October 06, 2009, 7:21 AM — Unlocking the Clubhouse (UTC) doesn't perform as advertised for me.
This has been hard for me to write. Several colleagues I respect urged me to pick it up, in words that approximated, "just read this; then you'll understand." Certainly the book has collected laudatory reviews, many echoing Scott Carlson's characterization of it as "highly influential and celebrated". The promised pay-off, as I understood it, was that I would better grasp:
- that women are grossly underrepresented in computing;
- that boosting women's participation in computing is desirable; and
- that the trajectory toward higher participation involves specific 'inclusive' behaviors that I can adopt and promote.
That's not the book I read.
Briefly, UTC is an adequate ethnologic popularization. Other claims made on its behalf are largely misplaced. It sketches an argument that computing will be a richer field as it benefits from the perspective of "outsiders", but I couldn't find a single paragraph that analyzed in even a preliminary way what difference it makes whether those "outsiders" are women, African-Americans, reserved foreign students, or even traditionally nerdish US "white boys" who are focused on applications.
There's plenty more that UTC does not do. It doesn't fully document computing's demographic uniqueness: while participation by women in the US in other professions, including law, medicine, pharmacy, chemistry, ... has definitely trended higher over the last fifty years, computing and very few other fields have seen nearly negligible gains over the same span. It introduces what strike me as its two most interesting ideas, but barely elaborates them:
- "outsiders" such as women, students of "color", and perhaps foreign-born students do comparatively worse in "unsupported" courses, that is, those with large classes and indifferent teachers. While the particular data UTC provides don't convince me, it sure sounds as though this area merits more investigation. In the meantime, it's fair to hypothesize that "vulnerable" students and co-workers benefit differentially from social reinforcement. As platitudinous as it sounds, I hadn't come across any meaningful analytic support for it before.
- UTC echoes the observation others have made that there are different cultural traditions for organization of educationally-pertinent interactions. Crudely, Asian students socialize and study with the same people, while African-Americans (but not foreign-born black students) keep the two realms separate. The former pattern is associated with significantly higher academic achievement.
UTC also doesn't teach parents how to raise daughters better, or how to organize an open-source project more wisely. Its anecdotes and narratives can be instructive, but they remained isolated observations that don't constitute a theory or plan on their own.
Carnegie-Mellon, the locale for UTC, has taken effective and publicized steps to boost women's participation in computing--and, in the process, as Carlson mentions, contradicted UTC in several regards. Lenore Smith, a professor at CMU, remarks, "To say that there are intrinsic male and female differences and you have to accommodate for those is absolutely wrong". As a contrapositive to what I wrote last week, gender needn't determine profession.
What's the conclusion? I'd love help from readers identifying what I missed in UTC. In the meantime, I'll let ride my instinct that it's a good thing to have co-workers who think differently about problems than I do. There is indeed, as Miguel de Icaza wrote recently, "a world of possibility", and plenty of good software still to write and research.
