W3C home > Mailing lists > Public > public-xg-webid@w3.org > January 2012

use proxy profiles , for use with foaf.me <-> foafssl.org websso protocol

From: Peter Williams <home_pw@msn.com>
Date: Sun, 8 Jan 2012 09:26:35 -0800
Message-ID: <SNT143-W130C7BA22E114D089325C9929B0@phx.gbl>
To: "public-xg-webid@w3.org" <public-xg-webid@w3.org>




Ive got a rediculous situation that none of my profiles work at Henrys site (using the foaf.me - foafssl.org handoff). But, they do work elsewhere. Im going to revisit the proxy URI that once wrapped my yorkporc2 blogspot account, and which once worked at foafssl.org. While foafssl cannot consume the profile directly, it could consume the content translation (and name remapping) produced by linkeddata.uriburner.com. That is success (and actually advanced). Once that works (again), Ill try to do the same for by having uriburner crawl my own idweb RDFa profile, too  The goal is for peter to be able to logon to foaf.me (and then make a new profile and new cert, I believe, since its acting as a profile importer cued by a webid signin event) Then Ill finish up on the RP side with some more equivalency work, mapping all profiles onto the one proxy URI (including the new one from foaf.me).  Using openid flows, folks can always use then the linkeddata endpoint as the "central verifier". Openid auth 2.0's name-delegation flow means that the proxy URI is hidden from relying parties, who only see the user-controlled name. This feels totally natural (and what made openid originally SO SPECIAL). Its great that its the semantic web that actually realizes the original openid. That just leaves the IDP side, with my own cert using system and validation logic. Ill see what time is left, to try again there, on a public endpoint.   		 	   		  
Received on Sunday, 8 January 2012 17:29:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 8 January 2012 17:29:37 GMT