W3C home > Mailing lists > Public > www-validator@w3.org > October 2000

Re: .html and nothing else

From: Terje Bless <link@tss.no>
Date: Mon, 2 Oct 2000 18:11:13 +0200
To: W3C Validator <www-validator@w3.org>
Message-ID: <20001002183035-r01010600-703f69cc@10.0.0.2>
On 02.10.00 at 10:02, Shane P. McCarron <shane@aptest.com> wrote:

>At the risk of sounding pedantic here, let me expand on something:

Pedantic is good! :-)


>XHTML 1.0 is somewhat wishy-washy on the issue of Internet Media Types.

Yes, this is one of my problems with it. It's also a "wishy-wasy"
semi-structural, semi-physical, inconsistent language that goes out of it's
way to have minimal impact on existing users and implementations. This
provides no benefits over HTML 4.0 but does have problems.

If you'd have gone with just reimplementing HTML 4.0 Strict -- using the
transition to XML as an excuse for getting rid of the leftover crud in
Transitional/Framset (and Strict too for that matter) -- I'd have bitten my
tounge. But to reimplement something with all the same problems, and few if
any benefits, in a new untried implementation language (XML); and then to
throw this MIME type confusion (which I still claim is unnecessary) on top
of it?

/me is *not* happy!


>The reason for this is that in earlier drafts of this specification it
>explicitly said text/xml or text/html.  The XML community went ape-shit (a
>technical term) because these documents, while well formed and indeed
>valid, have embedded semantic assumptions that cannot be expressed in pure
>XML.  This meant that even if an XHTML document were served up as
>text/xml, in order to process the document correctly a conforming user
>agent would have to have arcane knowledge of HTML and its semantics. The
>XML purists out there (you know who you are) objected strongly to this
>requirement.
>
>Rather than ruffle everyone's feathers, we decided to be silent on the
>issue of XML in the document itself, punting the issue to an IETF
>taskforce.

And the correct course of action, IMO, would have been to admit that we
weren't ready for XML to replace HTML just yet. There was still
specification work to do and a ghastly amount of implementation work. XHTML
could then have been released in finished form with substancial benefits
and no historical baggage to drag it down. Thus also keeping the XML
community happy.


>Now, this does NOT mean that you cannot serve up XHTML 1.0 documents as
>text/xml.

Yes, I'm afraid it does. Since there is no specified MIME type for it
except text/html, browsers aren't required to handle it. Further, there
isn't significant incentive for them to handle it because the whole world
is serving it up in "compatibility mode" text/html.


>What it means is that when you do so, you had better have an
>associated XML Schema or DTD that describes the semantics of the
>document in a way that is compatible with the browsers.  

I don't like the thinking that is implicit in the phrasing "compatible with
the browsers". I'll assume I'm misunderstanding you here.


>First, XHTML requires documents to be valid. This also means
>well-formed.  Documents that are not well formed will not even render in
>current versions of Netscape when a stylesheet is used, so requiring
>well-formedness has immediate benefit.

Point, but only barely. I suspect that this is incidental and subject to
change.


>Second, in my empirical tests well formed documents render significantly
>faster in all modern browsers.  I assume this is because they are easier
>to parse.

Point, but the same can be said of decently written HTML documents. Again,
it's incidental and not inherent to the current XHTML standard.



>Third, XHTML introduces the concept of case-sensitivity in elements

And this is a good thing?!?!?


>attributes and forces people to start lower-case now.  This helps to
>ensure that people are used to using the forms that XHTML 2.0 (and XML
>in general) requires.

Somewhat self-referential logic, if you ask me, but... :-)


>Finally, note that XHTML 1.0 documents are compatible with XML
>requirements and with XHTML 1.1 (all you would need to do is change the
>DOCTYPE declaration). This means that, in general, documents can migrate
>as the standard progresses.

If XHTML 1.0 documents can become XHTML 2.0 documents simply by changing
the DOCTYPE, I'm going to declare XHTMl 2.0 _dead_ as of right now!


>I'm sorry if this doesn't meet all of your requirements - it met a
>number of other peoples'.

Well, as I also said in off-list mail, this is obviously just me griping
here. I don't claim to speak for anyone else or to be the purveyor of the
One And Only Truth. It also occurs to me that we're off-topic and getting
further afield by the character. Sorry about that! :-|


-- 
As a cat owner, I know this for a fact...
Nothing says "I love you" like a decapitated gopher on your front porch.
Received on Monday, 2 October 2000 12:31:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:13:54 GMT