The fact that it's meant to be run in a browser is somewhat concerning. HTTPS is... okay... but for a GPG replacement I wouldn't really want to be dependent on the CA system or a remote server.
1. You can run it on your own server, and you can run it from a FILE: URL (though that part is still work in progress).
2. This implementation is not the only one there will ever be. The goal of this implementation is just to make something that non-technical people can easily use in order to establish the (open) file interchange protocol. Then anyone who wants to can write native clients, email plugins, etc.
I've never seen it before, I'll investigate when I get home, but it looks great at a first glance. PGP does need a bit of an update, at least from an implementation perspective, in my opinion.
Hmm, the demo doesn't work for me, but this looks like a great idea. I'll agree with the other commenter, that I can't trust my browser as much as I can trust my OS, but I understand that this is just a proof of concept.
It doesn't??? It should. Can you please tell me what browser you're using, and what symptoms you're seeing? (Feel free to send it to the email address listed on the github page if you don't want to clutter up this comment thread.)
Feel free to email me, my email is in my profile. I also toyed with the idea of implementing something like this with TweetNaCl, so props to you for actually going through with it.
That works, thanks. As far as I can see, there's not much glue code between TweetNaCl and the browser. Have you thought of writing/porting this to something like Python? Making it a command-line utility or a library might make adoption easier.
https://github.com/Spark-Innovations/SC4