@ihermit: The risk of trying to voice your opinion as accurately as possible, so that people understand what you are trying to say, is that you use a lot of words, and people tend to relate the number of words you spend on something to how important you feel it is. FWIW, I just use lots of words, all the time
The version number in and of itself may be not that important. Just a label, right? (Although I think, yes, the version number in and of itself is
actually more important than you appear to think - maybe just not as important as you think I feel it is, due to the number of words I used to voice my opinion
). Anyway, the version number wouldn't "suddenly" jump, it would be increased appropriately to indicate an advance in features, as opposed to just being a maintenance release, which is what the version number "3.0.6" indicates to anyone who has not read the announcement on phpBB.com.
You say "3.2 is already in development, so why change it now". That's reverse logic and, IMHO, is actually the problem (and the reason why this is more important than just the importance of the label that is stuck to the release); deciding on a very ambitious feature set up front
, putting the label "3.2" on that in advance, only to realize it's probably a bit too ambitious and now having to do an intermediate feature release that is not designated as one, by backporting stuff into the 3.0 branch. I think this being "necessary" ("necessary" definately being a relative term, but I think these advancements are very welcome) is an indication that maybe the way the project is planned needs some thought.
I think that maybe a better approach would be to come up with a long list of features that you would like
, and then going off and working on them (so far so good, exactly what the developers have been doing). However, instead of setting this feature list in stone, just see how far you get. Instead of making the process feature driven, it can be time driven. After a fixed amount of time you evaluate whether it makes sense to work towards a stable release with a subset of the features you've been working on (i.e. the ones that can be stabilized fairly quickly). Maybe you decide, no, we don't have anything yet that we can get stable very soon, or we really need feature X (and Y, and Z) to be complete as well. But it gives you the chance of deciding at a later stage on the features that will be in the next release, which makes the process more flexible. Actually, what is happening now isn't all that different, except that moving these features into the 3.0 branch is a workaround, a bandaid to relief the most pressing needs in the community. It's not a natural part of the process.
But alas, again, FWIW. I don't want to give this thread a completely negative vibe, because in and of itself, making these features available to the community is a great thing. Kudos to the development team for making this decision. I just hope that they will now also look into maybe turning the development process around a little, because these sidesteps really shouldn't be necessary if you want to get a more significant feature release out the door. Oh, and consider bumping up the version number a bit more than just 0.0.1, because the list of new features definitely warrants it