W3C home > Mailing lists > Public > www-style@w3.org > March 2004

RE: @useragent (was Proposal: version at-rule)

From: <Matthew.van.Eerde@hbinc.com>
Date: Mon, 29 Mar 2004 13:22:57 -0800
Message-ID: <B8B9965E9A4B1948839F37ADCB58E9D5813073@exchange.hbinc.com>
To: christoph.paeper@tu-clausthal.de, www-style@w3.org

> From: Christoph Päper [mailto:christoph.paeper@tu-clausthal.de]
> If all browser vendors had implemented CSS correctly, those hacks wouldn't
be
> needed. It's not the W3C's job to clean up the mess. (Okay, there are some
> minor parts where the spec is or was too unspecific or 
> self-contradictionary.)
> 
> Furthermore in my opinion only hacks that rely on parser 
> errors are evil.

At my organization there are three tenets for web development:

1) CREATE CORRECT CODE
code should be specs-compliant.  That way any browser has at least a decent
crack at generating respectable output given the code.

2) WORK AROUND BUGS IN MAJOR BROWSERS
If a browser with more than 2.5% market share (right now this is basically
IE5 through IE6) has a bug that breaks your correct code, create a hack or a
workaround for that particular browser.  In doing so be careful not to break
the code in such a way as to violate tenet 1).

3) IGNORE BUGS IN MINOR BROWSERS
If a browser with less than 2.5% market share has a bug that breaks your
correct code, explain (politely) to the user that their unusual browser
appears to have a bug.  Politely suggest that they appeal to their browser
manufacturer - or find a specs-compliant alternative browser - or go to a
major browser.

It seems to work pretty well - at least, until a bug in a minor browser used
by a major client shows up.  But such occasions are rare and can be dealt
with as they occur.
Received on Monday, 29 March 2004 16:23:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:27 GMT