W3C home > Mailing lists > Public > www-html@w3.org > August 2000

Re: XHTML Invalidity / WML2 / New XHTML 1.1 Attribute

From: Sean Palmer <sean_b_palmer@yahoo.com>
Date: Sun, 13 Aug 2000 08:05:40 -0700 (PDT)
Message-ID: <20000813150540.14858.qmail@web3505.mail.yahoo.com>
To: connolly@w3.org
Cc: www-html@w3.org, robin@isogen.com, bertilow@hem.passagen.se
Dan Connolly:
> I'm asking for one lousy attribute, 
> xmlns="<http://www.w3.org/1999/xhtml>"
> at the top (plus consistent lower-case 
> spelling of tag names), and some 
> faith that XML Schemas are going 
> to become mature for validation 
> purposes. 

SBP:
This kind of summed up your three page disagreement
that these new technologies are getting over
complicated. I agree that Schemas are better to use
than DTD's, which tend to be messy. I just don't think
that your collegues (the HTML gods") will buy into it
that quickly.

I quote:-
http://www.w3.org/TR/xhtml-modularization/xhtml-modularization-20000105.html

> This appendix will contain implementations 
> of the modules defined in XHTML Abstract 
> Modules via XML Schema [XMLSCHEMA] 
> when the XML Schema becomes a W3C 
> approved recommendation. 

I thought it was approved? Still, if not, it looks
like either Schemas or XHTML 1.1 are going to slow the
other down. One cannot finish until the other (XHTML
until Schemas). I'm not saying you should rush such a
masterpiece though, Mr. Connolly...

Having a close look at the XHTML 1.1 DTD (what there
is of it!), I found a startling fact - Schemas STILL
aren't being implemented properly (please check this,
I am nearly always wrong!). It only specifies
namespace attributes for HTML and XLINK. It doesn't
have any wildcard type namespace attribute.
I think this is due, in part, to some belief that
DTD's and Schemas are entirely separate. Of course,
you and I know that to be false, but what about the
HTML gods?It seems they are going to implement it in a
different way: XHTML Family Document Type Conformance.
In other words, In other words, XHTML 1.1 documents
with Schemas are going to be invalid. However, XHTML
Family documents (owr own Modularisation
customisations) are going to be valid.
In short, that sucks (p'on my language), and I
seriously think you should talk to the people involved
if this IS the case. If not, just ignore me.Of course,
you may not WANT Schemas included into the actual
XHTML 1.1 DTD. We seem to be moving away from DTD's in
this discussion, which I think is a good idea. I would
rather use namespaces than DTD's.

In essence, what you did by combining the "new comment
thingy" (lets call it The UTIL Comment Attribute,
sounds a bit better?), is very close to what they
expect you to do in modularisation, except that you
didn't write a complx DTD using XHTML 1.1 modules that
don't even exist yet. (I.e. I think it's impossible at
the moment!). That means if we want to use namespaces,
we have to make our own DTD's still, except they think
they have made it easier by using this new
Modularisation system, which nobody in the world
understands!

Which brings me to:-
> I hope somebody will step in and 
> show us the corresponding XHTML 
> DTD module, but I suspect it will 
> expose quite a bit of gobbledygook 
> in the module itself. 
> > I think somebody should write a laymens
> > guide to the modularisation of XHTML, 
> > just what it 
> > involves, how to do it, and, most 
> > importantly, why 
> > bother. 
> Yup... that's an important task. 
> > Maybe I'll have to do it myself, 
> I encourage you to give it a try! 
> > but I'd 
> > rather see it issued as a note 
> > at W3C.
> It's entirely possible that if you write up such a
> thing, W3C will publish it as a note. 

O.K. - I'll do it! I'll have it on your desk in the
morning, sir!

Seriously, I am going to spend a considerable portion
of my spare time wrting it, so I would be very very
grateful if you could have a look at it when it is
done, and (if good enough), refer it to who ever
publishes notes.It seems that every time I read the
modularisation Spec. it makes a bit more sense, and I
think by now I could write an XHTML Family DTD of my
own. The problem is that there aren't enough hard copy
modules, or tools to validate it with yet.This
discussion seems to be on the very forefront of Wb
technology, discussing using specifications that
haven't even been issued yet to their full extent. Any
document that mentions XHTML 1.1, XHTML Families, XML,
and Schemas is going to be pretty advanced (I'll look
back on this note in a few years and laugh).And
although I am one of the least qualified to take on
such a task, at least I do have a bit of a knack for
explaining things nicely and quickly (I have been very
highly praised for the WapDesign ORG U.K. site).

Anyway, back to the body of yor message:-

> Consider the beginner, somewhat like yourself

A beginner compared to yourself (Mr. "I wrote the HTML
2.0 spec with Time Berners-Lee, along with the XML
homepage, and goodness knows what else.). You even
describe Robin Cover as a beginner! We are in the top
1% of programmers, I would say (immodestly, altough
that still puts tens of thousands of programmers above
us), but you are most likely in the top 10 in the
world, so you would see us as beginners.
I agree I have a LOT to learn. Still, I can but try!

> The software for validating the 
> results is less mature, but I think
> it has a brighter future. 

I expect the main validator to have full compatability
for Schemas soon after XHTML 1.1 becomes a
recommendation (I predict Jul 2138).

