[hnov]

I think making sqlite distributed-capable is very interesting. Postgres isn't trivially distributed and is a full fledged database server that's not appropriate for scale to zero workloads, though others are working on that as well.

[maxbond]

These do all sound like interesting ideas, but in our current Cambrian explosion of new databases (especially "X open-source SQL database you know and love, but distributed" forks) - you can fork SQLite and build that without being needlessly confrontational about it, or including the language about how they'll rejoin SQLite if their terms are met, which makes me feel like this is a fork about leverage as much as realizing a different vision for a database.

[hnov]

I imagine they love the sqlite codebase but are not ecstatic about having to maintain a fork and would like to have sqlite upstream their changes such that they don't have to import patches in perpetuity. This is a valid frustration that most commonly happens with OSS that's built by a corporation and thrown over the wall. I think in this case the dig at the end of the README is a way to be like "look at those religious cult members, closed off from accepting contributions" which is a cheap shot but all's fair in politics. A lot of companies build on top Linux, PG and others and are able to reach consensus with the community.