I've worked with Bitpay while making https://www.incoin.io/ and their developer docs are average to good. It wasn't horrible, the documentation was good enough and Bitpay offered to help (and were super helpful when we asked any questions). In fact, other than having to implement BitAuth ourselves it was pretty painless.
You might be thinking of BitID, which tightly couples payment addresses to identity: https://github.com/bitid/bitid – BitAuth does not use Bitcoin keypairs.
We created BitAuth as a way to avoid coupling payments with identity, which is why the keypairs are independent and include a different prefix (0x02).
>Bitcoin addresses are intended to be single use, disposable keys rather a long term method of identification. A portion of Bitcoin's address design is intended to minimize publicity of EC public keys if at all possible.
A potential work around is to just BitAuth with a HD wallet? Thereby negating any address-reuse issues.
Hey Tony! Great work on your Go implementation of BitAuth: https://github.com/codelittinc/gobitauth – would you mind sending me an email about your AngelHack issues? eric@bitpay.com
Hey Eric! Thanks for the heads up, will send you an email now.
Hah... that library was awesome to work on. Ended up reimplementing the signature algorithm before getting a library that lets us use golang's built in ecdsa libs. Not clever!
Love BitAuth. It's super easy to use and works brilliantly.
Regarding your AngelHack prize... the same thing happened to me. I'd be more skeptical of AngelHack than Bitpay, a well-funded global company, though: http://www.businessinsider.com/greg-gopman-angelhack-lawsuit...
It's not the first time I've heard of AngelHack not following through.