W3C home > Mailing lists > Public > public-html@w3.org > October 2009

RE: A suggestion from the public

From: Justin James <j_james@mindspring.com>
Date: Mon, 26 Oct 2009 02:05:35 -0400
To: "'Tab Atkins Jr.'" <jackalmage@gmail.com>
Cc: <public-html@w3.org>
Message-ID: <0ad101ca5602$55a795d0$00f6c170$@com>
Tab -

The overall sentiment that I hear is that people want that style of HTML to not be merely "defined" an "obsolete" or "non-conforming", but to be considered "valid HTML". The alternative version I have been hearing, is to have a method of triggering the UA to apply an older version of the HTML spec to it (in other words, to behave the way DOCTYPE is frequently mis-understood to work). I think that the latter makes the fatal assumption that Web browsers are the only UAs out there; while it may be possible for Microsoft or Firefox (or whoever) to go through their code archives and put together a zillion engines that can be triggered somehow, I think that it would be a huge burden for anyone else without that code legacy.

But, to specify a subset of HTML as "presentational HTML" that works in the spirit of older specs... while I agree with you that using it would be shortsighted (I remember in 2000 converting a 200+ page site from HTML 3 to HTML 4 with CSS... well worth it in the long run!), I do understand why a lot of people would like to see something like that be marked as conforming and valid.

J.Ja

-----Original Message-----
From: Tab Atkins Jr. [mailto:jackalmage@gmail.com] 
Sent: Monday, October 26, 2009 12:50 AM
To: Justin James
Cc: public-html@w3.org
Subject: Re: A suggestion from the public

On Sun, Oct 25, 2009 at 11:38 PM, Justin James <j_james@mindspring.com> wrote:
> One of the readers of my column at TechRepublic sent in an item to me, that
> he asked that I raise to the group on his behalf. While I am sure that it
> will not go far in this group, I feel that it is my responsibility to pass
> it along.
>
> In a nutshell, he asked that we find a way to signal a more traditional
> style of HTML, one similar to HTML 3, in which CSS is not needed to perform
> presentational markup. His suggestion was to use a tag that would indicate
> "everything within here should be treated like older HTML", or possibly to
> use something like DOCTYPE (if not DOCTYPE itself) to indicate that a
> document be treated according to older rules.
>
> Again, this is something that I am passing along at the request of a reader,
> so take it as you will.
>
> That being said, while I *personally* understand why the
> presentation/semantic separation achieved with CSS is a major improvement, I
> have been getting a lot of feedback from readers who share frustration that
> the ability to generate a simple document using HTML is getting lost in the
> quest to transform HTML into a development platform. I agree with that
> notion, in that the bar has been raised significantly just to get a simple
> "Hello World" onto the Web since HTML 4. I know that at this stage of the
> HTML 5 game, it's a bit late, but I do feel a duty to raise it as a
> potential topic for discussion.

All of the old presentational markup is still documented in HTML5,
with instructions on how to parse and render it.  It has to be,
because a significant amount of documents on the web are coded in
those old versions of HTML and are very unlikely to be updated.  We
still want to be able to read them, so we still have to codify what
their elements mean.  We just mark them as obsolete and say that they
should not be used by authors, as they are horrible to actually use.

So if anyone wants to use those old elements, they can.  Their
documents will just be nonconforming.  And then, later, they'll be
angry at themselves for doing such a silly thing when they need to
make a site-wide change and every single <font> tag needs to be
updated.

~TJ
Received on Monday, 26 October 2009 06:06:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:50 GMT