- From: James A. Donald <jamesd@echeque.com>
- Date: Fri, 23 Jun 2006 07:31:09 +1000
- To: public-usable-authentication@w3.org
-- Chris Drake wrote: > The word "Chrome" is so cool that nobody wants to put > it back on the shelf where it belongs! In the paper world, letterhead conveys trust, though perhaps it should not, trust in the message, which is part of the reason that fake websites work so well. Obviously we have to make this instinctive reaction work for the user, rather than against him - which means that the web site's control of the stuff at the top of the screen should depend on the recipients trust in the domain - whitelisting as in the Netcraft toolbar, CA certificates, and past visits to the domain or messages from the domain that passed the spam filter. In particular, past successful logins to the domain, as in Trustbar and some incarnations of Passpet. (I fear that someone is going to tell me that the users reaction to general appearance of the top of his window is off topic, or "out of scope" also.) I rather like Phillip Hallam-Baker's phrase "Secure Internet Letterhead" - which unfortunately now means his specific proposal and RFC 3709. If it had instead meant a trusted path for conveying identity and trust information to the user, (trusted path assuming your computer is not owned) which is almost what we now mean by "secure chrome", this would have established the right user factor orientation to this discussion. I restrain myself from discussing Ebay's success and PKI's failure in providing trust information to users, since everything that Ebay does to succeed is quite clearly out of scope for this list. --digsig James A. Donald 6YeGpsZR+nOTh/cGwvITnSR3TdzclVpR0+pr3YYQdkG AVayE7tumsJ8TFYzGCST0ClHA6WeaX5jKV2cI8i6 4GlamBdlaVt12yNUEBEPZKGFd89jN4gveKxgwjYpM
Received on Thursday, 22 June 2006 21:31:22 UTC