W3C home > Mailing lists > Public > public-html@w3.org > November 2007

Re: The only name for the xml serialisation of html5

From: Mark Birbeck <mark.birbeck@formsPlayer.com>
Date: Thu, 1 Nov 2007 00:56:36 +0000
Message-ID: <a707f8300710311756r30406fbaq805c04c6615a2b5b@mail.gmail.com>
To: "Dean Edridge" <dean@55.co.nz>
Cc: public-html@w3.org

Hi Dean,

> And in that case, I hereby formally request that the editors include
> "XHTML5" into the title of the specifications.
> I do so in the firm belief that this is in the best interest of XHTML.
> And will encourage the adoption of the language by more people.

You really need to make your mind up about what your argument is.

Most of the time you say that no-one is using XHTML, and that it has
been a dismal failure. Then you claim somewhat bizarrely that the
WHATWG has done more for XHTML than anyone else in recent years. And
more, that despite the fact that no-one is using XHTML, the reason
that the WHATWG is so well qualified to further XHTML's evolution is
because they have real world experience of using XHTML.

(These are all things you have said in this thread.)

Unfortunately, despite this sterling work, Ian reckons that hardly
anyone will ever use XHTML in the future anyway.

So to recap; XHTML has been a failure--except of course for the
successful bits provided by the WHATWG--and it will remain a failure
into the future.

Yet in order to "encourage the adoption of the language by more
people", you want to use the word "XHTML" in the title of the HTML 5
specification. Given that XHTML is such a miserable failure, I would
have thought you wouldn't want to go near the word.

Something doesn't add up here...in your comments...in Ian's; it's
almost as if the aim is confusion.


LANGUAGES, SERIALISATIONS, AND MIME TYPES

But as it happens, the biggest confusions are on this list, in
particular around the difference between a language, a serialisation
of that language, and MIME types that are used as part of data
transfer and don't necessarily have to impact that language.

For example, if you have a language called BananaML, why would you
want to give it an additional name that depends on how it is
serialised? That's barmy. Are you going to also produce two sets of
documentation? Two definitions of the <a> element, for example? Of
course not; the meaning of the language is primary, and the
serialisation is secondary. So why not just come up with a name for
the language currently called HTML5, and stick with it?

But in the same vein, the argument that no-one uses XHTML 1.1 because
no-one uses the XHTML MIME type is ridiculous. The number of web-pages
generated by server-side systems that use XSLT is enormous. The fact
that those pages are almost certainly delivered to browsers using the
"text/html" MIME type, so as to ensure that they can be rendered tells
us nothing about the use of XML in generating mark-up.

And as for the major contributions to XHTML coming from the WHATWG,
that is a very odd thing to say; I have complete respect for the work
of some of the people involved in the WHATWG, but I have to say that
the major progress in XHTML in recent years has not come from those
people. The major developments have come in the form of:

 * RDFa
 * the role attribute, which is finding its way into HTML5, not the
other way round;
 * access key;
 * and the modularisation technique itself, which allows parts of
XHTML to be used
   in other languages.

The latter is being used to create a mobile profile of XHTML 1.1.

By the way, none of my comments should be taken as an argument for or
against any particular naming technique, but merely to point out the
shaky ground on which many of the assertions in this discussion are
standing.

Regards,

Mark

-- 
  Mark Birbeck, formsPlayer

  mark.birbeck@formsPlayer.com | +44 (0) 20 7689 9232
  http://www.formsPlayer.com | http://internet-apps.blogspot.com

  standards. innovation.
Received on Thursday, 1 November 2007 00:56:47 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:50 UTC