4 ms·
Firstly, Facebook really do "move fast and break stuff," and dealing with their APIs is both slow and aggravating. The app I'm working on has a lot of social fu
by andrewf 16y ago
Firstly, Facebook really do "move fast and break stuff," and dealing with their APIs is both slow and aggravating. The app I'm working on has a lot of social functionality, which is why I'm bothering. (You may have similar plans). I've never implemented Google account login, but if I didn't care about the social stuff I'd be leaning towards that.
Are you doing a "hard launch" with lots of PR, etc? If so, I would be thinking very hard about the risks in being FB-only. A lot of people will never visit your site again.
If not, then you have more freedom to experiment. I'd be inclined to go with my gut, but: metrics, metrics, metrics. From day one you should know how many people are being pushed to authorize your application to know about their Facebook identity, and how many of them are coming back after clicking the "Authorize" button. Facebook has a dashboard which I haven't investigated, but it may be good enough for you to avoid implementing your own metrics.
- onwardly 16y agoThis is a good point. We are in fact doing a "hard launch" as a result of the need for a critical mass of users. Its a videochat website, and if noone is there then you'll just leave. As I mentioned in another comment I made just before this, I'm certainly reconsidering the decision to do FB-only. The main thing I want to stay away from is allowing users to create accounts. We don't want pervs to be able to get booted off and then just quickly create another account. But it does seem to make sense to offer Google, Twitter, and perhaps another sign in option.