You'd need to read up on it but the TL;DR i got from previous comments was that previously it was being funded by a company that wasn't 100% aware of the entire situation around it (they used it in their commercial product, management didn't fully understand it was being released as open source, though I also understand they didn't own the rights to the entire code base so basically that was necessary effectively). The person developing it left that company and is seeking his own funding for it but that generally the development was significantly stunted as a result.

It seems a bit here-say-ish though, so please don't assume I'm entirely correct on all fronts there and I'd encourage you to research it further!

I've looked into BCacheFS extensively. Two problems I have -- First, it's not there in features yet. It doesn't have quotas, nor snapshots.

Second, it lacks a formally published design for these features. Because the author is the only person with knowledge of how these things might work, it makes it really difficult to mitigate the bus factor while the project is still in heavy development.

Looks neat. Nobody uses it though and LKML has zero discussion on the announcements.

However, HN has a previous thread on it here: https://news.ycombinator.com/item?id=12410798

I actually use bcachefs on my main data drive and so far it has been running solid for ~3months. That said, looks like the author has gone camping/moving/re-evaluating life. Hoping the development picks back up or that it gets rebased off of 4.12 (or maybe made into a set of kernel patches?). Wanted a COW fs - tried btrfs and kept running into issues where the drive would fall into ro mode so far not regretting bcachefs.

https://aur.archlinux.org/packages/linux-ck-bcachefs/