implementation/specification

Tom,
(+www-archive -public-tracking)

This is off-topic of ISSUE-81

Le 27 oct. 2011 à 19:51, Tom Lowenthal a écrit :
> Karl, I quite disagree. If we can imagine good applications in the
> future, we should try to enable them rather than prohibiting them, even
> if they don't actually happen. Generativity is an important goal.


We can imagine a lot of things. Really. And it is interesting to explore 
and test things for the future of technologies and interactions. There is 
even an avenue for this at W3C which is called Member Submission. [1] The 
Community Group has also been created for enabling exploration of techno-
logies with a low cost in participation [2].

For the purpose of W3C Specifications, the specifications are written to 
be implemented in an interoperable way. It's a delicate, ingrate exercise.
Whatever good the idea is, if it's not implemented it doesn't really exist.
The CR phase of W3C is asking for a double implementation of each features.
If a Working Group can't demonstrate that there is a double interoperable 
implementation, the feature is likely to be dropped. So the goal of a WG
is to come with good ideas (indeed) but ideas that will be implemented by 
the end of the specification. 

It is perfectly fine to have more ideas for future versions and to document 
them. This can happen on a wiki, a community group, etc. But for the purpose 
of having something realistically used, we also have to reach a consensus 
on what participants are willing to implement.

Hope it makes it clearer.

[1]: http://www.w3.org/Submission/
[2]: http://www.w3.org/community/


-- 
Karl Dubost - http://dev.opera.com/
Developer Relations & Tools, Opera Software

Received on Friday, 28 October 2011 12:14:56 UTC