On 2006-05-22 13:53:00 -0400, Sam Hartman wrote:

> I think we are in broad agreement that the interesting work
> in this space must involve the UI and is not principally a
> protocol problem.

I very much agree.

Incidentally, you may want have a look at the report from the
March W3C workshop:

  http://www.w3.org/2005/Security/usability-ws/report

We (W3C) are currently thinking about how to best charter work
that would specify some browser user interface components that
would have to be outside the control of web sites, and could be
used to make sure that users know (as opposed to look at on
their screens) where they are going to send their confidential
information.

Another element that we took as important from the workshop in
NYC is to enable user agents to reliably recognize HTML forms
that are used for authentication.  This ability would enable
user agents to manage credentials on behalf of the user. It
would also enable user agents to *not* submit credentials using
HTTP POST (even when entered through HTML forms), but instead
grab them and use them for whatever HTTP-level authentication
mechanism is used.  User agents could also do intelligent
things in the UI to make sure that users understand what they
are doing here.

PS: I'm currently at WWW 2006 in Edinburgh.  If any of you guys
want to chat more about this, please feel free to drop me a
line, so we can meet up somewhere.

Regards,
-- 
Thomas Roessler, W3C   <[EMAIL PROTECTED]>

_______________________________________________
dix mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/dix

Reply via email to