An acquaintance from CopyNight, Dennis McDonald, shared this post on his blog, some thoughts on a subject I have been giving some consideration, too. He points out that for albeit different purposes, there has been some work on interchange formats around relationship data. But a library of over 100 interdependent XML schemas? I think we can do better for purely social applications
Of his five points, I don’t have as much concern over the first three. I think he is envisioning a more comprehensive description of relationship than I think is practically necessary. At its core, I think any exchange or descriptive format needs to speak to privacy and maybe trust. As long as you are not using these for medical or financial applications, I think you can steer clear of regulatory messes. Or better yet, ignore them altogether as they are irrelevant to standards and have more to do with implementations.
The last two are the really sticky points, though. Right now, social applications can utilize the richness of their profile and relationship data and management as a differentiator. Push that down into a common interchange layer and you will no doubt meet resistance. I don’t see it as much on his final point, as they differentiators in a custom application are what you do with that data. In both cases, there is nothing to stop decoration or augmentation of standard data. Doing so would probably speak better to his concerns about the full semantics of any particular application of this data.
Dennis admittedly has a more organizational bias but these are some good points to consider when thinking about portability of profiles and relationships and what that may do to transform, for the better or worse, of existing social applications.
Technorati Tags: social networks