Category Archives: Ucsf

Opening social networks/ graphs up to Researcher Collaboration Tools, a UCSF Harvard Profiles collaboration project

For the past few months I have been consulting part time with UCSF and the department of Clinical and Translational Sciences. You might think that has something to do with my browser plugin Babelfin, but Translational Sciences really has nothing to do with language learning, but rather taking exciting patterns in other fields and translating the processes between them. In this case UCSF is focusing on how Social Media and Social Networking can be used in an academic sense for collaboration and messaging rather than games, photo sharing, or virtual resumes.

The UCSF OpenSocial project (–shindig-apps/ ) started as a Harvard project called Catalyst PROFILES ( ). Profiles (as we call it), is a simple social networking server that manages the graph relationships between colleagues, co-authors, and research interests. Profiles looks at relationships differently than Facebook, Linkedin, or event MySpace, but it’s pretty bare bones and limited in what it can do.

The innovative part comes in where UCSF thought it would be neat to extend Profiles without altering it’s code. So, Eric Meeks at UCSF bolted on an Opensocial container named Shindig to Harvard’s Profiles project which allows external apps to run on top of Profiles. This makes for an interesting mix of code, as Profiles is a Microsoft C# ASPX project, and Shindig comes in PHP and Java flavors. Eric rightly choose to implement the Java flavor of Shindig as it’s the most current.

So, this is where I come in, as, I am building the applications that run on the Shindig server accessing the Profiles social graph. In many cases it’s just like building an application that runs on Linkedin, Bebo or MySpace, however, there is no friend graph, but, there are 3 other graphs I can use, co-author, colleague, and interest graphs.

Initially we are keeping it simple, but we plan to extend Opensocial in a standard way so that other universities and research institutions can apply Opensocial to their graph servers. UCSF and Harvard are hoping that their work will make it easier to use Profiles as the graph server, but they are both very excited about creating an open platform that can develop a rich ecosystem of applications that extend their work, and are able to run on other platforms with small tweaks. In the end we want researchers to be able to better collaborate using social tools.