But for Bitwarden, the interface is a much smaller proportion.
Can you elaborate on that? Bitwarden’s apps use Bitwarden public API, similar to how the Voyager app uses Lemmy’s public API.
Sure. Bitwarden provides its own backend. So that backend represents some portion of their code base. In the case of Voyager, Lemmy provides the backend. So that backend isn’t a portion of your code. So Voyager is 100% frontend. Bitwarden is < 100% frontend.
I don’t really see how developing a backend or not has anything to do with the decision to build a native or cross platform app.
Because it changes the risk benefit profile of the choice. Imagine that your backend is 70k hours of work and your interface is 1k hours. Managing two interfaces isn’t going to seem like nearly as big an ask so other variables may get a higher weight. Of course those numbers are contrived for the sake of explanation, but if you still don’t think there are any circumstances in which others may value the benefits of native applications over cross platform applications, that’s fine. My point is simply that it may not seem like the trouble of managing two frontends is as insurmountable as you may think.
But I have a hard time believing you don’t think it is possible that there are any situations where one might reasonably believe it worth it.
