W3C home > Mailing lists > Public > www-html@w3.org > January 2002

solutions)ntations of CSS for years to come (was RE: CSS multicolumn layout solutions)

From: Etan Wexler <ewexler@stickdog.com>
Date: Thu, 24 Jan 2002 00:11:25 -0500
To: Web style list <www-style@w3.org>
Cc: <www-html@w3.org>, <www-style@w3.org>
Message-Id: <20020124051112.8A3532AD79@server5.safepages.com>
Christian Hujer wrote:

> So why should I already use CSS3 features that only work in 
[the dominant
> browser], if even HTML4 and CSS2, Recommendations from 
*1998*, do not work
> in [the latest version of the dominant browser, circa 2001]?

HTML 4 published as Recommendation prior to 1998 (1997-12-18). 
 But, hey, what's a difference of a few months at this point?

To answer the question, though, one should use CSS3 features, 
like the 'inline-block' value of 'display', because the features 
are useful.  What's more, CSS is constructed with attention 
to graceful degradation.  Combining declarations for 
CSS3-capable user agents with declarations for downlevel 
user agents, one can satisfy everybody.

Nevertheless, it is irritating that certain resource-rich 
companies can't muster the development power to implement 
four-year-old specifications.  (Or five-year-old specifications; has 
*anybody* produced a CSS1-conformant user agent?)

> It is always the same with [this monopolistic software giant]:
> New features everywhere, but no bug fixing,
> no or no proper implementation of standards, no security and 
no stability.

Get used to it.  Robust software engineering often fails to deliver 
the payoff for the vendor.  We will continue to endure software 
of beta quality, alpha quality, or worse.

CSS, in particular, is an area that vendors will be slow to 
implement conformantly.  Ambiguities and contradictions within 
the specifications, lack of a paying market, and difficult 
features make CSS unattractive to implementors.

It is not only commercial vendors who will produce 
non-conformant implementations, though.  There is 
significant pressure from dee-zyner folks to implement 
certain features, even if those features or those combinations 
of features do not correspond to a published specification.  
Ian Hickson wrote about Mozilla (<http://bugzilla.mozilla.or
g/show_bug.cgi?id=5693#c40> 2000-08-28):

    This is one of those cases where pulling the feature altogether
     is the alternative to fixing it, but pulling :hover and :active
     is likely to get us all shot. 

It's not a rosy picture, folks, so grab your bottle of gin (or 
whatever gets you through the night) and prepare yourselves.

Etan Wexler
Received on Thursday, 24 January 2002 00:11:17 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:58 UTC