X
Business

Learning to share: lessons from the Data Sharing Summit

This is a guest post by Ben Werdmuller who is the CTO of Curverider and one of the leads behind the Elgg open source social networking framework. This is his first report from the Data Sharing Summit in Richmond, CA, which aimed to address the issue of interoperability between social networks.
Written by Steve O'Hear, Contributor

This is a guest post by Ben Werdmuller who is the CTO of Curverider and one of the leads behind the Elgg open source social networking framework. This is his first report from the Data Sharing Summit in Richmond, CA, which aimed to address the issue of interoperability between social networks.

Marc Canter, the outspoken founder of social networking firm Broadband Mechanics, organised the Data Sharing Summit in part as a kick in the pants to services like Facebook which maintain silo-like control of the data their users generate, refusing to allow them to move it elsewhere. A noble cause, but nobody was surprised when they didn't show up; nor did Myspace or Ning, which Marc lambasted for not being open enough in July.

Microsoft, Google, AOL and Yahoo all made the trek out to Richmond, and were quick to point out the flaws in the Bill of Rights for Users of the Social Web - a document Canter admitted was naive and idealistic. In particular, international law issues loomed large: in some cases, it's legally difficult to share data from one country with a system from another country, as users have different privacy rights in each location. Solving the real-world problems imposed by governments and different legal systems is going to be key if the web is going to be anything other than American.

Microsoft and AOL in particular were careful to point out that becoming more open is a key part of their strategies. All four have open APIs that can be used today, and there were hints that Yahoo will pick up the pace within Jerry Yang's 100 day plan. That said, Yahoo's Randy Farmer suggested we relabel the issue data control rather than ownership, because the latter raises issues with their legal team.

Most services seemed to want social networks to open their data so that their networks could be automatically spidered. It's easy to see why this would be desirable to entrepreneurs, but social networks are extremely unlikely to give up the details of who connects to who on their systems; it's their most valuable asset. However, these issues diminish when a single user explicitly asks for their data to be moved from one place to another. Both users and service providers need to be convinced about the usefulness of interoperability, and allowing users to easily move their profiles between the different services they use is one way this could be done.

How data sharing will actually be achieved was left relatively untackled. David Recordon, of SixApart, is making an attempt to develop real-world best practices, alongside Crowdvine, Pownce, Broadband Mechanics, Curverider, and others. He made the point that there is probably never going to be one single technical standard for data sharing; as a result, services that bridge from one format to another are going to become more and more important. Cloudtripper gave a demo of their software, which is going to perform those kinds of translations.

Overall, no hard conclusions were reached, and it's even doubtful that all of the problems were identified. Some people will be disappointed in that outcome, but the summit was a landmark: the first time the centralised web service community has got together and seriously discussed opening up to each other. It's clear that this topic isn't going to go away, and another event is being organised in London; the conversation looks to continue for some time to come. In that, it was a success, and the first step down the road towards a richer, more user-centric web.

Editorial standards