I'm following the work that still is being done on Hivemind and I want to congratulate the whole team for this impressive effort! We are going to have a powerful and much more flexible tool at our disposal.

However, I think it would be unwise to perform the hardfork when all of these modifications have not been exhaustively tested. Having migrated all data and non-consensus APIs out of hived to hivemind makes the latter crucial for layer-2 applications. If it did not function properly, the entire ecosystem would be weakened.

So, although I am eager to deploying HF24, I don't think September 22 is still realistic. Waiting another week won't make a big difference but that could save us from a negative experience for users who would see their UI and (d)Apps stop working.

Better safe than sorry!