W3C home > Mailing lists > Public > w3c-sgml-wg@w3.org > June 1997

Re: requiring DTDs [was: RE Deleta est]

From: Peter Murray-Rust <Peter@ursus.demon.co.uk>
Date: Fri, 13 Jun 1997 08:22:05 GMT
Message-Id: <7950@ursus.demon.co.uk>
To: w3c-sgml-wg@w3.org
In message <9706130044.AA10557@sqrex.sq.com> lee@sq.com writes:
> If you do this, you've lost our good (if grubby) Dirty Perl Hacker.

Yes - I'm very conscious of this since it represents the borderline in
my own case (native JUMBO does not read and DTDs - it can, however use
NXP and other parsers).  The choice seems to be:
	- DTD parsing is too complex for some of the intended implementing
	- different parsers will produce different output with little or
		no understanding by a large authoring/reading community
	- some brilliant solution suggested by the WG/WRB

I favour 3; and can (possibly) accept a set of rules [If your DTD contains
X then it must be parsed, but not if it contains Y...]  Whatever the solution
it must not allow different ouptuts that are opaque to the redaer and change
the intent of the author.

> But now with every XML processor required to read an internal DTD subset,
> perhaps we've lost the dirty perl hacker anyway.  If every XML processor

I hope not, but I'd have to reluctantly agree if that was the price for correct
document processing.

> I'd like to see the programming team working on a relational database
> say
>     ah yes, importing and exporting tables, let's allocate a programmer
>     to XML suport for two weeks, that'll be good for marketing and
>     we'll be able to interoperate with zzz and www and...
> and not
>     ah yes, XML, that's a complicated macro processing language,
>     we don't need that, we'll just invent some extra tags and
>     call it HTML/DB.

Although the WG/ERB cannot require it, there is always the third:

	of course we don't have to worry about writing a parser because there
are several top-class ones already written that have no license restrictions
to our use, that track the standard (so we don't have to rewrite our parser
with each spec), and have the confidence of our customers.


Peter Murray-Rust, domestic net connection
Virtual School of Molecular Sciences
Received on Friday, 13 June 1997 04:16:33 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:25:10 UTC