W3C home > Mailing lists > Public > www-html@w3.org > May 2007

Re: Cleaning House

From: Jim Jewett <jimjjewett@gmail.com>
Date: Sun, 6 May 2007 20:05:40 -0400
Message-ID: <fb6fbf560705061705l67e2b7e3nbcb608c4f01bfb52@mail.gmail.com>
To: tina@greytower.net
Cc: "Maciej Stachowiak" <mjs@apple.com>, "Anne van Kesteren" <annevk@opera.com>, "Philip & Le Khanh" <Philip-and-LeKhanh@royal-tunbridge-wells.org>, www-html@w3.org, public-html@w3.org

On 5/6/07, Tina Holmboe <tina@greytower.net> wrote:
> On  6 May, Jim Jewett wrote:

> > If the parsing fallbacks were dropped, and b were defined strictly as
> > a fallback for strong -- it would continue to be prescriptively wrong,
> > but it would still produce something that many people would see as a
> > synonym for header, just as they do in printed pages.

>   How do you read it out loud?

You clipped the relevant portion.

There isn't any known-correct aural version of a pixel-perfect
absolute position.  A heroic User Agent might recognize it as a header
(or branding material, or advertisement), but if it doesn't ... well,
that's what you get with crummy markup.

And that is the *existing* situation today.

Saying that browsers should treat <b> as a poor synonym for <strong>
doesn't fix that, but neither does it make things any worse than they
already are.

>   Doing it wrong in a different way doesn't mean that either way of
>   doing it wrong is a good idea.


True -- so why does it matter which wrong way is chosen?  Sometimes it
doesn't.  In this particular case, one treating <b> as <strong> seems
(statistically, unless your heroric browser happens to have more
specific information) does seem to be the least bad.  We can pretend
this fallback is a trade secret, or we can document it.

>   Please - and it goes for all of us - try to keep in mind that
>   "graphical rendering of a page" is not the only physical reality in
>   which a webpage and its content is supposed to exist.

When discussing <b>, semantically well-written pages are a tiny
minority.  If you *know* your input is likely to be corrupt, then it
makes sense to consider what sort of errors are most likely to have
occurred.  Authors assuming that everyone will have an environment
equivalent to their own is by far the most common source of errors
that give *me* practical trouble.

-jJ
Received on Monday, 7 May 2007 00:05:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:10 GMT