(Please send followups to debian-devel, not debian-devel-announce)
Potato looks ready to freeze. Its primary goals have been achieved, and the only things left to do are to finish the bootdisks and fix lots of bugs. I think it is advisable to freeze now, before we start major new developments in potato. Last weekend has shown that the bug count can be reduced rapidly in intense sessions. We'll need more of those, and probably a large number of packages will also have to be removed from frozen. The freeze will be the coming weekend, on Sunday, November 7th. Before the freeze, we will have to deal with the backlog in Incoming somehow. There are more than 200 packages in it now and it's growing. Help is on the way, but probably not in time. In any case, I do not think it is wise to install a hundred new packages just before the freeze! My plan is to handle all the packages that fix bugs, and leave the rest for the new unstable. After the freeze, I expect it will take a week or two for frozen to settle down. A lot of bugs can be fixed in that time. This period will be similar to the traditional freeze. Then we can start with Test Cycles. These will address the problems we had with the previous two freezes. A Test Cycle looks like this: 1. Boot disks and CD images are created. 2. The distribution is tested for a fixed amount of time. No changes of any kind will be made to frozen during this time. Fixes for problems that are discovered will of course be prepared, but they will not be installed yet. 3. The results are evaluated. If the distribution is good enough to release, it is released as it is. 4. Otherwise, fixes are installed, and if necessary, extra time is taken to fix the problems. 5. New boot disks and CD images are created, and the cycle begins again. Richard Braakman