Mike Kupfer <mike.kupfer at sun.com> writes: >>>>>> "Mark" == Mark J Nelson <Mark.J.Nelson at sun.com> writes: > > Mark> Prioritization is NOT implied by ordering. > > It sounds like E (documentation) should be the top priority, followed > perhaps by C (Automated management of the split repositories), since > it's a major dissatisfier.
Some notes on what needs to be said would be good, the initial documentation we put together, I believe, covers everything we could think of. If people have read that and need more, it'd be handy to know what else. > Mark> C. Automated management of the split repositories > Mark> 1. In the spirit of hgext.forest, but need not be generalized. > Mark> 2. Major dissatisfier for internal developers, > > Are there specific aspects of this that people are complaining about? > That is, do they want to pretend that the open and closed trees live in > a single repository, or are there just certain rough edges that need to > be smoothed? Yes, I share the same question. Can one of you guys also ask Java how forest is working for them? Our concern was the complete lack of atomicity, if I recall correctly. But given the biggest reason that would have been a problem, the two halves of the tree being in different places, opensolaris.org and onnv.eng, seems to now sadly be gone seems reasonable to think that's less of an issue, if still an issue at all. -- Rich
