@Ekaterina: yes, maven currently; my mistake. But I'd prefer moving it to gradle if we brought it in.
@Mick: ish. submodules are finicky in CI; have to re-initialize it and kick git occasionally to get it to behave with accord. We'd probably need to do the same w/jamm but wrapping any submodule operations with retries and reinit shouldn't be too hard. The bigger issue I see is that having changes be in a separate unmaintained repo makes coordinating changes w/C* significantly harder. Hard enough that almost all of us (excepting Ekaterina and Benjamin) just do our best to avoid working with it and working around it. Since Jonathan doesn't have bandwidth to maintain jamm and nobody else is actively keeping up with it afaict, bringing it in-tree would massively lower our friction to maintaining and extending it. There doesn't seem to be a viable, permissively ASLv2 licensed alternative in the space. On Tue, Dec 30, 2025, at 4:34 AM, Mick wrote: > > > > I am not sure modules vs in-tree. Probably someone has to play a bit with > > it to provide the pros and cons here? (If it was not already done) > > > It's what we are doing with accord. It allows cassandra-accord to live as a > separate codebase and library so others are free and can more easily use and > contribute to it, but allows us to use it as one codebase part of in-tree and > not have to go through the rigamarole of releases. > > I'm not aware of any reasons why we wouldn't take this approach (now that we > have accepted the use of git submodules in the project) …? > > >
