+1 to Nat's point about supporting "virtual aggregation" under user control.
Chris, on the larger question you ask, the Data Sharing Summit was not about how others that have your data can share it, but rather about how you can control and share that data the way you want with whom you want. To be clear, not all the attendees of the DSS subscribe to your vision of "100% user-centric aggregation". A number of them were services that provide some type of aggregation in which the user is in control, but the site is still able to offer services/applications/other-value-adds based on doing this aggregation on the user's behalf. Several new social network aggregation services like Pownce have this philosophy. They would say that the user elects to trust them as a service provider just like the user elects to trust an IdP (or OP as OpenID calls them). Anyway, I just wanted to clear up any misconception about what the "Data Sharing Summit" was about. Without that context, I agree that name could make it look like a really "Bad Idea". =Drummond > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf > Of Nat Sakimura > Sent: Sunday, September 09, 2007 10:18 PM > To: Chris Drake > Cc: 'OpenID specs list'; 'ID Schemas' > Subject: Re: [Idschemas] identity schema element metadata:using > existingspecifications > > Hi, > > Instead of having one single master copy at the IdP, I would prefer one > single piece of each information disparsed over the network (optionally > with opaque identifiers) and having IdP managing the "links" so that I > can control all the pieces from one place. I feel that having everything > at one place is too risky. > > =nat > > Chris Drake wrote: > > Hi, > > > > Having missed the summit - can anyone tell if there was any dissent or > > scaremongering going on? The idea of assisting everyone who's > > collecting information about me, to share it easily, seems like an > > exceptionally "Bad Idea" (tm). > > > > If anyone's building anything that assists these companies to "give > > up" or relinquish their collection of my data, in favor of letting me > > hold the one single master copy if it myself, on my chosen IdP, and to > > let me arbitrate who can use it, when, and how - now *that* would be > > something to congratulate people about. > > > > Search google/google-news for "paper shredder" ... > > > > Kind Regards, > > Chris Drake, > > =1id.com > > > > > > Saturday, September 8, 2007, 5:33:20 PM, you wrote: > > > > DR> Mark, > > > > DR> I just wanted to say that based on what I learned about them at the > Data > > DR> Sharing Summit (http://datasharingsummit.com) today, and what I read > on my > > DR> first pass tonight, these are fine pieces of work that really push > the ball > > DR> forward. > > > > DR> Hats off to you. > > > > DR> =Drummond > > > > > >>> -----Original Message----- > >>> From: [EMAIL PROTECTED] [mailto:idschemas- > >>> [EMAIL PROTECTED] On Behalf Of Mark Wahl > >>> Sent: Thursday, September 06, 2007 6:05 PM > >>> To: ID Schemas > >>> Cc: OpenID specs list > >>> Subject: [Idschemas] identity schema element metadata: using > >>> existingspecifications > >>> > >>> > >>> I'm reformatting the table of identity schema metadata, located at > >>> http://idschemas.idcommons.net/moin.cgi/MetaData, into a > >>> pair of more compact and usable specifications. One spec describes > >>> where the existing well-known metadata (e.g., Dublin Core) should be > >>> used when describing identity schemas and their schema elements (i.e., > >>> attribute types and claim types). The other spec will describe how > >>> to represent identity schema metadata for which there is no > >>> pre-existing well-known specification. I've attached a copy of > >>> the first draft of the "Identity Schema Element Metadata: Using > >>> Existing Specifications". > >>> > >>> Mark Wahl > >>> Informed Control Inc. > >>> > >>> > > > > > > DR> _______________________________________________ > > DR> specs mailing list > > DR> specs@openid.net > > DR> http://openid.net/mailman/listinfo/specs > > > > > > > > _______________________________________________ > > specs mailing list > > specs@openid.net > > http://openid.net/mailman/listinfo/specs > > > > _______________________________________________ > specs mailing list > specs@openid.net > http://openid.net/mailman/listinfo/specs _______________________________________________ specs mailing list specs@openid.net http://openid.net/mailman/listinfo/specs
