Beki Grinter

Lucky Physicists and the Devolution of Computer Science

In academia, computer science, discipline on May 26, 2009 at 9:43 pm

There are two types of books one can read about academia.  One genre is the academic “how to” book.  I’ve just written about an example in that genre that I found particularly insightful.  The second is a genre that studies academia itself.  It’s in this second genre that I conclude that Physicists are lucky people indeed.

Many years ago I read Sharon Traweek’s Beam Times and Life Times.  It certainly explained the person I was dating—a High Energy Particle Physicist. But, to the book. The book was a multi-sited ethnography of the way that Physicists work. The division of work differs from computing, and those differences emerge at the graduate school level. Working in teams is the simplest way to describe it, hierarchical teams composed of theorists and experimentalists. Traweek’s book did a marvelous job of explaining how these teams produce the science of physics.

And yesterday I started reading another book about Physicists. This is a longitudinal ethnography, following 55 Physicists over some years, understanding how their career evolves over time and as they climb through the ranks of Assistant, Associate and finally Full Professor. I’ve only just started it, but already I’m finding it quite instructive in thinking how an academic career evolves, and what that might me for me.  (It also makes me think that I should assign the two books as an example of how ethnography is not just one thing, but a methodology that consists of numerous approaches to data collection and analysis. If you add in Latour and Woolgar’s Laboratory Life, you have a third approach to studying scientists and the doing of science).

Computer Scientists by contrast have been woefully neglected in study. There are some books about engineering in practice (Kunda, Bucciarelli for example). But not in the academic context. And this is a shame, because Computer Science is in an interesting transitional period, and one with particular implications for the organisation of the academy. (It almost certainly has other implications but lets begin here for now).

Computer Science is in a period of devolution. Devolution into entities that will be the intellectual home for an increasingly divergent number of research problems and pursuits that go under the banner of Computer Science. One can view this devolution as entirely pragmatic. The field is still fairly young and has enjoyed explosive growth and rapid change (both can be attributed to not just the increased presence of computing technologies, but also the radical transformations that they’ve undergone in terms of size, power, form factor, that fuels a series of alternate possibilities, there are certainly other reasons, these just strike me as some obvious ones). In response to this explosive growth and rapid change, so Computer Science has expanded to accommodate the new research problems and domains made possible. And, here’s the really pragmatic part, the organization of Computer Science as a discipline is now stretched, well it seems to be. Perhaps it’s because no-one wants to have faculty meetings with 100 people that you see an increasing devolution of previously single, monolithic Computer Science entities (such as the College of Computing at Georgia Tech, or the Information and Computer Sciences School at Irvine into a series of smaller organizational entities). (There’s more to say about Computer Science organized inside Engineering and the rise of the ISchool movement, and perhaps I will at some future date).

These smaller entities interest me greatly. Again comparing GT and UCI, we see three different sub structures: Computer Science, Informatics, and Statistics at Irvine, while Georgia Tech has Computational Sciences and Engineering, Computer Science, and Interactive Computing. Do the names matter. Yes, I think so. For many other departments these names are the names of their disciplines (such as finding Physics, Chemistry, Civil Engineering departments that characterise what goes on in them and with what disciplinary entity that research work is associated).

So, lets take a case of a particular speciality in Computer Science, Software Engineering. At GT you’ll find it in Computer Science, but at Irvine it shows up in Informatics. What implications does that have for the practice of Software Engineering? Right now, I don’t know. But will it, that’s what I wonder. When will what we have become institutionally become what we do, our community of practice professionally.  Then there’s a question of contested turf. The name Computer Science is not without problems, particularly when applied to one piece of the organization. Is the implication that what is not within Computer Science is not Computer Science research?  If my boss read my blog, I think he’d groan about this particular comment, one way I have experienced this is rather personally (yes, I certainly do research that’s unusual in Computer Science, I thought that that’s what taking risk meant, but I have always taken the risk identifying myself and my purpose as being invested in the future of Computer Science research).

But there’s far more on the table here than my own ego (big as that is). I think what might be on the table is institutional legitimacy and disciplinary organization and discourse. If an organization is structured to make something possible, it simultaneously hinders other things. What is organizationally near (i.e., in the same organization, sharing a common reporting structure, the more the better) is far easier to accomplish than what requires crossing organizational divides. So, we can view the organization of Computer Science as a set of bets, bets about what will happen if we put some things in the same place and separate others.  I wonder how to place those bets, but that’s because I find organizations absolutely fascinating. At the same time there’s an undercurrent—a suggestion—of a separation that might result in a smaller discipline of Computer Science, and something larger that encompasses the full span of activities that formerly used to be Computer Science (and a question of what that larger entity is, what it’s project is, what space does it occupy at the national level). Transition can be difficult, graduate school is a preparation for a career of identity management and assessment (my vita speaks volumes about Computer Science research, and says nothing about Physics research, not just in content, but also in the mechanisms used to produce that content like whether I choose to write books, journals or conference papers). The Institutional legitimacy I think turns on this somewhat, that identity is not just what we do, but what is given back to us by our ability to affiliate with a particular research discipline.

So, I wish someone had studied us. I wish I could pick up a multi-sited ethnography, preferably also longitudinal, to get some ideas about what if anything my colleagues are doing about the devolution of Computer Science. How they handle identity and legitimacy. What organizations are being created and how that gives rise to new areas of research, and what it might potentially close off. Lucky Physicists as their field evolves they have a series of guides to stimulate and frame questions. But, perhaps we are lucky enough even despite not having these studies because what we have an abundance of right now is opportunity. Opportunity to rethink it all, if we are quick, thoughtful, and open to devolution.

  1. Have you seen “Studying Those Who Study Us: An Anthropologist in the World of Artificial Intelligence”? It’s not a million miles off Beam Times & Life Times for CS. (What’s your physicists book, btw? Sounds interesting.)

    Other than that, the closest stuff I’ve seen falls into two camps. One of them is originally informal and then increasingly formalized advice that’s CS-specific: Phil Agre’s old stuff on how to do being a grad student/faculty member/etc is probably the most visible example of this. The other set of work is work on women in computer science: Ellen Spertus’s discussion of when she was transitioning from being a grad student to a professor at (I think) Mills, for example. It’s one of the few places that you regularly find the sort of self-critical assessment as a field that makes for really interesting reading.

  2. Have you read Tony Becher on “Academic Tribes and Territories”? You’d find it relevant.

  3. Hi there –

    Margolis and Fisher’s Unlocking the Clubhouse also has pointers to CS culture, though from an undergrad perspective. I also have a paper I wrote based on my undergrad thesis specifically about undergrads entering a phase where students got grouped as hardcore, talented, and regular:

    Lucy Suchman’s papers over the years also add up to ethnographic notes on academic CS at Xerox PARC. I believe she is working on something more book length on PARC,but I haven’t heard about that project for two years.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: