Re: [ogpx] A Review of Multi-Domain Use Cases [Was: Re: OpenID and OGP : beginning the discussion ...]

[Mike Dickson <[e-mail]>]

I have to agree with Carlo on this.  This is a case where virtual life
needs to model real life.  Lots of people with the same name in the
world.  The protocol needs to support some other way to uniquely
identify a "user" other than FN/LN. 

When crossing into a new region it seems highly possible that a region
owner may want to require a re-authentication also. 

I'm not suggesting a solution becaus this isn't my personal area of
expertise.  Simply stating I see the problem and agree a good solution
needs to be found. 

Mike

On Mon, 2009-06-29 at 11:45 +0000, Carlo Wood wrote:
> On Mon, Jun 29, 2009 at 04:13:08AM -0700, Meadhbh Siobhan wrote:
> > On Mon, Jun 29, 2009 at 3:51 AM, Carlo Wood<[e-mail]> wrote:
> > as mentioned in the draft, deployers MAY choose to identify users via
> > account ids rather than first+last.
> 
> But that is local to the deployers.
> 
> What is used to ID an agent in the server-server protocol
> when a user walks from one region into another, or teleports
> to a different region?
> 
> Firstname Lastname should be merely payload, not interesting
> to the servers, not being used for anything except passing
> it on the the viewers so they can use it to put in the tags.
> 
> There must be some other string, that is globally unique,
> that is used to ID the agent. I was missing that string
> in the first post that started this.
>