X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["6699" "Fri" "30" "June" "1995" "16:38:23" "+0200" "Frank Mittelbach" "mittelbach@MZDMZA.ZDV.UNI-MAINZ.DE" nil "154" "Re: SGML & LaTeX3" "^Date:" nil nil "6" nil nil nil nil] nil) Received: from MZDMZA.ZDV.UNI-MAINZ.DE (vzdmzi.zdv.Uni-Mainz.DE [134.93.8.15]) by trudi.zdv.Uni-Mainz.DE (8.6.12/8.6.12) with ESMTP id QAA27710 for ; Fri, 30 Jun 1995 16:43:34 +0200 Received: from DIRECTORY-DAEMON by MZDMZA.ZDV.UNI-MAINZ.DE (PMDF V4.3-12 #4432) id <01HSBO9P1O8G96W03X@MZDMZA.ZDV.UNI-MAINZ.DE>; Fri, 30 Jun 1995 16:43:29 +0100 Received: from mxrelay.gmd.de by MZDMZA.ZDV.UNI-MAINZ.DE (PMDF V4.3-12 #4432) id <01HSBO9N1W009AMLU7@MZDMZA.ZDV.UNI-MAINZ.DE>; Fri, 30 Jun 1995 16:43:27 +0100 Received: from vm.gmd.de by mxrelay.gmd.de (LSMTP for OpenVMS v0.1a) with SMTP id 10B7A650 ; Fri, 30 Jun 1995 16:43:17 +0200 Received: from VM.GMD.DE (NJE origin LISTSERV@DEARN) by VM.GMD.DE (LMail V1.2b/1.8b) with BSMTP id 8815; Fri, 30 Jun 1995 16:42:51 +0200 Received: from VM.URZ.UNI-HEIDELBERG.DE by VM.URZ.UNI-HEIDELBERG.DE (LISTSERV release 1.8b) with NJE id 3866 for LATEX-L@VM.URZ.UNI-HEIDELBERG.DE; Fri, 30 Jun 1995 16:40:18 +0000 Received: from DHDURZ1 (NJE origin SMTP@DHDURZ1) by VM.URZ.UNI-HEIDELBERG.DE (LMail V1.2a/1.8a) with BSMTP id 7848; Fri, 30 Jun 1995 16:40:09 +0000 Received: from kralle.zdv.Uni-Mainz.DE by vm.urz.Uni-Heidelberg.de (IBM VM SMTP V2R2) with TCP; Fri, 30 Jun 95 16:40:07 CET Received: (from uucp@localhost) by kralle.zdv.Uni-Mainz.DE (8.6.12/8.6.12) id QAA00455 for LATEX-L@VM.URZ.UNI-HEIDELBERG.DE; Fri, 30 Jun 1995 16:39:11 +0200 Received: (from latex3@localhost) by frank.zdv.uni-mainz.de (8.6.9/8.6.9) id QAA04039; Fri, 30 Jun 1995 16:38:23 +0200 In-reply-to: <199506281946.FAA14308@cs.anu.edu.au> Reply-to: Mailing list for the LaTeX3 project Message-id: <199506301438.QAA04039@frank.zdv.uni-mainz.de> X-Envelope-to: schoepf@goofy.zdv.uni-mainz.de MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT References: <199506281946.FAA14308@cs.anu.edu.au> Date: Fri, 30 Jun 1995 16:38:23 +0200 From: Frank Mittelbach Sender: Mailing list for the LaTeX3 project To: Multiple recipients of list LATEX-L Subject: Re: SGML & LaTeX3 Status: R X-Status: X-Keywords: X-UID: 1678 Richard writes: > In response to the previous posts, I've just read through the ltx3info > file and I found it very interesting. It does *not* suggest that the > input format will be SGML. right. as Nelson and others pointed out providing an sgml parser within TeX is - other than an interesting excerise - something which is rather impossible, if you want to have a general solution. what can be done is to have a cleaner separation between surface and internal syntax which would allow to replace LaTeX's top-level parser to accept some other syntax than \begin{foo} ... \end{foo} by just replacing the parser module. This could then very well allow for having a syntax that resembles sgml's concrete syntax as one possible form. but that does not mean that you could parse arbitrary sgml documents: you would still need to provide classes that can typeset the elements from the dtds for your documents and it is very likely that a lot of constructs allowed by the sgml standard could not be directly parsed. LaTeX's primarily goal is to typeset documents. it imposes structure on the document but due to its primary goal it is not so much concerned with enforcing that structure but concentrates on typesetting it. eg while most dtd's would probably disallow section headings in a footnote no current LaTeX class (and most probably no future one) will actually prevent a user of trying to do this. the result is most certainly strange or one might end up with some strange error but actually enforcing correct structure according to some general dtd would be very very inefficient in TeX. In other words, LaTeX does assume that a document for article class is well-structured according to some article-dtd (in fact the daphne project specified dtd's that correspond to the standard LaTeX classes of 209 (and thus basically 2e) some years ago) but if it isn't there is no well-defined reaction (you might get strange typesetting or you might get some TeX error or ...) i do believe that this is the right kind of separation. if you intend to typeset some sgml document then the sgml parsers task should be to validate your document and it