RE: ACTION6: URL display as anti-pattern

 
Stephen Farrell wrote:
> Close, Tyler J. wrote:
> > My perspective is that the difference between chrome and page area 
> > should be the difference between "browser says" and "web site says".
>
> Nice distinction.

Thanks.

> However it immediately suggests a generalisation to:
> 
> "<<foo>> says" for foo== browser, web site, PKI, reputation service,
> corporate-AAA, content-delivery-network, etc..
>
> It'll be interesting to see which of those and/or others get traction
> in the group and whether or not we'll end up having to allow or
> recommend some kind of pluggability to handle that, S.

Good point.

A possible direction is emphasizing the need to make it clear to the
user who is presenting a particular piece of information, and providing
a set of GUI conventions for communicating who is speaking.

I am somewhat hopeful that we can get significant protection from
phishing using only ceremonies involving the browser and the web site. I
think it's feasible to represent third-party recommendations using page
transitions. For example, my employer's web site provides a hyperlink to
my 401k manager. If we can secure such introductions, we can support use
of arbitrary recommendations.

Tyler

Received on Wednesday, 6 December 2006 19:00:03 UTC