> So I think XHTML, XML, 
> namespaces, and schemas 
> are a good mix... they make 
> the easy things easy and the 
> hard things possible.

They sure do.

> I'm suggesting that folks 
> who *design* new 
> attributes write schemas, 
> not folks that just use the 
> new attributes. 

Point Taken.

> Not at all. I think that 
> schema-valid XHTML is 
> quite neat and clean. 

I agree - what I really meant was that it is either
Valid XHTML, or it uses Schemas.

> Many of our specs have a 
> lot of gobbledygook in them. 
> I'm struggling to understand the 
> XHTML modularization spec, 
> and some of the folks in the 
> HTML WG are struggling to 
> understand the XML Schema spec. 
Hmmmmm.....that's quite interesting. I understand why,
though. Maybe that's why the two aren't coming to
gether as well as they should do (see earlier in the
text for an explanation).

> restrict yourself to attributes

That brings me to an important point: do you think the
notion of the util:comment as a means of including
META information directly into elements is a good
idea? I agree that RDF should be used for longer META
information, but for quotes such as "Was blue" and
"Generated from XML source", wouldn't it be better to
use util:comment?
This to me is part of the cornerstone of the
discussion. It is alright having means of extending
XHTML, but what is the point if it is perfect anyway?
Of course, it ISN't perfect, and that's probably the
answer.
I'll put more into my Note (see point 3) of my summary
to this reply, near the bottom).

> I'm not a fan of regulation, 
> myself. I hope you don't 
> think that just because a 
> document's address starts 
> with http://www.w3.org/ 
> it's somehow magically good. 
It's only magically good if it starts with
http://www.w3.org/TR/ :-)

I have a lot of respect for the W3C's work. Without
them, the Web wouldn't be as good as it is now. In
fact, I'm not even sure if it would be in existence.
W3C do a lot of good regulatory work, I just think you
guys are too modest too "over-regulate". There is no
such thing as over-regulation, as long as the
regulations are technically sound.

> The whole point of XHTML, and 
> my comment example, is that XHTML *is* XML

It IS XML, but it doesn't include schemas. In short,
aren't schemas a part of valid XML? I-.e. - shouldn't
they be allowed into any document, regardless of it's
DTD? I am most likely wrong on this point. I agree
that XHTML is XML, but it isn't really extensible. It
should just be called HTML in XML. I don't think that
it is HTML though, because HTML is SGML, not XML.
XHTML is XML, not SGML.

Can we copy your XSL file at
http://www.w3.org/XML/2000/04schema-hacking/comment-extract.xsl?
Not that I would need to, I'll just reference it from
scripts: but I need your permission.

> Huh? I don't follow at all. XHTML 
> cerainly is XML. Which part of our 
> conversation suggests it's not? 

I meant it doesn't allow Schemas - and this means it
isn't STRICTLY XML. You cant specify other Namespaces
with out the Validator messing up.

> > - 
> > this all supports the text/xhtml MIME 
> > type suggestion. 
> No, I don't see how it does.

Ah! - but it does! If we had a new MIME type, you
could say that every tex/xhtml document has a valid
DTD for xhtml, and therefore we would only have to use
namespaces!
Your example:-
<html xmlns="http://www.w3.org/1999/xhtml">
 <head> 
<title>Virtual Library</title>
 </head> 
<body> 
<p>Moved to <a
href="<http://vlib.org/>">vlib.org</a>.</p> 
</body>
</html>
Would then be legal in the forseeable future. At the
moment, no way will that validate whilst the HTML gods
have their say.

> > I am at least going to 
> > try to make Schemas and RDF 
> > more accessible to 
> > programmers. 
> I salute you! 

Cheers. I will do my best. Of course, I don't have
much sway, but I can write a lot, and then publish it
on my site.
I don't know muchabout RDF, but I'm getting into it.
I'm more interested in Schemas now.

> and given that modularization 
> and schemas are still not 100% 
> cooked, I think your logo "accurately 
> reflects the STATUS associated 
> with the W3C products." 

Thank you - nice loophole.
I don't expect anyone to use it, it was just a joke!
Actually, I asked Gerald Oskoboiny if he could make
and allow an XML logo for pages such as the
comment-test.html one that you prepared for us.
Maybe we should also make a "This page uses Schemas"
logo, and point it to your Schemas page.
All just thoughts...
Thank you for your examples, work, research, and
continuing support on this topic!
How shall I wrap this reply up this time?

How about with a summary:-
1) XHTML won't include Schemas. XHTML Families will.
2) Why not???????
3) I'll write a simple guide to "Extending XHTML -
Using Modularisation and Schemas to Good use in
Current and Future XHTML/XML Documents". I would be
vehemently grateful if you could review it and pass it
on.
4) XHTML is XML. But it isn't perfect.
5) I think we should have a "text/xhtml" MIME type to
cover these errors up(?)
6) Is the util:commet attribute a good idea for META
information?

Kindest Regards,
Sean B. Palmer
WAP Tech Info - http://www.waptechinfo.com/
Take a sneak peek at
http://www.waptechinfo.com/index.asp!!!


__________________________________________________
Do You Yahoo!?
Yahoo! Mail  Free email you can access from anywhere!
http://mail.yahoo.com/
Received on Sunday, 13 August 2000 11:06:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:43 GMT