In Context

Andy Hodgkinson, Novell Bandit and Higgins committer extraordinaire, announced the availability of DigitalMe for Mac OSX. Our colleague Dale Olds also wrote about it a bit here. Very exciting stuff!

DigitalMe is Novell’s name for the “client-based identity agent/selector deployment.” Each of the four identity alternative agent configurations use variations on the same basic architecture, although there’s still work to do to harmonize them better.

If you’d like to join the cloudtripper.org conversation, click here. We’ve also got the source code to some of the demo posted to the site. More to come.

There was a well attended session on the Higgins Data Model at the data sharing summit. I presented some slides that I’ve since cleaned it up a bit here (PPT) and a great discussion.

My favorite question was from Andy Dale who pointed out that I shouldn’t be drawing Digital Subjects in two different colors (depending on whether they were, loosely speaking, about “you” vs. being about “someone else”) since in truth all “things/people” are represented by one class, namely Digital Subject, and that the only thing that differentiates them is whether they are linked by “SubjectRelation” or “SubjectCorrelation” links (links to friends, etc. vs. links to other representations (facets) of the entity known as “you”.

Marc Canter was clearly uncomfortable with the lingo that the data model uses. It’s based on the Lexicon developed during an intense 18 month period of peer co-production on the identity gang list. That project leveraged earlier work by Kim Cameron, SAML, and many others. Marc may be unfamiliar with that work. Here’s a mapping that might help apply the more generic Higgins/IdGang lingo to Marc’s world:

• A node in a social graph = Digital Subject
• An edge in a social graph = a SubjectRelation or a SubjectCorrelation (see above). Kinds of Identity Attributes
• A profile as commonly used = the set of Identity Attributes of a Digital Sujbect minus the SubjectCo/Relations

I can’t think of a better way to kick of a new blog than to post about this weekend’s data sharing summit. There were folks from big sites like AOL, Yahoo, Google and Microsoft as well as lots of small cool sites Pownce, RapLeaf, and lots more. Though Marc had planned it for a while, the buzz just before the meeting around Brad’s thoughts on the social graph couldn’t have come at a better time. Facilitated in open space by Kaliya, who as usual did a wonderful job, the meeting was a chance to kick around a topic that I’ve been passionate about for years. The original question I asked myself in 2003 was, how could I merge the social sub-graphs of which I was a member into one unified graph so I could see it all in one place? So I couldn’t have been more excited to go, and I wasn’t disappointed.

For my colleagues Drummond and Markus Sabadello and I who attended with me it was a chance to:

1. Launch a new community site, cloudtripper.org, focused on this area
2. Demo some code that we’ll post to cloudtripper.org (as soon as we get SVN up!) that showed how we can access using a common API data from both Facebook and an OpenID OP service.
3. Demo the new Community Dictionary Service (CDS) that we’ve put up as a live (though experimental) service at http://cds.idschemas.idcommons.net