Identity and Sociability at the MIT Media Lab

March 4, 2013


The Media Lab Mixer
The Media Lab Mixer project is encouraging and measuring person-to-personal socializing among members of the MIT Media Lab this semester... and the results are both promising and raise a set of new important questions.  
The project is an initiative of Shaun Salzberg of the Changing Places research group and it involves bringing the people of the Media Lab out into shared space together - specifically, into our new community living room and cafe area called "the Atrium".  We use MIT Media Lab cards (like the MBTA "Charlie Card" or any building access or identity card) to identify ourselves and whoever else we are sitting with in our little 10-forward recreation and multipurpose space.  Basically, we just put our card onto a little box in the middle of the tables that are spread around the Atrium while we are hanging out and we take the cards with us when we leave.  There is a game overlay to the experience, with "winners" announced each week reflecting people who spent the most time with others in the community space during the prior seven days.  And there are more points for hanging out with additional people and even more for being with other people that you had not spent time with in the past (an easy formula to execute by the program calculating "time spent").  

Is it Working?
My office mate started spending more time in the Atrium the week before last as part of "That Game in the Atrium" (it's other name) and told me about it.  So, last week I also joined the fun.  I can report the results are impressive: I was able to get work done in the sitting areas (hey, there's fast wifi, a full-boat coffee set up and plenty of good natural light) and also met several people I'd seen over the years but never really spoken to before!  A couple of new collaborations were even hatched, as well as a measurable bump UP in the quality of my incoming Facebook activity stream.  In short - this use of identity technology is showing early signs of value, fun and new types of sociometric team and group cohesion or performance measures.  

From the Lab to the Marketplace?
I have suggest the Cambridge Innovation Center consider exploring whether the project's gear would be a good fit for our building.  I've had an office at CIC for over a year and appreciate the effort made by the staff to create attractive and usable open spaces for tenants.  There is a plan for a meeting to discuss the idea soon and look at how this project could be leveraged for thriving community of entrepreneurs and innovators.  Community and social events like the Venture Cafe and receptions, speeches and training sessions bring CIC tenants together to some extent and there is broad interest to find more ways to encourage cross-community socializing and hence connections, networking and more value.  In CIC, we already have powerfully implemented building access cards that double as MBTA Charlie cards, Venture Cafe membership cards and other applications and uses as well.  Using our identity and community cards to also measure time spent together would be a natural fit.  

Dimensions of "Social"
What if the Media Lab Mixer program measured more than one thing?  Is it possible to add some meta-measures that shed light on other dimensions of the sociability being promoted?  One idea is to track the aggregate time spent in the Atrium by members of MIT Media Lab research groups.  This would both show the measures of research groups directly and also likely have the secondary effect of stimulating more group cohesion and teamwork.  Similarly, it would be possible to measure - and perhaps reward through the gamified point system - the time spent together by people from different disciplines or people working on complementary aspects of different projects.  The Media Lab is inter-disciplinary at it's core and these types of overlay measures might support and reflect some aspects of what makes the lab so special and effective at generating creative ideas and novel approaches.  In the context of potential applications in workplace environments like CIC, potentially the additional dimensions of measurement could include whether members are spending time with people from other companies or whether people involved in creating new products and services are spending time with people who purchase, extend or fund such endeavors.  

Areas for More Research and Development
This past week I did some deeper testing and exploration of how the Media Lab Mixer code and system work, both out of pure interest and also to prepare for a higher value discussion with CIC and others regarding potential fits of this approach in workplace environments.  One thing I learned: it's important to deploy systems like this with some security and controls.  I discovered that it's simple for Media Lab members to print multiple cards that can be used at one time (boosting points) and to leave card on, below or near the readers for long periods (even while the card-holder is back in their office) and to tactically seek out tables with many new people to turbo-charge point accumulation, among many other techniques.  However, even after subtracting points generated as part of testing and evaluation, I was still astonished to see that my time in the Atrium was of long and frequent duration and this pointed to a need for other enhancements as well as simple security and anti-gaming tweaks.  Supporting OAuth2 based integration with social and productivity apps (like Facebook, twitter and Linkein or Evernote, Hangout and your contacts) would further the depth and breadth of value.  By supporting, simplifying and strengthening the social outcomes enabled by a "mixer" experience, the entire life-cycle and impact of the experience can be amplified or transformed into something of a new and significant value.  For instance, what if the identity data associated with people around your table - by permission - could be easily connected to making status updates based on cool new things discussed and learned, adding friends and contact easily, accepting invitations to events or catching links and attachments in a single stroke, and so on?  The code to allow these types of integrations is not especially complex thanks to widely deployed standards such as OAuth2 and soon OpenID Connect.  Which types of integrations to enable, what data to collect and how interactions are best stimulated are among the chief open research questions to be explored.  

Next Steps
I'm getting together with Shaun later together to make a short "demo" video of his project.  Be on the look out for that posting here at and learn more by watching the Changing Places group at the MIT Media Lab.