[OpenID] Google+ and Unique Identifiers -- different again?

[Steven Livingstone Pérez]

Totally agree, though i well remember the amount of effort in that went into element and attribute normalization [1] in defining algorithms that meant everyone "saw" the same value.
When those algorithms (loosely defined as in this case - as many - the algorithm is something in a closed environment that just knows what the mapping is) are hard to define i can't see how else, in simple terms, consumers of the service or api,  are expected to make queries on behalf of a users account, other than by having a way of asking if two things are alike via a service provided api.
/steven
[1] http://www.w3.org/TR/REC-xml/#AVNormalize

Subject: Re: [OpenID] Google+ and Unique Identifiers -- different again?
From: henry.story at bblfish.net
Date: Sun, 3 Jul 2011 14:25:09 +0200
CC: melvincarvalho at gmail.com; jernst+openid.net at netmesh.us; openid-general at lists.openid.net
To: weblivz at hotmail.com




On 3 Jul 2011, at 14:17, Steven Livingstone Pérez wrote:Probably make some sense for them to have a "Normalize()" graph api call ... i have often been worried about this in storing identifiers as keys etc in a local data store.
I do think it it the responsibility of the account provider to provide the mapping rather than us trying to prejudge the next migration choice.
You can't escapte the fact that in an open world the same thing will be named with more than one name.
That's why the semantic web was developed.
Henry

/steven

> Date: Sun, 3 Jul 2011 13:58:43 +0200
> From: melvincarvalho at gmail.com
> To: jernst+openid.net at netmesh.us
> CC: openid-general at lists.openid.net
> Subject: Re: [OpenID] Google+ and Unique Identifiers -- different again?
> 
> On 2 July 2011 05:48, Johannes Ernst <jernst+openid.net at netmesh.us> wrote:
> > It seems Google has changed their unique identifiers for people again.
> >
> > Apparently I'm now:
> >        https://plus.google.com/104555285104903729468
> > as opposed to
> >        http://profiles.google.com/Johannes.Ernst
> > and so many other variations over the years.
> >
> > My relying party implementation does not recognize me any more although I use the same URL as identifier. Which means I can't access my account!
> >
> > Is it me who is doing something wrong here? What's the official Google migration path?
> 
> I've just realized facebook have 5-6 IDs all tied together
> 
> 1. Original email address
> 2. facebook.com/foo
> 3. facebook.com/UID
> 4. foo at facebook.com
> 5. graph.facebook.com/foo
> 6. graph.facebook.com/UID
> 
> This is very clever stuff, imho. I think the FB graph is extremely
> well organized, and possibly gives them a competitive advantage.
> TimBL always says, 'give everything a URI and let them link to each
> other'. Facebook have done exactly that, and I think it's the design
> model to follow.
> 
> >
> > Thanks,
> >
> >
> >
> > Johannes.
> >
> > _______________________________________________
> > general mailing list
> > general at lists.openid.net
> > http://lists.openid.net/mailman/listinfo/openid-general
> >
> _______________________________________________
> general mailing list
> general at lists.openid.net
> http://lists.openid.net/mailman/listinfo/openid-general
_______________________________________________
general mailing list
general at lists.openid.net
http://lists.openid.net/mailman/listinfo/openid-general


Social Web Architect
http://bblfish.net/


 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-general/attachments/20110703/242d4d20/attachment.html>