- From: Peter Williams <home_pw@msn.com>
- Date: Mon, 30 Jun 2014 15:46:12 +0000
- To: Kingsley Idehen <kidehen@openlinksw.com>, "public-rww@w3.org" <public-rww@w3.org>, "public-webid@w3.org" <public-webid@w3.org>
- Message-ID: <SNT404-EAS105B9B4CF16740C7524697392040@phx.gbl>
I will play with this in detail, later in the week. As usual, you are redefining the baseline. Ive become well immersed in the mobile security tool chains planned (a year or two ago) that align with the “official” line on commodity security, set in the days when vendor cels and product managers at microsoft enjoyed regular cosy coffee sessions general Keith. Of course, that is all on some disrepute, now. So im interested in fresh thinking, that takes into condideration the fact that “now everyone knows…” that untrustworthiness is endemic. How does one design, given this new reality - where any and all assurance is now inherently suspect (since explicitly subverting assurance sources is part of the reality). Unlike the corporations still stunned into silence at the public exposure of the reality of the threat environment (now known to include generalized surveillance, systemic dossier collection, and layer upon layer of untrustworthy messaging making any us based entity - including w3c - an untrustworthy body on crypto/security), we dont have any such compromised history. Not trying to sound like a nut job, more one of asking a science question that recognises the reality (rather than doing the larger vendor’s ostrich act wishing for the status quo of a year or two ago). I dont necessarily expect any solution (and perhaps dont really even want a technical answer to what is a political question). But registering the threat of state sponsored subversion seems the right thing to do. How does the web react, to being militarized? Sent from Surface Pro From: Kingsley Idehen Sent: Monday, June 30, 2014 6:18 AM To: peter Msn, public-rww@w3.org, public-webid@w3.org On 6/30/14 8:22 AM, Kingsley Idehen wrote: > On 6/29/14 7:24 PM, Peter Williams wrote: >> We cannot have a “more” list of 3 million icons. And I have no >> intention of using an American brand (like Microsoft or Google, or >> ...l) for anything that has the slightest value. >> >> What do we do? > Peter, > > You don't have to remember or type in a URI when accessing a protected > resource using the Virtualized Authentication Layer (VAL) referred to > in my earlier post. I've produced a screenshot from my ODS (OpenLink > Data Spaces) based Briefcase (our equivalent of OneDrive, Dropbox, > Google Drive etc..) that displays the current authenticated identity > associated with a user agent session [1]. > > If I wanted to make a more fine-grained acl, scoped to a specific URI, > I would simply copy and paste that URI for use in my ACL. As for > users, they never need to type anything when accessing protected > resources, they simply click on a button. > > If you wanted to use your Microsoft URI in the SAN of an X.509 cert > you have two choices: > > 1. Simply generate your x.509 cert (Digital Identity Card) using YouID > -- take the Microsoft PdP (Profile Data Provider) route with one of > the following as the IdP (Identity Provider -- service that stores > public part of your Identification oriented Claims) OneDrive, Dropbox, > Google Drive etc.. > > 2. Do it by hand using provider certificate generator provided by > relevant operating system. > > Either way, our NetID-TLS (a superset of WebID-TLS) protocol with > handle identity claims authentication. In short, that's what happens > when you click on the buttons presented by the VAL dialog. > > Links: > > [1] http://susepaste.org/35303595 -- My Identifier from Microsoft's > Data Space (which is comprised of millions of other user accounts for > every Microsoft app/service user) > Here's WebID-TLS based authentication results page based on a YouID generated Identity Card where the WebID is derived from interaction with a Microsoft Account via OAuth and the WebID-Profile document is deployed using Microsoft OneDrive. Net result: 1. a WebID -- scoped to Microsoft OneDrive (following successful interaction with Microsoft Identity Provider services) 2. an X.509 based Digital Identity Card (Cert.) . When generating the Identity Card the user simply clicks on a button that triggers the handshake and profile data exchange with Microsoft's services. Note, even if the Microsoft user doesn't have an actual Digital Identity card, I can still make an ACL based on their Microsoft URI (e.g., <https://profile.live.com/cid-3a02f98c12fc49f5>) using the Microsoft Accounts specific authentication API. Basically, this is where the NetID-TLS superset of WebID-TLS comes into play. Links: [1] http://bit.ly/TwdZ10 -- actual WebID-TLS based authentication results page [2] http://bit.ly/1sRcg70 -- YouID generated artifacts collection for my Microsoft Account, deployed via OneDrive and mounted to my ODS-Briefcase (since these storage service providers don't provide open directory browsing without social media contacts data related payments) [3] http://kingsley.idehen.net/DAV/home/kidehen/Public/OneDrive/YouID/IDcard_Live_140630_084317/index.html -- HTML based variant of my Digital Identity Card . -- Regards, Kingsley Idehen Founder & CEO OpenLink Software Company Web: http://www.openlinksw.com Personal Weblog 1: http://kidehen.blogspot.com Personal Weblog 2: http://www.openlinksw.com/blog/~kidehen Twitter Profile: https://twitter.com/kidehen Google+ Profile: https://plus.google.com/+KingsleyIdehen/about LinkedIn Profile: http://www.linkedin.com/in/kidehen Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this
Received on Monday, 30 June 2014 16:04:10 UTC