Re: CR exit criteria and features at risk for HTML5

Smylers, Fri, 17 Aug 2012 13:54:27 +0100:

> I definitely agree with that. Thinking about it in those terms, even 1-2
> years sounds like too long.
> 
> If ceasing to recommend HTML4.01 is the priority, then it makes sense to
> publish as soon as HTML5 would make a better recommendation than
> HTML4.01. In which case, it would make sense for the publication
> criteria be phrased in terms of that, and not mention tests and interop
> at all. Is that possible?
> 
> Indications that HTML5 is not 'better' than HTML4.01 in some areas could
> be things like:
> 
> * For feature X implementers still need to refer to HTML4.01 rather than
>   HTML5 to produce useful software.
> 
> * For feature Y authors still need to refer to HTML4.01 rather than
>   HTML5 for a useful explanation of the feature.
> 
> * Feature Z in HTML5 is a change from HTML4.01 (or wasn't in it at all)
>   but it isn't stable and there's a chance that we'll want to revert to
>   the HTML4.01 behaviour (or remove the feature).
> 
> In the absence of any cases along those lines, HTML5 would be
> unambiguously an improvement on HTML4.01, so, however bad HTML5 is, it
> would make sense to start recommending it instead of HTML4.01.
> 
> Could the working group instead work towards identifying if there are
> any areas of the spec which statuses like X, Y, or Z above, and if found
> then prioritize fixing (or eliminating) those, then declare it suitable
> for publishing as a recommendation?

+1 +1 +1
-- 
leif halvard silli

Received on Friday, 17 August 2012 13:16:44 UTC