should be the task of this application to translate it into something LaTeX understands. what we do intend to make easier is the translation task which (in case you do use a different surface parser for LaTeX) might actually be nearly a noop. > On the contrary, it states (I quote) > > It will provide a syntax that allows highly automated translation > from popular \SGML{} DTDs into \LaTeX{} document classes (these > will be provided as standard with the new version). > > The syntax of the new \LaTeX{} user-interface will, for example, > support the \SGML{} concepts of `entity', `attribute' and `short > reference' in such a way that these can be directly linked to the > corresponding \SGML{} features. > > This suggests to me that it will be good old LaTeX as before (or close > to it), but with some added commands to support the SGML concepts in a > clean way. this summary by Richard is basically what we think > And further: > > It will provide an enhanced user-interface that allows expression of > the typesetting requirements from a large range of subject areas. Some > examples are listed here. > > (Sounds like a sales pitch?) yeah, anything wrong with that? it also tells you where to send your contribution to if you want to support LaTeX2e maintenance and LaTeX3 development :-) > \begin{itemize} > \item The requirements of technical documentation (\eg offset layout,\\ > change bars, etc). > \item The requirements of academic publishing in the humanities\\ > (critical text editions, etc). > \item The requirements of structural formulas in chemistry. > \item Advanced use of the mathematical-typesetting features of \TeX{}. > \item The integration of graphical features, such as shading, within text. > \end{itemize} > > (A critical text edition of the source of LaTeX would be an > interesting project :-) let's get it first correctly documented. then we can think of somebody writing a critical text edition of it :-) > It seems to me that you can't do this in pure SGML. If you could, you > wouldn't need LaTeX, would you? SGML and LaTeX do serve completely different purposes, right but i really don't want to expand on that topic as it doesn't really belong here. > I even wonder if all of the aims can be achieved in TeX: > > Error handling will be improved by adding > a more effective, interactive help system. > > Interactive help system? This sounds almost as if a graphical user > interface is planned - or at least that you will have to run it under > Emacs :-P no actually not. we did some tests with using a "ascii terminal" based one. > And this: > > It will provide access to arbitrary fonts from any family (such as > the \PS{} and TrueType fonts) including a wide range of fonts for > multi-lingual documents and the specialist glyphs required by > documents in various technical and academic areas. > > This suggests that LaTeX3 will be based on Omega (or a successor), not > TeX. This would seem to be a little dangerous (particularly since > Omega doesn't work yet - and it's not clear when it will). well actually i guess we meant to say something like "for any font for which you can obtain tfm files" and in fact with input/internal/output encoding interface LaTeX2e already holds up on this quite well > The resulting new \LaTeX{} will, like the present version, > be usable with any standard > \TeX{} system and so will be freely available on a wide range > of platforms. > > I don't believe it. Either it won't use TeX or it won't live up to > all these promises. (Or the team will prove me wrong and it will run > like a dog.) well, perhaps not on a small emTeX (where already 2e has a problem) > I haven't been on NTS-L for ages, so perhaps someone from NTS can > suggest how that project relates to this proposed LaTeX3? as Phil said they are independent. we don't intent to make use of features provided by Omega/NTS... unless one of those programs finally becomes a defacto standard in the TeX world. However, depending on the status of such projects we might actually emulate some of the features they provide or do support alternative coding so that the use of, say, e-tex would be beneficial for the user and thus make the acceptance of a TeX successor faster. but our major goal will be to provide LaTeX3 in a form suitable for the normal LaTeX user. > Richard (enjoying summer (at last) in Germany!). where? frank LaTeX3 project