xml:id RE: [sc34wg3] Compact syntax requirement question

[Mason, James David (MXM)]

LMG wrote, speaking of BNF: The one exception I'm aware of is SGML, and =
that
really is an anomaly that took a lot of flak for being an anomaly.=20


I think that had we had, a quarter of a century ago, a decent =
understanding
of BNF across the committee and an agreed syntax, we would have used it. =
At
that point, BNF was known in theoretical circles, and some of us =
stumbled
across it, but SGML was largely written by people not in contact with =
the hot
theories. So we created our own formalism over the course of several =
years.
(If we'd been into fashionable theories of the day, we'd have done ODA =
rather
than SGML. ODA did eventually attempt a kind of BNF in their =
specification,
but only after I ridiculed them for not having even so much formalism as =
our
homemade production grammar from 8879.) At this point, all I can say is =
that
we won both the battles and the war, even with anomalous techniques.

What is most anomalous in 8879, however, is not its production grammar =
but
the separate lanugage for DTDs. We looked at doing it in SGML, but by =
the
time we got to that, we were at such a late stage in ISO processing that =
it
would have meant major delays (and we were in a race with ODA to get a
standard approved).

I think many of the evils of the present XML world stem from our not =
having
gone back and recast DTDs into SGML. That became one of the first open =
(as
opposed to covert) excuses for creating W3 Schema. And so we have
proliferating schema languages, with this committee in the midst of the =
fray.

Just as some of the proponents of CTM say they can write LTM or AsTMa=3D =
a
whole lot faster than they can write XTM, I can say I write DTDs a whole =
lot
faster than I can write W3 Schema. (Indeed, I don't write W3 Schema at =
all.
If I find I need one of those things, I write a DTD and translate it.)

But I don't think that's an ideal state. I'd rather be using an XML =
notation
for my schemas. That isn't likely to happen soon because (1) I'd have to
become more proficient in something else (like ISO/IEC 19757-2), (2) I'd =
have
to change a lot of software that's written for DTDs, and (3) I'd have to
change the user community in which I work, because they're DTD centered. =
(And
of those three, I'd say the last would be the most difficult.)

I'm generally opposed to standardizing new syntaxes unless there are
technical reasons why it is unavoidable. Verbosity is not a technical =
reason.
For this reason, I have very little enthusiasm for CTM. I likewise have
little enthusiasm for a TMQL that is not in XML syntax. If you can't =
notate
it in XML, is it even in the scope of SC34?

Verbosity is not in general an excuse for not using XML. I will readily =
admit
that XSLT is hideously verbose, and I paid good money for an editor that
speeds up creation of the stuff. But I'll put up with that. (I didn't =
learn
DSSSL because I didn't want to branch out into Scheme, and I learned =
only
enough OmniMark to make small changes in others' code. But I learned =
XSLT
willingly.)

So I wonder whether CTM is really worth SC34's time.

Jim Mason