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

Re: Proposed list of predefined class names

From: Jonas Sicking <jonas@sicking.cc>
Date: Sat, 05 May 2007 06:29:35 -0700
Message-ID: <463C86BF.9030502@sicking.cc>
To: Rene Saarsoo <nene@triin.net>
Cc: "public-html@w3.org" <public-html@w3.org>

Rene Saarsoo wrote:
> 
>> The more I think about it the less I think that the current list of 
>> predefined class names in the spec is a good idea at all.
>> [---]
>> What is special about the class names in the list?
> 
> I don't think it is a bad list. Many of them seem quite
> useful to me.
> 
> copyright:
> A screen reader user could easily skip to the copyright
> information of the article he is reading.
> 
> error and warning:
> Non-CSS-aware browsers like Lynx could display errors and
> warnings in distinctive way (e.g. in red).
> 
> search:
> A browser could implement a shortcut for jump-to-search.
> 
> example, note and issue:
> Without CSS it might be otherwise quite hard to tell, where
> a group of example paragraphs end and normal content starts.
> The same with "note" and "issue", although I'm not quite sure,
> should the "issue" really be included into that list - doesn't
> seem such a common use-case to me.

You could certainly give them semantic meaning by doing the above. But 
you could do that even if the list wasn't in the spec. The thing is that 
they are no more or less likely to have the semantic meaning that you 
think it did than if we said "here are some class names people seem to 
use a lot, treat them like they have semantic meaning based on their name".

Basically what we are doing is that we are guessing that most people 
that currently use class=error in their pages are using it to indicate 
an error message. But anyone could make that guess.

/ Jonas
Received on Saturday, 5 May 2007 13:29:38 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:44 UTC