Desert Island Research Groups

This is probably too arcane to be an official Dorky Poll, but I thought of it yesterday, and sort of like the idea. In pop-culture circles, it's very common to find people making list of "Desert Island Books" or CD's or DVD's, or whatever. The idea being to list those pop culture items that are of sufficient quality that they would be sufficient entertainment if you were stranded for years on a desert island.with nothing else to read/ listen to/ watch.

So here's an extremely nerdy variant of the "Desert Island" idea: imagine that you're being exiled to a remote island or a a space colony or some such, and for contrived reasons will be allowed to read new papers by only one research group: what group would you pick?

(This may be more strongly biased towards experimental work than most, which is part of the reason why it's not a full "Droky Poll" question...)

My answer:

This boils down to "What is the most essential research group in your field?" The selection criteria really ought to involve a combination of quality and quantity-- you want to read papers by people who do really good work, but you also want people who are really productive. A group that produces one absolutely brilliant paper every ten years isn't as good for these purposes as a group that produces three merely excellent papers per year (at least, that's how it seems to me).

In the cold atom/ BEC field, there are really three contenders:

• The Phillips group at NIST in Gaithersburg. This would be the obvious sentimental favorite, given that I did my graduate work with them, but they're unquestionably one of the top research groups in the field, and have a range of different projects going on, from molecular spectroscopy to quantum computing, and even a bit of biophysics here and there. They don't publish as much as they might, owing to the slightly burdensome government approval process, but the papers that come out of that group are always well written, and always on the cutting edge.
• The Cornell group at JILA in Boulder. Another NIST enterprise, and another very large and diverse research group. They've got a little bit of everything going on, from studies of BEC properties to Casimir force measurements, to some precision symmetry tests. Again, they're not incredibly prolific, but when you see a new paper from them, you know that it will be good stuff.
• The Ketterle group at MIT. On some level, these guys are just frightening-- they produce something like 3-4 exciting results per year, like clockwork. They publish enough that they ought to offer an RSS feed. They've got a somewhat tighter focus than the other two-- all BEC, all the time-- but if you want to know what's hot in the BEC community, this is the place to look. Odds are, one of their experiments is in the top two or three of the past year.

If you put a gun to my head, and told me to choose just one, I'd probably let you shoot me. If that weren't an option, I'd go with the Phillips group, with sentiment tipping the balance. But it'd be a close call.