You are /partly/ correct. The intention behind this proposal is to have a spec that matches what implementations do (or that we can reasonably expect that they will do, given minor effort), so that users know what idioms will work.
It isn't just a matter of PR for implementors. If the spec says that certain idioms work, but those idioms actually fail for all real implementations, then that is bad. Moreover, given the problems with the storage mutex as currently specified, some browsers don't currently implement it at all - which is really bad. This proposal makes the storage mutex easy enough to implement that it would hopefully be more widely implemented - although in a weaker manner. -Rob On Tue, Nov 24, 2009 at 5:19 PM, Jonas Sicking <jo...@sicking.cc> wrote: > On Tue, Nov 24, 2009 at 4:46 PM, Rob Ennals <rob.enn...@gmail.com> wrote: >> In spec language, I think it should be "MAY" release. > > It seems to me that the only difference between what we have now (in > the spec), and your suggestion, would be that implementations could in > more cases claim to follow the spec. I.e. the only difference seems to > be one of PR for implementations. Nothing that would actually help web > authors. > > / Jonas >
