Re: some thoughts on objections to publishing ""HTML 5 differences from HTML 4"

On Fri, 29 Jun 2007, Steven Faulkner wrote:
> The rationale for dropping the headers,longdesc and summary attributes 
> was presumably formulated by the WHAT WG when the decisions to drop 
> these attributes was made.

No such decision was ever made. We started with a blank slate, and added 
features based on use cases. The use case "I'm blind and I need to be able 
to tell which headers apply to a data cell somehow" was originally handled 
by just having <th> elements, scope="", and implicit scoping with an 
algorithm, since that seemed to handle the 80% case. It has, since then, 
been suggested that it might not cover enough cases, but that thorough 
research hasn't yet been applied to the specification. None of the use 
cases considered so far resulted in the adoption of longdesc="" and 
summary="", and so far there is no thorough research transcribed to the 
wiki containing use cases suggesting that longdesc="" is useful.

(The wiki pages for summary="" and headers="" could be improved by 
decoupling the proposals for those attributes from the explanations of the 
problems that the spec doesn't cover, by the way. It's nice to see those 
wiki pages exist, though.)

Ian Hickson               U+1047E                )\._.,--....,'``.    fL       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Friday, 29 June 2007 21:03:44 UTC