This blog has been left in a little bit of a hibernation mode these past few months (well…how to the Eskimos call it, when the duration of nighttime so much overtakes the duration of daytime?). I’ll spare you the usual excuses though — I suppose my mind was elsewhere. Which didn’t prevent me from generating ideas for future blog posts though, and by now I have a huge backlog that I hope to slowly go through. (Of course everyone knows where these kinds of hopes go…)
A couple weeks ago I permanently left my office in MIT’s building 32, the Ray & Maria Stata center (which not only houses CSAIL but also MIT’s linguistics and philosophy departments, including such celebrities as Noam Chomsky). The past couple weeks I was there, but I will soon be over here, before going all the way over there and then spending some time here and here, before finally reaching there…lots of fun traveling in perspective, and hopefully good science as well!
I spent 18 months in Building 32, as a postdoc under the wise supervision of Scott Aaronson. The Stata center is a unique building with a unique history (and unique occupants of course!). In guise of farewell I went around the building and took a few pictures that I’ll share here with some impressions of the building.
Wikipedia has an interesting entry on Building 20, the building that occuppied the same space as the Stata center before it was built in the early 2000s. Building 20 was erected in just a few months in 1943, and meant to house research directly related to the ongoing war effort. Originally designed to be taken down 6 months after the war ended, it continued to be used until 1998…some 53 extra years! Due to its lack of official purpose, the building ended up housing all sorts of small, “speculative” projects. In spite of — thanks to — the difficult conditions, the building had an extremely successful run. The flexibility that comes with the almost total absence of design (and associated rule-keepers) enabled occupants (including, already, Chomsky) to transform their working space at will, bringing a wall down here and erecting another one there. The inscrutable floor plan encouraged unexpected collaboration as occupants from different areas were bound to get lost in the same dark corners. Difficult conditions (heat, leaking windows, etc.) encouraged creativity.
The architects behind Building 32, Frank Gehry and his firm, sought to keep *all* these elements in their design — the flexible spaces, the abtruse floor plan, as well as the leaky finishings. Its current occupants regularly complain about some of the conditions (who has never seen a mice trapped under their desk? Gotten lost? Received drops of water from a leaking window? Gotten lost? Been unable to fit furniture due to the weird angles? Have their lighting/AC/projector shut down automatically because “the building” said so? Gotten lost?), but overall it works remarkably well. The building is very bright, with numerous common spaces that open one onto the other, including one floor up or down. There are no long corridors, but rather small “work areas” the open one onto the other. Multiple office doors open up on any of these areas, encouraging spontaneous discussion.
Possibly the hardest, and most important, aspect to encourage in a building devoted to research is interaction. It should be the case that occupants are literally forced to pass by one another on a daily basis. This can be encouraged by placing strategic items such as the coffee machine, the elevators, or the bathrooms, in central and open locations. Building 32 does this; unfortunately it is possibly a little too big (and tall), so that there is a large proliferation of such areas and in the end smaller communities form around each of them. For instance, the theory group itself lives on two floors, with multiple distinct spaces. The group is large, and the space is needed, but it means that it easily lives onto itself as a small community. During my time there I had very little interaction with other large occupants of the same building — AI, optimization, etc — not to mention the linguists!
Second to interaction, a good research space should encourage creativity. I suspect the elements that work for one person or the other may vary wildly, and are harder to control for. For instance, in my case this includes a lot of light and open views (say, on the mountains), but others may need darkness to concentrate. The ability to focus also necessitates some amount of calm & quiet (though not necessarily perfect), a requirement not always easy to reconcile with interaction. Finally, I personally need what I would call a “diversity of working spaces”. I usually feel that I cannot work for more than a couple hours on the same problem at the same place. So, in addition to an office desk, I may also need a couch, a library, a cafe, a bench in a courtyard, etc.; the more the better!
All in all, building 32 provided me with all this and much more (the one downside being a little more noise than I would have asked for), and it is one of the most original and welcoming spaces I have had the chance to work in. I am very lucky to have many other great working environments lined up for the coming year: the Newton Institute in Cambridge, Berkeley’s Simons Institute‘s brand-new space in Calvin Hall, and finally the beautiful Annenberg Center on the Caltech campus!