I disagree with Goob and Fla. I would love to see an Android specific app. The mobile version of the site is usually incredibly slow to load for me and like Koen said, sharing is too inconvenient. We do not currently have an API though, so that might put a kink in this project.
Part of what needs to happen to get an API going would be to set up Diaspora as an OAuth2 provider. Alternatively, we could try OpenID instead, but OAuth seems to work well enough with some decentralized social platforms already. (Tent and Pump come to mind).
A quick search brings up two gems that might be worth taking a look at:
That’s only half of the battle though. We also would need to expose our APIs in such a way that developers could easily leverage them. It’s worth noting that much of D* uses Backbone.js already, which comes with a RESTful JSON interface.
Don’t get me wrong: I’d be more than happy for someone to create an app for Android, iOS, Firefox OS or any other mobile OS. I just don’t think it should be a core development priority.
It would be better to put the limited resources we have into improving the mobile site for all devices, and ironing out any issues there are with it, such as those you raise. But if an independent developer wants to develop and app, that’s fine. As long as they can commit to keeping it up to date or opening up development to others, so that we don’t end up with yet another defunct, out-of-date app which people find on search and try to use, as many Android users are still doing with the Android Webclient alpha, and then thinking that Diaspora itself doesn’t work properly. That’s the danger with apps.
@koenmartens , someone is working on an Android app at the moment. If you want to take part, contact him via his profile page.