Re: Minutes from 16 November 2000 WCAG WG telecon

> > It is a violation of SGML to specify markup for presentation: end of
> > story.
> Not really. I've been making web pages since HTML 1.0, what is
> SGML and why is what they say significant? If I don't use SGML,
> but use HTML, why should a violation of SGML affect my work?

I'm going to give up on this soon. SGML is the protolanguage for all other
markup languages. HTML is an SGML language. XML is an SGML meta-language.
MathML is an SGML language. Anything that involves markup is an SGML
language. HTML and XHTML use an SGML/XML Document Type Definition, therefore
it is an SGML language Q.E.D. If you use HTML, you are using SGML. You wrote
"If I don't use SGML, but use HTML": that is a non-sensical statement. It's
like saying, "if I don't use public transprtation, but use a public
train...". (X)HTML = XML = SGML.
Further to that, it *is* a violation of SGML to specify markup for
presentation. There is no ambiguity about that phrase.

     "In the general case, behavior is not permanently (and
     exclusively) bound to data objects (i.e., the SGML
     content vs. style model). [...] implementation[s] of the
     suggested behavior cannot be universally enforced)."
       - http://www.w3.org/XML/9712-reports.html

i.e. don't specify that elements have a behaviour because they clearly
don't. That is not an opinion, it is the ground rule of the ISO standard on
which all markup languages are based. If you want to create your own markup
standard, and create your own presentation based format, be my guest, but
for our sake don't bring HTML and the WWW down with you. Sorry if I'm being
spurious (I certainly don't mean to offend), but I'm sticking up for the
entire foundation principles of the WWW and HTML. [BTW, technically there is
no such thing as "HTML 1.0". It stated as an IETF RFC at HTML 2.0. Before
that there was just Tim's draft...]
I'll bet you still doubt me. Well how about this then: HTML and the WWW were
created by a man named Tim Berners-Lee. He is the director of the W3C itself
now. When he originally wrote HTML, he wrote this:-

     "It is required that HTML be a common language between
     all platforms. This implies no device-specific markup, or
     anything which requires control over fonts or colors, for
     example. This is in keeping with the SGML ideal." [1]

Surely you must get it now :-)
That quote was written by the inventor of HTML, when he invented HTML. All
of this <b>, <i>, rubbish was added at a later date. My guess is that the
HTML WG will deprecate it soon anyway (they probably acknowledge that they
will, otherwise they wouldn't have given us m12n). Presentational markup is
*wrong wrong wrong*, unless you have some other semantic purpose in mind for
it. This is not an opinion, it is the truth, rigid solid fact written into
years of standards and Web architecture. What would happen if you tried to
add a behaviural style(s) such as presentation to an XML document in IE5? If
I used <bold>, IE5 wouldn't render it as bold. If I used CSS it would. What
does that tell you? If you are going to say it would display <b> as bold,
save your breath: only Web browsers trying to bring in more customers
actually attach behaviours to markup. It is my guess that in the future they
won't. Note that Netscape and MS are both W3C members...
This all goes against the principle of the Semantic Web. What if I wrote a
new SGML/XML language and specified a <b> element in it. What does the b
stand for? It stands for baker, it's a list of bakers in my local area. How
would I get a processor to realise what is is for? (I'd use RDF and schemas,
FYI). More to the point, why on earth would I want to specify a behaviour
for it. Exactly, I wouldn't.

> Maybe it's well to start with what specifically constitutes mark-up. Use
of
> Bold, Italic and Underline, plus "different" color and "different" fonts,
> all used as visusal aids.

How is a visual/presentational aid semantic? It is the exact dictionary
opposite of semantic!

> Why can't the alternate outputs just read these
> variables in HTML and act like the style sheet does?

Good question. The answer is because that is a poor use of markup, getting
proprietary software to grok your proprietary behavours attached to
proprietary markup. If the WWW is going to be based on that, then I'll just
sit back in my easy chair and laugh at it whilst it collapses. Behavioural
and presentational markup does NOT constitute sound Web architecture, no
matter what people have been deluded into thinking!

> The thoughts of purists aren't my concern, accessibility is. When
> accessibility is used to advance "purists" principles, it's being misused.

I can't see how accessibility can be on your agenda at all when you ridicule
the soundest possible WAI principles and guidelines there are. I am a
purist, but the WAI covers a very broad spectrum of
people/beliefs/influences. However, I am totally behind what the WAI has
achieved.

> This seems to presume that web pages are created by teams,

No, use of markup for structure and semantics presumes that web pages are
built by either compenent Web designers, or competent Web tools.

> Everyone has their own working style. Some pages are best if the visual
> effect is planned first, then add the text to round it out.

So write the style first. I often write independant styles. If you're not
able to create styles independant of markup, I suggest that is because you
have spent too long following anachronous, limited, and blatantly illegal
Web programming habits, and I suggest you change pretty soon before the WWW
changes first.

> >Does this make any more sense now?
> >> >      Guideline 2. Separate content and structure from
> >> >      presentation and explicitly define significant structural
> >> >      or semantic distinctions in markup or in a data model.
> >> >    - http://www.w3.org/WAI/GL/WCAG20/
> Perhaps it would say it better if it said that to simply define
significant
> structural and comprehension aides in markup. I've no clue what "data
> model" adds to the guideline.

I fully agree. In my opinion it should be re-written to:-

     "Guideline 2. Create a given document's content based
     on structure and semantics, and add the presentation as
     a separate entity based on the afforementioned content."

I shall hope to raise that with the rest of the group before our next
teleconference (I don't think you're a member of the GL are you? If you
would, you could raise your point yourself). Although it doesn't mention
markup, the definition of content will include markup, I hope :-)

Kindest Regards,
Sean B. Palmer
http://xhtml.waptechinfo.com/swr/
http://www.w3.org/WAI/ER/
http://www.w3.org/WAI/GL/
"Perhaps, but let's not get bogged down in semantics."
   - Homer J. Simpson, BABF07.

[1]
http://www.w3.org/History/19921103-hypertext/hypertext/WWW/MarkUp/HTMLConstr
aints.html

Received on Sunday, 19 November 2000 12:19:53 UTC