I'm having some sticking points with how to do something in Glass. The current flow I'm thinking of for my application is almost identical to the existing Mirror API flow for things like Evernote or Sharing photos. Except for one major caveat, the first thing a user would do with my Glassware is to search from something and then be presented with a list of possible results. Then they could select one of those and post it, along with a comment to my service.
To post to me, they will need to be authenticated or at least somehow provide identity information. The two ways I came up with to do this is to either provide a QR code with an API key tied to that user and scan that with glass so that it can store and send that key with the request, or to use the Mirror API somehow. If the user logs in on my end, then goes through OAuth with Google, I can immediately subscribe to some sort of custom event or add a Contact for posting updates through my Glassware or something, and Google will give me the user identity with the callback. However, I was planning on creating an immersion to allow the user to page between results rather than inserting static cards and don't know how to start a Mirror API call from there.
The QR code thing seems hacky, though there is precedent with the WiFi Settings...but it just doesn't seem appropriate. I'd much rather go through the mirror API, any ideas?
I'll accept that the capability just doesn't exist yet and I should file or expand an existing issue...but I just wanted to check beforehand how everyone was dealing with this now.