Re: Charter for Native Apps

On Monday, November 28, 2011 at 1:41 PM, Charles McCathieNevile wrote:

> On Fri, 25 Nov 2011 02:44:22 +0100, Marcos Caceres <w3c@marcosc.com (mailto:w3c@marcosc.com)> wrote:
>  
> > (apologies for posting this twice, but want to make sure Art and Charles  
> > are CC'd as they are Web App WG's Chairs and their input is critical)
> >  
> > On Thursday, 24 November 2011 at 20:55, Marcos Caceres wrote:
> >  
> > > Putting it up now.
> >  
> > http://www.w3.org/community/native-web-apps/charter/
>  
> As far as I understand the CG cannot *take over* errata, nor can it bin  
> the Web Apps group, who are bound by their charter that is made by W3C and  
> formally agreed by members.

Hmm … the intention was to convey a sense of close relationship between the two groups, not that we are binning anything.  
> I suggest that you phrase the relevant bits as "will provide input on  
> errata, ..." - it enables the same stuff.

That seems better, yes.   
> I'm not happy about the idea of a specification not being allowed to have  
> a version number - in particular this interferes seriously with the notion  
> of "done" and what you do next.

We would use different criteria for that.   

[[

A specification is “done” when:
* All issues have been addressed, as recorded by the issue tracker.
* Every conformance requirement has a tests, and that test has been independently verified.
* At least two implementations pass every test in the test suite.
* A W3C Working Group takes ownership of it to progress it along the Recommendation track.



]]

I think we should be allowed to experiment with a new model (been burnt too hard by the old model).   
> Even worse is not being able to put a date  
> there - it's actually important to be able to review changes through time,  
> which means you need some notion of how to address things, and this seems  
> to prohibit it, which strikes me as unnecessary bureaucracy.

If we only have one version that lives in a version control system then that should not be a problem. Every check in is versioned.  
We can make it clean in the document as to which version in the version control system we are referring to and ask commenters to do the same.  
> I'd like to see the charter say that drafts *must* be kept on W3C systems  
> - someone telling me I have to log into some third-party service (which  
> might well be a competitor's proprietary system that I would choose not to  
> use) seems to me unacceptable for a community group.

I agree. We can use the W3C's HG server.

Received on Monday, 28 November 2011 13:03:34 UTC