Re: System Level APIs draft proposal

On Thursday, 3 May 2012 at 06:33, N.V.Balaji wrote:

> I agree. I am not aware of the cost and effort involved in setting up a WG,

~2-6 months. Potentially reduced set of RF commitments.    
> however, small and focused working groups are likely to gather more mass
> than one group with wide spread focus.

Maybe, though WebApps WG and the WHATWG serve as strong counter examples. Depends if there is a legitimate interest in implementing (over, say, "wouldn't it be cool if we did spec X?").

For example, if none or few the people that jumped to the new group actually can contribute code to a browser's codebase, then whatever is standardised will see little or no uptake (though you might get a "standard"… historical example: the BONDI and JIL specs). That might be ok, as it might meet the needs of a number of members that want to work together and don't care about getting their code landed, for example, in Webkit. So, it really depends on what the goals are: if the goal is to get onto the Web platform, then at a minimum one would need endorsement from one of the dominant browser engines. But if that is not a big concern, and two or more companies just want to work together on some APIs because the W3C provides a good patent policy and lively community, then a new Working, Incubator, or Community Group is the right way to go IMO.  
  
--  
Marcos Caceres
http://datadriven.com.au

Received on Thursday, 3 May 2012 09:04:53 UTC