On 6/26/06, Keith Moore <moore@xxxxxxxxxx> wrote:
> > If the errors are sufficiently grave, it is easy to fork the WG > documents and have them replaced or completely rewritten. it's not easy at all - because even if you replace the WG you'll have most of the same individuals active in the new one as in the old one - only they'll be angrier than the first time around,
I meant replace the documents, rather than the WG. But you're right, they do get angrier, at first. For example, in the somewhat-successful Atompub group, this happened at least twice. The protocol document still contains a lot of cruft that's only in there so that a few vendors can call whatever it is they're building "Atom". It might not be fatal stuff, so forking it won't work. The resulting document will be easy to improve on, if someone wants to, but it sure is a heck of a lot more practical than WS-*.
I'm much more interested in trying to figure out how to get WGs to stay on track in the first place and to accept useful clue from elsewhere.
I maintain that no process will accomplish that. The only way to get a WG to accept a clue is to demonstrate that their output is irrelevant by concrete example. -- Robert Sayre _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf