🚀 go-pugleaf

In article <1992Aug12.050720.5458@news.eng.convex.com> connolly@convex.com (Dan Connolly) writes:
>I got this the other day, and I wanted to play with the
>toys that it mentions:

that was quick..


>The archive contains the _OSF Documentation SIG Recommendataions for
>the OSF DTD_. This includes some machine readable SGML files, including
>an SGML declaration and a few delaration subsets.
>
>I wanted to play around with them, but I couldn't get sgmls to
>digest the SGML delaration. First it complained about having
>comments before the SGML declaration. Then it complained about
>the character sets. Then it said something about only supporting
>fuction characters from the reference concrete syntax.
>
>This appears to be the emphasis of the OSF Doc SIG DTD. At
>least they have done significant work to separate the semantics
>from the presentation of the information. Their design is
>extensive and makes some interesting abstractions.
>
>Anybody else had a look at the design?
>
>Anybody else have an SGML system that can process documents
>that conform to the proposed DTD?


I work at OSF and have been involved in this work from its beginning
here.  I would like to clarify what you've seen in the recommendation,
and explain a little of the process that we're in the midst of.

OSF's Documentation SIG (Special Interest Group) is a group of OSF
members who have been meeting to discuss, among other things,
requirements and specifications of an SGML DTD that will meet the
needs of OSF's technical publications, mainly books.  Hopefully,
you've seen our "legacy documentation" in book stores somewhere,
if not everywhere (the OSF/Motif, OSF/1, and DCE books are published
by Prentice-Hall).

The way our SIGs work is to generate requirements and to feed
them into OSF, which generally has responsibility for implementation
of the requirements.  This is the process that you're seeing --
the SIG has made its recommendation visible outside of the OSF
family (in this case, through the Davenport consortium).

OSF is itself reading and evaluating the recommendation as it
works to implement the "official OSF DTD(s)."  We've participated
in most of the SIG's activities, and have contributed to a lot of
the work, so we're probably further along in understanding the
recommendation than people who have just seen it for the first
time, but we're still evaluating it.

OSF owes the SIG a response to the recommendation -- this will
come in the form of meetings, plans for implementation and
rollout of the implementation, DTD(s), and an explanation of
how well the DTD(s) fit the specific recommendation.  This will
happen over the next couple of months with the SIG.

Without jumping too far ahead of the situation here, I think I
can say that the OSF-implemented DTD(s) will meet the requirements
voiced in the recommendation very well.  However, it is
definitely too early to say how closely the prototype, "proof-
of-concept" DTD that you saw in appendix C of the Recommendation
will match the OSF-implemented DTD.  I suspect that, at least in
certain places, it may diverge significantly.  For example, we're
looking seriously at the Davenport/HyTime work and may use it
for implementing indexes, glossaries, etc.

We are now moving ahead with our implementation, and it won't be
long before it is done.  At that point, and after we've worked
through the OSF family evaluation of the implementation (including
review by the SIG), we'll make the OSF DTD(s) available for
comments by the larger SGML community (including, presumably and
among others, Davenport and this list).  At that point, you'll be
seeing the "official" OSF DTD(s), at least in beta form.

Eventually, after enough review and testing, OSF will make the
final DTD as widely and freely available as possible.  We're
hoping that will be in the first quarter of next year.

Please follow-up with me either through email or this newsgroup
if there are questions about this process.  Follow-ups on the
recommendation itself should be pursued with the author of the
recommendation (see the "Availability" section of the Preface).

Thanks--

Fred Dalrymple

fred@osf.org
617-621-8855
the Open Software Foundation