Phone-A-Friend

@rrhoover Thanks for featuring it! We're sort of targeting hands-free use cases like driving with it. So, on your morning or evening commute, you can just say something like "Siri, call a friend on Phone a friend", and we'll automatically connect you to the first available friend who is also free.

@dredurr It's a double opt-in. One person taps the button to multi-cast their availability to their friends in a light way. The recipients can tap the notification if they're also free, or just ignore it.

@alicelthwaite Nope! You can choose to have the app send them a text message for you when you're free. If they're also free, they can tap on the number in the text to connect to you (or your group call)

@rohanseth @alicelthwaite That's super interesting. Think you should mention that feature more!

@jimt1277 Yes, you get a chance to pick who to send it to before it goes out! It even remembers your last selection to make it easy to send to the same set of people again! Thanks for trying out the app!

@rohandang Awesome, thanks. Just an on-boarding comment: I'd suggest the "How does this work" swipe-screen at the beginning of onboarding rather than at the end. I'm already sold, but seems like it might boost adoption for those wondering what the app is. Cheers.

@gregcohn Hey greg, thanks for your comment. Do you think something like this works better as an alternative to scheduling time / office hours? We've been brainstorming quite a bit about what this unlocks, would love to chat with you more since you've spent time thinking about the problem. Also, please let me know how the app worked out for you: rohanseth@cs.stanford.edu

@rohanseth Happy to chat and will PM you, but to address your public question, I think scheduling and office hours models are just different planning paradigms, and that there is distinct value to the ad hoc paradigm. Office hours, for example, always seem more like a 'one-to-many' solution for someone who's high-demand/low-availability to say 'come at me', rather than the right model for peer-to-peer. One of the challenges is managing the tradeoffs between trying to have a highly brokered process (lots of context and filtering) vs a highly loose process (lots of potential for unwanted calls). Looking forward to seeing how you work through those and to talking.