By Michael Olson | Posted on July 26, 2010
Janrain software engineer Ivan Pulleyn attended the Federated Social Web Summit in Portland, Oregon last week. Below is a recap of Ivan’s experience at the event:
Imagine you’re backpacking through the mountains in a distant country. You catch the only bus from a small village that is headed toward the coast. The bus is crowded with locals and their luggage, but you coincidentally find yourself seated next to another expatriate like yourself. She’s traveling in the same direction, but planning to head further South. The two of you hit it off and you soon find yourself lost in conversation. Hours later, you suddenly realize you’re at your stop. The bus driver won’t wait; he has a schedule to keep. You have to get off immediately, but you want to continue the conversation. You have a pen and nothing to write on but your arm. What’s the one piece of information you ask for so you have any hope of getting back in touch?
This was the thought experiment presented by Blaine Cook at the Federated Social Web Summit (FSWS) here in Portland. It’s meant to illustrate a central theme that occupied much of the discussions at the summit: how do people identify each other and share information on the social web? If someone asks me for my contact information, do I give out my cell phone number, my personal email address or my work email address? How about my Twitter handle, or my employer’s URL? Would I give out different information to a friend than to a colleague? Would the information vary if I was overseas rather than home here in Portland? The answers to these questions seem simple on the surface, but once you dig a little deeper, things become more complicated. The social web is imbued with varied and dynamic context and, as such, emerging methodologies, protocols and user experiences must adapt to a new set of challenges.
The FSWS was facilitated by the technical team from Status.net to “to coordinate the development of the federated social web.” It was organized as a one day event preceding OSCON, with invitations going out to a varied group of technical innovators at companies and academic institutions working on relevant protocols or architectures. I was curious to see how this rather broad agenda would manifest itself given the short amount of time we had together. The day opened with Evan Prodromou, Founder of StatusNet, sketching out a plan for the day. Each group gave a short presentation showing off their work in social federation. By lunch time I had seen presentations from StatusNet, Google, DiSo, Appleseed, Facebook, Diaspora, Six Apart, Mozilla, Microsoft, GNU Social, the Personal Data Store Project and @versionvega, among others. I presented a brief history of Janrain’s involvement in OpenID and described some of the challenges we’ve had to overcome in order to support a large array of social media platforms.
The afternoon agenda for the summit was generated democratically, in real-time through the use of Google Moderator. Participants posted ideas for breakout sessions and we voted the topics up or down. Some common themes bubbled to the top and we eventually chose nine different tracks for detailed discussion. My interest in is protocols, so I attended three of the protocol-related tracks. The discussions centered around two major areas of pain: how to do end-user identifier discovery and how to avoid the problem of OAuth pre-registration. Both need to be solved before a truly open, federated social web can emerge.
In the first session, Blaine Cook proposed a method for private PubSubHubbub subscriptions using WebFinger for discovery and server dialbacks to verify the identity of the subscriber that involved an innovative use of the HTTP From: header. He proceeded to walk through an example of how person A could subscribe to person B’s address book in an open, yet verifiable way, with person B being able to approve or deny access asynchronously. Following that presentation, Brett Slatkin of Google proposed a more complicated version of the same use case, this time using OAuth. Martin Atkins, of Six Apart, presented a third protocol idea, this time where the end-user approves a domain for access rather than another user. This was followed by a presentation on distributed comments using Salmon Protocol by Joseph Smarr of Google. Other participants included Maxwell Salzberg of Diaspora, David Recordon of Facebook, Rob Dolin of Miscrosoft and Bear of XMPP fame.
At the close of the day, we reconvened and Evan proposed a challenge to the group. Dave Recordon had recommended a simple social federation use case and we decided to call it the Social Web Acid Test, level 0 — SWAT0 for short. The challenge is “to have as many discrete pairs of implementations pass the test by September 30, 2010.” SWAT0 is a simple use case that has been implemented many times by proprietary social networks: two users sharing and tagging photos. The real challenge is to build this in a truly open, federated manner. I strongly encourage those with interest in the technical aspects of the social web participate in this effort by joining the mailing list or even building independent implementations. As FSWS closed, we enjoyed a demo of Diaspora in action, followed by drinks and conversation at a local venue. I left feeling inspired to participate in this emergent collaboration.
Now back to our thought experiment: As you watch the bus pull away, a cloud of dust settles onto the dirt road. The sun is beating down on you as you struggle with your overstuffed backpack. You look down at your arm. What’s written there?
How to tell if your identity management is ready for the new data protection regulations…
We just released the latest member of the Janrain product family: Janrain Advanced Policy Manager…
Janrain Information Security Manager, Lisa Nicholson, shares her thoughts on why CSA Level 2 and…