Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!rpi!uupsi!sunic!ugle.unit.no!nuug!ifi!enag
From: enag@ifi.uio.no (Erik Naggum)
Newsgroups: comp.text.sgml
Subject: Re: What is SGML?
Message-ID: <ENAG.91Feb8030457@holmenkollen.ifi.uio.no>
Date: 8 Feb 91 02:05:05 GMT
References: <494sis-b@massey.ac.nz>
Sender: enag@ifi.uio.no (Erik Naggum)
Organization: Naggum Software, Oslo, Norway
Lines: 78
In-Reply-To: E.Ireland@massey.ac.nz's message of 6 Feb 91 22:35:06 GMT

Anybody want to take up the task of writing an authoritative answer to
this question?  Perhaps we need a comp.text.sgml FAQ, just like some
other groups?

Evan, this is of course not to deride your question, but it seems that
it's very difficult to write a good summary of what SGML is, and
spread it around sufficiently.  I'll try, after a brief pause (a.k.a.
sleep).  Stay tuned.

To answer your questions:

(1) ISO 8879:1986 + ISO 8879A1 (year forgotten).  You can order ISO
    standards from your national standards association or body.  I
    have no idea who this is for New Zealand.  However, see below
    (particularly the lines with --> on them).

(2) Forthcoming...

(3) SGML: The/An Author's Guide, Martin Bryan, Addison Wesley, 1987;
    a book which tries to encompass everything in the standard and
    doesn't succeed very well.  I wouldn't recommend this book, but
    it was one of the earliest out, and many people have read it.
    It's fixation on keeping the language and acronymitis of the
    standard itself is its main drawback.

    Practical SGML, Eric van Herwijnen (sp?), Kluwer, 1990; a more
    practical book (!) with a more limited scope and much more suc-
    cessful than the above at meeting its stated goal.  There are lots
    of small, annoying inaccuracies in it (I'm preparing a bug list),
    but you get a real good grip on SGML through it.  It's quite easy
    reading, comes in three parts for escalating user needs (SGML
    user, SGML application, SGML parser), but is a little "simplistic"
    on some of the difficult points.  It skips the things you don't
    (or won't) need to know, and this is very important with an uruly
    beast like SGML.

    The SGML Handbook, Charles F. Goldfarb, Oxford, 1991; an
    incredibly throrough book on the standard itself.  I've found one
    very minor (but consistent) cosmetic bug, and a reference to H.400
    which should have been X.400.  That's the bugs.  Now the reasons
    I've fallen in love with the book: ISO 8879 is a jungle.  ISO 8879
    has an amendment, no more than a diff in Unix parlance, which you
    have to consult to be certain of what "the" standard says.  ISO
    8879 has no index, no cross-references, separates the concrete
    reference syntax and the syntactic variables to an extent which
    makes you need a blackboard on which to scrawl "MDO = '<!', MSC =
    ']]', ...", or take numerous copies of pages from the standard and
    hang them on your wall and squint at, and you end up cluttering
    the standard with Post-It notes all over the place.  CFG has done
    all that for you, with an expert hand.  His book is the first I've
    seen with "paper hypertext", having "buttons" with page numbers
    and syntax production numbers on them, which makes it possible to
    look them up easily.  It also has small boxes with the reference
    concrete syntax representation of the delimiter roles in the right
    margin whenever the roles appear.  The book contains a heavily
    annotated, complete ISO 8879 plus amendment, plus suggested
    improvements.  The text is full of wit and humor, which is badly
    needed with ISO standards in general.  You'll find what you're
    looking for in this book, if your interest lies with the standard
    itself.  It is no beginner's book, however.

--> Several people have predicted that The SGML Handbook will replace
--> the ISO standard as the reference work of choice.  I fully support
--> this prediction.

    The bibliographies in Eric van Herwijnen's book leave little to be
    desired (except, perhaps, issue numbers for <TAG>The SGML News-
    letter), and the section on SGML sources in The SGML Handbook is
    nearly complete.  It's some time ago that I read Martin Bryan's
    book, but I was not that impressed with the bibliography, either.

I hope this has been of some help.  Your request for a "reasonably
comprehensive information" I shall attempt to fulfull by the end of
the day (T minus 21 hours, and counting...).

--
[Erik Naggum]					     <enag@ifi.uio.no>
Naggum Software, Oslo, Norway			   <erik@naggum.uu.no>