Okay. This has gone too far. Both sides made mistakes. How about we all admit that? Mitch and Sven should have talked to Daniel offlist first. Perhaps requiring patches posted to gimp-dev is going too far. Daniel: GIMP isn't ready for optimizations. You should have talked to Kelly before changing the tile stuff although she hasn't commented in a while. So you made some mistakes. But some of the stuff you are doing might be more useful in the future when the code has stabilized. So how about you wait for those changes? Now.. to keep these things from happening as much in the future, I propose the following: 1) EVERYONE (including Mitch and Sven) provide frequent updates to gimp-dev mailing list regarding work in progress. It doens't have to be much, just a quick overview. 2) When an important discussion happens on IRC, those involved try to post a summary or excerpts to the mailing list for those who were not present at the time. 3) When someone's changes interfere with someone else's or the overall plan or just generally cause problems, it is initially brought up to that person in private. Most issues will need to be discussed on list, but it would be better if the person was informed privately first. 4) It be made publicly clear now who is the official maintainer of the CVS HEAD branch. It is important that the project have someone who can make final decisions when discussion does not lead to a solution. It would be best if everyone involved knew who that was. 5) When planning a new major project, discuss it on list first. This will eliminate major reversion of patches. 6) The todo list be expanded to include a listing of who is working on the items in progress so work isn't needlessly duplicated. There have obviously been lots of failures in communication lately. I think everyone here should be able to see that. I also think pretty much everyone is willing to try to improve it. Can't we all bask in the gimp love? Seriously. We need to make improvements and move on. Silly battles and ego clashes don't help anyone. bex
