RE: And the DTD says, "I'm NOT dead yet!!"

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Dan had to insist on people cleaning up a mess 
the HTML users insisted on making. 
Dan worked his buns off to make sure HTML 
was valid SGML.  No question of that.  If he 
regrets that, he is angst-ridden without cause.
Markup discipline comes with benefits but any 
rigorous practice consumes resources.

So let's be clear:  the problem was not in  
DTDs; it was in the application design.  For what 
they are designed to do, DTDs work.  They don't 
do enough.  Thus, schemas.  As application 
languages with DTD-valid ROAs, meta-schemas work. 
If the DTD is ever to be replaced for that purpose, 
someone has to explain why it is broken.

The article doesn't disturb me.  It was sent 
to me from a different list discussing schemas 
as "food for thought".  I was being asked to 
consider DTDs as unnecessary.  It prompts me to 
react because as Simon observes, we end up 
having to explain this stuff.    Henry says 
DTDs will be around for quite a while.  Ok. 
That makes it easy to explain why the DTD 
is there even if not why it is non-normative. 
That still makes no sense to me.  

Len
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Sean B. Palmer [mailto:sean@mysterylights.com]
Sent: Monday, January 08, 2001 7:26 AM
To: Bullard, Claude L (Len)
Cc: Henry S. Thompson; xml-dev@lists.xml.org
Subject: Re: And the DTD says, "I'm NOT dead yet!!" 


Len Wrote:-
> The official is Dan Connoly and the article is
> http://xhtml.waptechinfo.com/extxhtml/

Ugh, that's one of mine (nice to know that people are still reading it).
Onwards:

> Huh?

I know, I know, it's not the clearest article I've ever written, and it's
hideously out of date now anyway. What I was trying to say is that XML
Schemas have a much wider function than DTDs: they do different things. The
fact that they both are systems for constraining data models of XML is a
bit confusing to some people (including myself when I wrote the article,
evidently), but the simple truth is that XML Schemas are more expressive
than DTDs. Doesn't mean they will "eradicate" them..... but imagine if the
initial Web had had XML Schemas and namespaces etc. in the first place? I
think that the Web might be a better place for it, with more people
validating their code etc. Don't ask me to back that up with any facts; no
one can change the past, and any speculations about it are just that -
speculation...but that is my opinion (subject to change every 5 minutes).
However, remember how much trouble Dan (Connolly) had in getting people to
validate their code, and make it SGML compliant? I expect that he regrets
doing that now, and is trying to make "amends" by stabilizing things
somewhat and renouncing the SGML past. But that is just speculation once
again on my part... Why don't you ask him?
So I think he is right, but only in the context of his 10 years of Web
experience... here from the very beginning. Like all of us, he has seen the
corruption of the Web through non-validation crappy HTML coding... I wonder
what's next?

Kindest Regards,
Sean B. Palmer
http://infomesh.net/sbp/
http://www.w3.org/WAI/ [ERT/GL/PF]
"Perhaps, but let's not get bogged down in semantics."
   - Homer J. Simpson, BABF07.