The Federated Social Web and Drupal - Notes from the Drupalcon BOF meeting

The meeting started with a series of introductions by participants including the reasons they were interested in a federated social web. The answers were very encouraging because in addition to the expected concerns regarding breaking apart silos of information and control (I am thinking of you facebook.com) there were also some very practical needs regarding interoperability and decentralisation in enterprise environments. This is great because noble thoughts (and efforts) on their own a federated social web will not make. We need practical business uses to justify funding and further development and adoption of the technologies.

With regards to the technologies and were things currently stand we got a first-hand introduction from James Walker (otherwise known as general trouble maker "walkah" in Drupal-land). In summary, the OStatus effort is making good progress and working in collaboration with the key people from each of the constituent protocols to improve the protocols and make them fit together better. The first Federated Social Web Summit showed that there is great interest from many different corners, although they would have liked some more input from the Semantic Web technologies crowd. One of the more tangible items to emerge from that meeting is the rather ominously named SWAT0 test - the Social Web Acid Test. This is an integration use case that involves multiple parties and will make use of all the relevant standards.

So where does Drupal stand in all this? Well, there was a very interesting presentation from Alex Barth of Development Seed on the work they did with the Pubsubhubbub protocol. Although some aspects such as PuSH publisher support are missing there is work on that being done by Palantir. One of the uses of the the Pubsubhubbub protocol that Alex talked about is pushing user accounts and their updates from one Drupal installation to another - this was in response to a very practical need they had from a client. Scenarios like this are really exciting because they provide sensible use cases for the technologies and I hope we will see more uses of the sort coming up.

Now, the challenge ahead for Drupal is moving beyond Pubsubhubub support to the other protocols of the Ostatus family. James showed us a working version of his work for discovery and salmon support, while we discussed the challenges in getting Activity Streams implemented in Drupal (a longer post and more ideas on that should come some time soon).

So to sum up - the good news are that there is a lot of interest in the Drupal community to turn Drupal in a fully-subscribed member of the federated social web community. The not so good news is that a lot of work needs to be done. There is great progress on the Pubsubhubbub front and more should come soon. If you are interested in all of this there should soon be a groups.drupal.org group about it (if the powers that be realise that a federated social web effort is not the same as discussing generic social networking and as such a separated group is required).

If you were at the BOF and want to keep updated drop me an e-mail (ronald at istos dot it) and I will let you know as soon as the group is created or something relevant has come up.