As in, sure you can connect the AIM and MSN chat networks, but you yourself have to run a server to connect the two.

Currently one also has to have migration functionality written by themselves. Instead, amazon could provide it, just like AIM could provide an MSN gateway themselves.

I'll probably get downvoted for it, and I'd like to be mistaken here, but I'm getting the feeling that some of you get offended by critique, justified or not, of some of US politics.

I think many of us would actually make the same critiques.

But not on this site. We don't want to talk politics here.

- show me a 2d barcode (QR-code, datamatrix, whatever) on your Sign-in page;

- I scan this code with my mobile phone application (or simply use desktop application that takes a screenshot);

- this application (password manager) makes a request to your service and authenticates me (using SRP);

- your Sign-in page uses AJAX in the background to check if I'm already authenticated;

- if the auth process was finished - I'm redirected to login-only area of your site;

So this 2d barcode would contain some kind of nonce that would uniquely identify current Sign-in page (we obviously can't use cookies because we're using different channels).

This whole idea could also make sign-up process much more simple (you could, for example, encode password requirements for your site, and my password manager would automatically generate the password according to these requirements, and would sign me up and record credentials in itself).

Just an idea. I'm not sure if it's viable at all (there might be some hard-to-overcome security issues), it would be nice if you let me know what you think about it.

2 factor auth though is very much possible with this, actually with any service. I may try to hack that up as a demo this week. Maybe use twilio for it and have it call you. Fun!

Once again, I'll try to express myself more clearly:

- we show a username/password input boxes and 2d barcode (QR/datamatrix);

- the user may simply enter his username/password and then everything works your way;

- he may also choose to use some kind of password manager that is compatible with this protocol;

- in this case we encode some data in the barcode (it could be some login URL, e.g. https://example.com/srp/yourservice.com/?nonce=some_random_n...);

- this password manager scans the code and performs full SRP authentication on the above URL; if it's successful we simply send some kind of notification to your service - or you poll our service for the result;

- we check the result on the actual login page as well (e.g. using AJAX): we may simply poll some other URL, like: https://example.com/srp/yourservice.com/result/?nonce=the_sa... (of cource results signing/etc. comes into play here);

- if the result is OK we simply set session data in the cookie as usual and from now on the user is authenticated;

This has additional advantage of the actual crypto going on in the password manager and on the server. It also gives the possibility to perform the sign-up process semi-automatically as well (think of password generation - the user doesn't have to know it at all). And then think of public wifi or other networks you don't trust: you may perform the auth via GSM using your mobile. But the biggest improvement, in my opinion, is the usability aspect. You could have your password manager lying in the background and when you are shown a login page you could simply double-click an icon in your tray area and you are immediately logged into the app. Or you could use password manager on your mobile phone to take a shot with its camera.

I hope this time it's clearer.

Of course we're 'overthinking' it again, but is the above solution viable?

    import hashlib
    hashed = password
    for i in range(50000):
        hashed = hashlib.sha1(salt + hashed)

  import bcrypt
  hashed = bcrypt.hashpw(password, bcrypt.gensalt(log_rounds=13))

    import timeit
    
    t = timeit.Timer(stmt="""\
    def test(pwd, n_iter):
        for i in range(n_iter):
            pwd = hashlib.sha1(pwd).hexdigest()
    test('hello', 50000)
    """, setup='import hashlib')

    print t.timeit(100) / 100
    
    >>> 0.126629960537

EDIT: sorry, it actually does change the path on Chrome. It doesn't on Firefox.

It's a simple discussion board, w/o registration, very (too?) clean.