue, 11 Apr 2000 15:50:50 -0500
From: Peter Newcomb <peter@techno.com>
To: lubell@cme.nist.gov
Subject: Re: XML Base Architectures in SP

[Josh Lubell <lubell@cme.nist.gov> on Tue, 11 Apr 2000 16:24:43 -0400]
> I tried running the solaris binaries and got the following:
> fatal: libSPt.so.1.4.1: open failed: No such file or directory

Try adding the directory that contains libSPt.so.1.4.1 to your
LD_LIBRARY_PATH environment variable, or creating a LD_LIBRARY_PATH
environment variable whose value is the directory that contains the
library.

> Could you provide statically linked binaries?

If you really need them...  but it might be a while before I can do
it.

Alternatively, you could create a small script that sets
LD_LIBRARY_PATH appropriately.  For example, for nsgmls:

--- cut here ---
#!/bin/sh

SPt_DIR=/wherever/you/untarred/SPt

LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:${SPt_DIR}/lib"
export LD_LIBRARY_PATH

exec "${SPt_DIR}/bin/nsgmls" "$@"
--- cut here ---

BTW, the same problem can occur under Linux, and the same solution
applies.

-peter

Peter Newcomb                           TechnoTeacher, Inc.
peter@techno.com                        http://www.techno.com/


Date: Tue, 11 Apr 2000 17:47:27 -0400
From: Josh Lubell <lubell@cme.nist.gov>
To: Peter Newcomb <peter@techno.com>

Subject: Re: XML Base Architectures in SP

I amended my LD_LIBRARY_PATH, and the binaries now work. Thanks!

Josh
_____________________________________________________________________
           Joshua Lubell, NIST | joshua.lubell@nist.gov
   100 Bureau Drive, Stop 8260 | (301)975-3563
Gaithersburg MD 20899-8260 USA | http://www.nist.gov/msidstaff/lubell

Prepared by Robin Cover for The XML Cover Pages archive. For additional background and references, see "Architectural Forms and SGML/XML Architectures."

