Blogger: Bob Blakley
Folks have been commenting on our relationship work since Phil posted his Technometria interview with me on the subject. Adam picked up on the talk with a characteristically excellent entry, and I agree with him that identity is created through relationship; in fact this is one of the points I was trying to make in the interview. Gunnar read Adam’s post and added his thoughts here, and Tom Scott worked the relationship theory into a discussion of OAuth and OpenID here. In the comments to Tom’s post, Chris Messina asked him for a reference for the three types of identity I discuss in the Technometria interview. Until now, the reference (our “Relationship Layer for the Web” paper, which Kaliya referenced in her blog entry on the addition of Facebook to the OpenID board) has been available only to Burton Group customers. I’m pleased to announce that as of today, it’s freely available to anyone at this link.
The conversation about this paper has actually been going on for a long time. Drummond kicked it off last May after reading a pre-publication draft.
Ryan, who also reviewed a draft of the paper, was confused by my proposed relationship data structure; he thinks I’ve introduced an extraneous third party into the picture and turned what should be a bilateral relationship into a multi-party affair. Not so! The relationship data structure is created by someone (in the case of my example in the paper it’s created by Facebook) and it asserts information about the relationship between a set of parties. But there’s no constraint on how many parties are involved: the relationship data structure could be created by one party to describe his (her) relationship with another party. Or it could be created by a third party (Facebook) to describe the relationship between two individuals who meet “under the auspices of the third party”. In this case the third party sets the rules, and the individuals who subscribe to the third party’s services agree to abide by the rules. This is simply a three-way relationship. These arise in non-commercial contexts as well as commercial contexts; societies which practice arranged marriages are a good example.
If I wanted to I could create my own relationship object and offer it to the public as the basis for forming online relationships with me – this would not require introducing any third party Identity Provider into the picture, as Ryan claims. And proposals to do this already exist; individuals setting the terms on which they will provide information to merchants is part of what Adriana Lukas has in mind with her Mine! Project.
Now that the paper is freely available, if you’d like to join the conversation, post a comment here or put up your own entry and track back to us. We’d love to know what you think.
Credit where due: though I wrote the relationship paper, the work it's based on was done by our entire IdPS team, and builds on the "laws of relation" work of our former fearless leader, Mike Neuenschwander.


When I try to access the paper using the link given above, I just get an error message telling me that Adobe Reader can't open the pdf file "because it is either not a supported file type or because the file has been damaged." Could you please look into this?
Posted by: Disappointed | February 13, 2009 at 11:44 AM
Those of you who, like Paul (http://connectid.blogspot.com/2009/02/when-is-log-in-not-log-in.html), would prefer not to register to get the paper can email me (Bob) and I'll send you a copy; the address is my last name at burtongroup.com.
Posted by: Bob Blakley | February 13, 2009 at 12:50 PM
I've also had reports of errors trying to download the document. If you get an error, you should also feel free to email me (Bob) and I'll send a copy out to you.
Posted by: Bob Blakley | February 13, 2009 at 12:55 PM