This is Lorrin Garson at the ACS. Bob Kelly has kindly put me on the sgml@aps.com listserv. This is my introductory message. The ACS has been working about five years on developing a DTD for journals. We started with the AAP standards and worked from there. Our latest activity was to contract with Atlis Consulting to develop the DTD structures for tabular material, based on a careful analysis of our tables and how they are handled in Xyvision. That work has been completed. Tim Ingoldsby at the AIP kindly provided us with their DTD structures for math, which they also derived from an analysis of Xyvision math. Thus, we now have a complete DTD. Our next effort for the journals SGML effort is to create a program to translate from Xyvision file format to the corresponding SGML encoded format, and reverse. This work needs to be completed by the end of this year for electronic product production in 1995. We currently publish 23 journals collectively containing ~100,000 typeset pages. After we have developed the kernel of this translation program, and are reasonably satisfied that we can go back and forth between Xyvision and SGML, i.e., there truly is a correspondence between all the data elements on both sides, we plan to make the DTD available to the public. We did consider adopting ISO 12083, but felt that the DTD we had developed, which was based on careful analysis of our journal data, would better serve our needs. Incidently, the ACS has been preserving its journal data in a database structure since 1975. These data are not in SGML, but nevertheless highly structured. For example, we have every paragraph and sentence uniquely identified, as well as authors, abstracts, footnotes, references, etc. Our journals DTD development work has been done in part with cooperation with the Royal Society of Chemistry in Great Britain, the Chemical Society of Japan and the AIP. Not that any of these organizations will adopt what we have developed, but we wanted and appreciate their input. As best I know, the RSC, CSJ and Elsevier have decided not to adopt ISO 12083 at this time. Other SGML work we have been involved with is as follows: (1) We translate Chemical & Engineering News files on Xyvision to SGML. This is a weekly news magazine and has a completely different DTD than that for the journals. The SGML files are subsequently processed and data are loaded weekly on STN Internationsl as the file CEN. (2) In July we will release two CD-ROMs, for the journals Biochemistry and the Journal of the American Chemical Society. For this effort we are using SGML, but with an abridged DTD from the one described above. (3) For the CORE project, our archival journal data has been translated to SGML (yet again a different DTD, one developed specifically for this project). The objective of this project, which is being done in cooperation with Bellcore, Cornell University, OCLC and our colleagues at Chemical Abstracts Service, is to create a prototype electronic library at Cornell University. (4) We have created DTDs and SGML files for a number of other products and publications, mostly "throw away" items, but have found this a worthwhile method for handling data. Enough of an introduction. I look forward to hearing from others on the list. Lorrin -- ************************** FROM: Lorrin R. Garson ************************** Publications Division, American Chemical Society, Washington, D.C. E-mail: lrg96@acs.org Phone: (202) 872-4541 FAX (202) 872-4389 *****************************************************************************