Hacker News Reader

Comment by alt227 | Hacker News Reader

So if we now have Vaultwarden + keyguard can these things move away from Bitwardens api and pursue their own?

Is there any reason to? It's kinda nice that they all stick to the same contract. I don't really like the Bitwarden desktop app, but because there's so much code out there, it was pretty easy to have an LLM write a detailed spec for a client and then implement something that works for me.

The Bitwarden web client has actively broken Vaultwarden compatibility at least once <https://github.com/dani-garcia/vaultwarden/issues/7607>. It’s reasonable to expect they’ll do so more, carelessly or deliberately. They already have features and UI that don’t make sense for Vaultwarden, and the rate of such things feels to me (as a mere user) to have increased. When Bitwarden continues in the direction they’re going, I think a hard fork of the entire ecosystem for Vaultwarden is fairly likely, and it will almost certainly be a good thing, and in hindsight we’ll probably wish the fork branched off earlier.

> The Bitwarden web client has actively broken Vaultwarden compatibility at least once

I skimmed the thread, so maybe I missed it, but from what I saw, saying they deliberately broke Vaultwarden is a bit if a stretch. They likely develop their apps without taking into account others (and rightfully so), and it happened to break an unsanctioned 3rd party implementation. Nothing nefarious.

I didn’t say deliberately, but I did say actively. It sounds like this was something that would have broken older Bitwarden self-hosting servers too (no idea how old, or whether this is true at all, for I haven’t checked).

The “nefarious” dividing line is often hard to assess. For a better-known example, it’s interesting just how many times Google has done things like degrading their stuff on Firefox by user-agent string detection, and then been all apologetic about it, it was a mistake, &c. Malice and incompetence can be indistinguishable.

Reply on news.ycombinator.com