Re: ACTION-56 Standardizing presentation of Site Risk Information

I think this would be useful material to add to the Goals section of
he Note, specifically to the last paragraph (as assigned to Hal in
ACTION-62).  What do others think?

Cheers,
-- 
Thomas Roessler, W3C  <tlr@w3.org>






On 2007-01-02 06:47:07 -0800, Hal Lockhart wrote:
> From: Hal Lockhart <hlockhar@bea.com>
> To: public-wsc-wg@w3.org
> Date: Tue, 2 Jan 2007 06:47:07 -0800
> Subject: ACTION-56 Standardizing presentation of Site Risk Information
> List-Id: <public-wsc-wg.w3.org>
> X-Spam-Level: 
> X-Archived-At:
> 	http://www.w3.org/mid/D0C847B2BD75414090045D8C7EA3D59402E1469E@repbex01.amer.bea.com
> 
> 
> ACTION-56 
> 
> I am re-titling this action to better reflect what I had in mind.
> 
> During the Dec-19 call I suggested that although there seems to be a
> strong consensus around not specifying any kind of algorithms for
> determining if accessing a given web site represents a higher or lower
> risk by examining its history, content or other information, we should
> at least keep open the possibility that we specify a standard way to
> indicate to the user what the browser thinks the risk level is.
> 
> Possibilities include: colors like red, yellow, green; a thermometer
> type display; numbers between 1 and 100; etc.
> 
> My reasons for allowing for this kind of approach are:
> 
> 1. Users will not be exposed to the underlying algorithm directly;
> therefore browsers are likely to use distinct means to indicate
> essentially the same risk semantics. It would be more desirable, if
> possible, to show some standardized display which users could be trained
> to look for in all browsers.
> 
> 2. The algorithms underlying existing displays, such as the padlock are
> not completely defined or identical across all browsers, yet we can
> agree that they fundamentally have the same significance. Even if risk
> algorithms change over time, either to improve accuracy or to counter
> changes made by black hats, as long as the practical meaning remains
> essentially the same, a standard indication will be beneficial.
> 
> ---
> 
> That said, I am by no means certain that we can actually come to
> consensus about a common meaning which we expect to remain reasonably
> stable and relevant for say 5 years. Thus my argument at this point is
> merely NOT to rule this indicator out of scope.
> 
> Hal
> 
>

Received on Tuesday, 2 January 2007 15:27:50 UTC