Hehe, I don't think the editor makes that much of a difference (unless it has something really handy like MSVC's intellisense).
I'm not sure if this is exactly what Bain was suggesting or not, but here's what I'm thinking if we go the MD5 route:
- Password hash stored on the server as MD5(salt+pass).
- When generating the login page, the server creates a random string (I think you can handle this with a session, but I'm not sure -- never used sessions before).
- User enters password into form, javascript hashes password as MD5(random+MD5(salt+pass)).
- Server compares that to MD5(random+stored_hash).
While probably not as secure as RSA, it should be sufficient enough to stop people from sniffing a reusable MD5 hash and logging in. The only downside is that the salt string can't be kept secret.