>> I _would_ like to do off-site backups of the CVS repository, >> but assuming everything goes through, I'll work that off-line >> with Doug. > >Definately... It would be a shame to loose all the history. One >nice way to accomplish this would be to allow cvsup access, then >anyone can have a full mirror of the cvs repository.
Well, I guess the only problem with that is the classic cvsup problem of the Modula-3 runtime. We'll need to figure that one out. I'm not sure it's the best mechanism, but it's certainly one possibility. >> At this point, we create a release branch. > >I don't see a need for a branch at this point. All the work will >be moving along the same path and thus can be done on HEAD. Well, I don't want to lose any momentum from people that want to contribute new work; this lets us start release engineering, but people can start committing new stuff right away. >I am not sure the changes thus far warrant bumping the major >version number, but in the scheme of things I don't think it is >really worth arguing over so whatever everyone else thinks. >Aiming for the end of Janurary sounds reasonable enough to get >all the patches tested and commited. Remember that the "end of January" release isn't for any NEW code ... that release will (under my plans) be whatever is currently in the CVS repository plus any fixes to that needed to create a stable release. Part of my motivation for a "2.0" release is to draw attention back to nmh. "Hey, look, they haven't been sitting on their asses for two years, they have a whole new major release!". I don't think shameless self-promotion is necessarily bad in this case. Also, I think that there ARE a number of significant changes in the code (e.g, all of the SASL security stuff I added way back when) that it's a legitimate argument that perhaps a 2.0 release is warranted. A diff against the 1.0.4 release shows 34,573 lines of context diffs since the 1.0.4 release tag; that seems like a fair amount of changes. --Ken
