> Depending on the goals. > > If we take the minimal goals, of keeping the current > GNU Arch 1.x going, without aiming at solving some of > the bigger problems like the Windows port and > performance for Linux kernel-size projects, tla 1.x > should be able to continue with much less work. Of > course it is not ideal to many people, but the current > implementation is usable for many projects.
You're looking at four problems you'll need to solve: * The storage mechanism is not right. This raises problems in hydra style form in several areas - history growth is the most biggest. * The user interface needs a full redesign. * The code base could really use a solid refactoring. * The namespace decisions should arguably be removed. * In my opinion, the merge concepts in Bazaar-NG should be looked at closely. Its a bit better than the mege in tla. Btw, the first four of these are addressed in the revc spec. > Tom, do you see that as something requiring much less > efforts? > > And in the mean time, we hope someone will pick up the > 2.0 codebase and continue with it. Maybe Tom will go > back to it. That would be neat and much more promising than maintaining tla. _______________________________________________ Gnu-arch-users mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnu-arch-users GNU arch home page: http://savannah.gnu.org/projects/gnu-arch/
