For an example, jq has gone without maintenance for a long time now, but it's very challenging to gain critical mass around a maintained fork, so the un-maintained fork remains state of the art.
Used to be open-source, became the defacto OAuth implementation, but now it's closed source.
https://github.com/timescale/timescaledb/blob/main/tsl/LICEN...
There's a marked difference between what they allow you vs Mongo (SSPL) and Elastic (Elastic license) -- see (b).
Also generally the Postgres ecosystem has in general seemingly found a way to strike a balance between creating immense value (point to a F/OSS DB that's better than Postgres) and enabling commercial success (Citus, 2nd Quadrant, EDB, Timescale, Supabase, Neon are all very successful)
I do agree with the overall trend though -- I already think the future is going to be AGPL[0] (but that's based partially on a misunderstanding of what AGPL does).
It remains to be seen if people will cynically make projects open source to start and then switch to something like SSPL/BSL/Elastic later, but the unintentional rug pulls will probably continue for a while.
[0]: https://vadosware.io/post/the-future-of-free-and-open-source...
Timescale's community manager.
The problem with the SSPL, as I see it, is that it's incompatible with pretty much every other open source licence and software, possibly even running it on linux. Something similar that requires everything to be released under some kind of FOSS licence seems desirable, but I don't know if that could be defined legally.