SEARCH Advanced Search ABOUT Site Map CP RSS Channel Contact Us Sponsoring CP About Our Sponsors NEWS Cover Stories Articles & Papers Press Releases CORE STANDARDS XML SGML Schemas XSL/XSLT/XPath XLink XML Query CSS SVG TECHNOLOGY REPORTS XML Applications General Apps Government Apps Academic Apps EVENTS LIBRARY Introductions FAQs Bibliography Technology and Society Semantics Tech Topics Software Related Standards Historic	XML Base Architectures in SP Date: Sun, 9 Apr 2000 15:53:35 -0500 From: "Steven R. Newcomb" <srn@techno.com> Subject: XML Base Architectures in SP [...] You can now use SP to validate the conformance of XML documents to base architectures (meta-DTDs). TechnoTeacher has created a version of SP with full industrial-strength support for the alternative PI-based "Base Architecture Declaration" syntax. The enhancement builds on pioneering work done by Luis Martinez while he was working at TechnoTeacher, and it has recently been brought up to industrial strength by Peter Newcomb. Because of urgent need in certain industrial quarters (mortgage, healthcare, etc.), we've placed binaries of this version of SP at our FTP site: ftp://ftp.techno.com/TechnoTeacher/SPt There are builds for Linux, Solaris, and Windows. There is an announcement at . We're not making a lot of noise about this because we should have completed Luis's pioneering work long ago; it's just that Peter didn't have time to do it. Anyway, we now have industrial-strength validating parser support for inheritable architectures (meta-DTDs, base architectures, architectural forms, etc.) in XML, using SP. Hooray! Unfortunately, there are as yet no really good examples and no installation instructions at the TechnoTeacher FTP site. We're assuming that anyone who is interested in inheritable architectures is sufficiently clueful that he or she can install executables and libraries on his or her own machine. In [the file] `demo1.zip` (and in the exactly equivalent `demo1.tar.gz`), there is a very simple example. (For those who are receiving copies of this note, the documentation for SP is available at http://www.jclark.com. The OpenSP website is http://peano.mathematik.uni-freiburg.de/OpenJade, but, as I write this, I can't get through to it.) In demo1, there is also some very ill-organized explanatory material about inheritable architectures that I have written for various purposes at various times over the past couple of years. Good luck with it. All those examples should be tested, now that we have a working test bench. They are probably full of dumb errors. Our intent is to fold our SP enhancements into OpenSP, which is where they belong. It's a big job to do that, though, and Peter is one seriously overloaded software guy. I'm telling you this so you won't worry about our intentions. We have NO intention of creating anything that would compete with OpenSP. We fully intend to contribute our SP enhancements to OpenSP. Please excuse any delay in our making that contribution on the basis that we're simply unable to do that work immediately. I spoke about inheritable architectures at the Healthcare Information Management Systems Society (HIMSS) meeting here in Dallas this morning, at the invitation of Consolidated Business Forms/VMP/Electronic Laser Forms. I mentioned the pioneering architectural work done by the Kona group of HL7. The inevitable question, "But isn't this all taken care of by XML Namespaces?" was asked at the end, as if on cue. I enumerated the ways in which XML Namespaces are bogus, concluding with: "When information interchange fails, XML Namespaces don't provide a way to point the finger of blame at the software that caused the failure." The questioner responded, "But then it's really an industry problem." I agreed with him, saying that the healthcare industry must take control of its own information architectures. It cannot expect system vendors, including consortia of systems vendors like the W3C, to act in the best interests of the healthcare industry voluntarily. The inheritable information architectures paradigm is the only way that the healthcare industry can hope to impose the rigor necessary to allow the finger of blame to be pointed, while at the same time accommodating its own ongoing evolution, complexity, and diversity. (It was fun to sing HIMSS on a Sunday morning.) -Steve Steven R. Newcomb, President, TechnoTeacher, Inc. srn@techno.com http://www.techno.com ftp.techno.com voice: +1 972 517 7954 fax +1 972 517 4571 Suite 211 7101 Chase Oaks Boulevard Plano, Texas 75025 USA [From "HyTime News and Events": "INHERITABLE INFORMATION ARCHITECTURES: Amendment 1 to ISO/IEC 10744:1997: http://www.ornl.gov/sgml/wg8/document/1985.htm (ISO/IEC SC34 document N1985). This amendment provides an XML-friendly PI-based syntax for declaring the use of SGML and XML inheritable architectures (this is the element subtyping paradigm that is also variously known as 'meta-DTDs', 'SGML architectures', 'architectural forms' or 'AFs', 'base architectures', 'client architectures', etc.): <?xml version="1.0"?> <?IS10744 arch name="somearch" public-id="+//IDN me.com//NOTATION Some Architecture//EN" dtd-public-id="+//IDN me.com//DTD Some Architecture//EN" doc-elem-form="somedoc" renamer-att="somedocnames" options="option1 option2" ?> <mydoc somearch="somedoc"/> This new syntax is supported by a enhancements (created by Luis Martinez and Peter Newcomb of TechnoTeacher, Inc.) to the "SP" SGML/XML parser. Compiled versions for Linux, Solaris, and Windows are available at ftp://ftp.techno.com/TechnoTeacher/SPt. (It may interest some of you to know that this is the version of SP that is used in GroveMinder.) The SP parser itself was created by James Clark (www.jclark.com). The PI-based base architecture declaration syntax is also supported by David Megginson's XAF package..."] Date: Tue, 11 Apr 2000 15:50:50 -0500 From: Peter Newcomb <peter@techno.com> To: lubell@cme.nist.gov Subject: Re: XML Base Architectures in SP [Josh Lubell <lubell@cme.nist.gov> on Tue, 11 Apr 2000 16:24:43 -0400] > I tried running the solaris binaries and got the following: > fatal: libSPt.so.1.4.1: open failed: No such file or directory Try adding the directory that contains libSPt.so.1.4.1 to your LD_LIBRARY_PATH environment variable, or creating a LD_LIBRARY_PATH environment variable whose value is the directory that contains the library. > Could you provide statically linked binaries? If you really need them... but it might be a while before I can do it. Alternatively, you could create a small script that sets LD_LIBRARY_PATH appropriately. For example, for nsgmls: --- cut here --- #!/bin/sh SPt_DIR=/wherever/you/untarred/SPt LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:${SPt_DIR}/lib" export LD_LIBRARY_PATH exec "${SPt_DIR}/bin/nsgmls" "$@" --- cut here --- BTW, the same problem can occur under Linux, and the same solution applies. -peter Peter Newcomb TechnoTeacher, Inc. peter@techno.com http://www.techno.com/ Date: Tue, 11 Apr 2000 17:47:27 -0400 From: Josh Lubell <lubell@cme.nist.gov> To: Peter Newcomb <peter@techno.com> Subject: Re: XML Base Architectures in SP I amended my LD_LIBRARY_PATH, and the binaries now work. Thanks! Josh _____________________________________________________________________ Joshua Lubell, NIST \| joshua.lubell@nist.gov 100 Bureau Drive, Stop 8260 \| (301)975-3563 Gaithersburg MD 20899-8260 USA \| http://www.nist.gov/msidstaff/lubell Prepared by Robin Cover for The XML Cover Pages archive. For additional background and references, see "Architectural Forms and SGML/XML Architectures."