[gary_0]

Even if the price wasn't secret, maybe when you hit $10M revenue 3 years later, they'll have raised the price 300%. Maybe they've changed the license again and now its priced per-seat or per-cpu-core or per-employee-tooth. Maybe they've decided to focus on larger enterprises only, and they no longer want your business.

[vbsd]

This is true but you’ll have these risks with any paid service/dependency that has non-zero migration cost. Startups still use such products all the time, assuming that a huge price spike is not that likely or that there’s going to be some viable alternative with a migration path. I’d say it’s quite different when you can’t even guess what your starting point is going to be once you hit those $10M.

But I do agree with the general argument that lock-in as this for your core technologies (a database certainly counts) should be avoided.

[subarctic]

With a database I'd say it's especially risky and lockin is extra strong. Compared to, say, OpenAI where you're just using it to generate text and its behaviour is already unpredictable anyway

[dartos]

IIRC cockroachdb support about every Postgres client.

It “speaks” Postgres.

Not a ton of vendor lock in, since you could always migrate to another Postgres hosting service (although you’d lose some of cockroaches regional features)

[vbsd]

Even if CockroachDB uses Postgres wire protocol, it doesn’t mean you can easily migrate all workloads supported by CockroachDB to Postgres. The scaling properties are different.