Hi Yves, On 09/01/2015 11:30 AM, Yves Lafon wrote: > On 31 Aug 2015, at 20:12, Ms2ger <ms2...@gmail.com> wrote: >> On 08/31/2015 03:28 PM, Yves Lafon wrote: >>> In fact, I would prefer to have the editors’ copy published as >>> TR/WebIDL/, and let -1 -2 … -n be pointers to the stable version >>> (aka, what is implemented, not what has to be implemented). >>> >> >> Who do you propose will construct such a "what is implemented" >> specification, and what useful work would you have them drop for >> it? >> > > Well, we need tests, of course. There is a test suite that needs to > be checked again, and lots of other specs have tests that actually > test parts of WebIDL. So apart from new functionalities that don’t > have test yet (and may by mean of other specs testing those), most of > the work is to gather existing tests, and I am ready to help on that. > So I don’t ask anyone to stop doing their usual job, especially as it > can also help here. Of course people reviewing tests is always a nice > feature, as you know. >
I'm not sure I understand your point. Tests will need to be written and reviewed, whether "what is implemented" specifications are published or not. My question is specifically about the editing of the documents you propose to publish as -1 -2 … -n. Thanks Ms2ger
