Re: [Ietf-http-auth] WAE topic/problem scope (was: Stab at grouping of problem (was: Re: [dix] Agenda bashing))

[Lisa Dusseault <[e-mail]>]

The outcome of WAE could be 0, 1 or more WGs.  The conflation in the  
BoF does not necessarily mean conflation in WG(s).  The breakdown is  
appreciated.

Lisa

On Jul 13, 2006, at 11:59 PM, Jeff Hodges wrote:

> Just to attempt to help clarify, in my understanding WAE was begot  
> by the conflation of three topic areas..
>
>  1. in-protocol authentication (for HTTP)
>
>     e.g. that which is defined within the HTTP spec set itself and is
>     used by an HTTP client to authn directly with an HTTP server.  
> Presently
>     the two available mechanisms are Basic and Digest. There are  
> use cases,
>     eg CalDAV, that would ostensibly benefit from a wider range of
>     in-protocol HTTP authn mechanisms. My understanding is that  
> this was the
>     motivation for creating the ietf-http-auth@ list.
>
>
>  2. application-level authentication/SSO and attribute sharing (for  
> HTTP-based
>     web-oriented (eg portal-based) apps)
>
>     This is a well-trod area with a plethora of existing solution  
> approaches
>     which are deployed to varying extents: SAML web sso profiles,  
> Liberty
>     ID-FF, Shib, OpenID, SXIP(/DIX), LID, WS-Federation, RoboForm,  
> etc. The
>     primary reason this topic is on the table in this venue is a  
> perception
>     that perhaps "more" can be done in order to facilitate wider  
> and more quick
>     adoption amongst websites in the wider Internet, eg "the  
> blogosphere".
>     This was the motivation for creation of the dix@ list.
>
>
>  3. anti-phishing
>
>     The motivation why this is on the list is obvious.
>     Effective overall solutions will involve a large component of user
>     interface (UI) approaches. Some would argue that the UI aspects  
> are the
>     first-order ones (and this is not a typical IETF problem  
> domain). Though
>     as well as UI, any solutions will likely rely on capabilities/ 
> properties
>     obtained from solutions to 1 and/or 2 above, and may require  
> specific
>     capabilities/properties that 1 and/or 2 don't otherwise provide.
>
>
> At this point, it isn't clear to me that the WAE BoF represents  
> just one overall "problem" to solve. Each of these are large  
> distinct topic areas in their own right, though they do intersect.  
> It will be a challenge to not short-shrift one or more of them. It  
> should be an entertaiing discussion.
>
>
> JeffH
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Ietf-http-auth mailing list
> [e-mail]
> http://lists.osafoundation.org/cgi-bin/mailman/listinfo/ietf-http-auth


_______________________________________________
dix mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/dix