I am full time on Apache at the moment, split between CouchDB docs and CloudStack. Agree with everything I've seen in this thread, especially the scope of what's included in 1.3.
I have one thing to add. I think, after this release, we should adopt the release cadence and merge procedure that we came up with in Dublin, that has been proposed to the list before. (This ties into Bob's suggestion to use 1.3 as the stable base on which to start merging in all the forks.) It's not really had much discussion since then. Only problem is that I need someone to lend a hand with the Git side of things. So I'm looking for someone to join me in some release team activity to get things properly set up and communicated out to the community. (Looking at you Paul, Bob. One of you, don't need both.) So, who wants to buddy up with me to get this done, post-1.3? Don't all step forward at once. (I'm planning to buy a book on Git. I'm fed up of being so clueless...) On Thu, Oct 4, 2012 at 3:53 AM, Benoit Chesneau <[email protected]> wrote: > On Oct 3, 2012 10:48 PM, "Dave Cottlehuber" <[email protected]> wrote: > > > > On 3 October 2012 21:41, Paul Davis <[email protected]> wrote: > > > > > Only other thing I'd add is that we talked about importing Jiffy. > > > > Good, I'd be up for that. Um did you find (in cloudant land) that it > > handled parsing large docs better? The current ejson struggles > > sometimes I think. Better would mean without spitting the dummy > > completely. > > > > A+ > > Dave > We arw using jiffy in rcouch. It solves some errors too. It is also faster > on large docs, On thing I'm not sure with using a nif here is how it > behaves with dbs containing obky relatively larges docs? If it blocks the > scheduling too much or not. > > Maybe cloudant can answer here? > > benoît > -- NS
