W3C home > Mailing lists > Public > public-bpwg@w3.org > March 2008

Re: Statistics on mobileOK Basic

From: Sean Owen <srowen@google.com>
Date: Wed, 5 Mar 2008 11:46:26 -0500
Message-ID: <e920a71c0803050846n2fc1f082y7f00476b7739339c@mail.gmail.com>
To: "Dominique Hazael-Massieux" <dom@w3.org>
Cc: public-bpwg <public-bpwg@w3.org>

Fascinating, fascinating. Thanks Dom. This below I think confirms what
we all suspected, that the problem is largely compliance with the
basic standards out there alone, not even higher-level issues. I'm
still glad we have mobileOK, another test suite asserting that
standards compliance, and more, is good.


On a philosophical tangent -- are the XHTML specs unnecessarily
difficult to comply with? I like the idea of well-formedness without
question, but do there need to be so many tags and rules about what
can go where? The lang vs. xml:lang item particularly raises this
point. Should that be required? I'm not taking a position, just noting
that the level of non-compliance out there, coupled with the fact that
things "basically work" in practice, raises these questions.

Then again I have heard anecdotally that the vast majority of browser
code is in place because it has to expect all kinds of variation from
the standard. A browser that only accepted valid markup would be
lighter and faster. I think Firefox already has a sort of two-track
approach to rendering that is pickier but faster on docs that appear
to want to be standards compliant, and IE8 seems to be headed the same
direction. So maybe that will finally put out some clear incentive to
developers to do it right.

Sean

On Wed, Mar 5, 2008 at 11:22 AM, Dominique Hazael-Massieux <dom@w3.org> wrote:
>   * the most common errors were
>   - not XHTML Basic valid (74% of pages)
>   - not valid (70% of pages)
Received on Wednesday, 5 March 2008 16:46:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 5 March 2008 16:46:43 GMT