We tried to put so many new features into b2evolution v4, and we tried to make everything so perfect, that by asking us for a release date, we just had to make a wild guess... Especially: a guess about when we'd be satisfied enough with the outcome to make a public release.
Well, it turns out that at the end of April, we were not satisfied!
Sure, we were satisfied about everything that had already been done. But we were not satisfied about the big picture and the global coherence of the development at that stage. Making an alpha release at that point wouldn't have made much sense beyond "being on time" for the sake of "being on time".
There would have been features partially implemented and others not working because they depended on something else to be finished... So, instead of releasing a version that would have been "broken but it's an alpha, what did you expect?" we decided to wait a little and release an "alpha version you can actually run on production"... because that's just we way we like our alphas to be! ;)
So, we took the to-do list, combed through it carefully and came up with a feature set rather than a date. That feature set is not perfect, but it will still make a coherent alpha. Based on that, our best guess would be a release early June.
Later than expected, but you'll get so much more out of it that you'll probably agree it was worth the wait. After that, we'll resume to incremental changes and more regular releases.
Btw, in your opinion, what would be the ideal delay between successive releases?