Gavin King

> It happens because, at some point, it calls `org.hibernate.type.DbTimestampType.seed`. I think there's some other missing code for database-generated versions actually. (Noticed earlier today.)

(So it's not so much a bug, just that we don't support the `@Source` annotation currently.)

> I thought I caught them with that `UnsupportedException` I left but it seems it was a wrong assumption. Apparently. I say we document this as a limitation and leave...

(Unless you think it's like really easy to fix.)

> I suspect this might be complicated because it deals with an ORM type and I'm not sure how much work it takes to have a reactive version of that...

Uffff. This looks like it is going to be *painful* to untangle. The whole type hierarchy of batching loaders has completely changed, and it looks like our batch loading support...

@DavideD @Sanne in terms of the roadmap ya'll are working out, this is perhaps the highest-priority major issue after the schema export stuff.

@gbadner has offered to work on this one, which is *awesome* because she knows this part of the codebase much better than we do.

@gbadner it's about reworking the code we have for batch loading to adapt it to the newly-refactored code in ORM. But the schema export stuff is definitely a higher priority.

Ah yeah, this is a good question. There's no "bottled" way to do it rn, it's something I forgot we had to come back to. I also forget whether the...