composition.al

Upsides and downsides of "women in..." lists

A couple months ago, Jean Yang started a crowdsourced list of women doing programming languages and software engineering research. I added my name and advertised the list in a couple of communities I’m part of, mentioning that although I think there are upsides and downsides to such lists, on balance I think they’re good. A friend asked me to elaborate on what those upsides and downsides are.

Here are what I see as the upsides of lists like this:

  • They help people who are putting together a program committee, keynote speaker roster, or the like, but who are having trouble thinking of women to ask. (This was the stated purpose of Jean’s list.) Often, the same small number of highly visible women tend to get asked to serve on program committees again and again, which does a disservice to those women by overburdening them with work, while other qualified women never come up for consideration. So, lists like this are especially useful for getting the names of less-well-known women in front of decision makers, so that those names will be more likely to come to mind when it’s time to choose committee members or speakers.

  • It’s helpful for women in the research community who feel lonely or isolated to be able to look at a list of names and see that there are a lot of other women out there in the community who they might not have met yet. I’m excited when I look at a list of women in my field and see names I don’t recognize.

  • Some people have a habit of always citing the same handful of very famous names whenever they’re asked for an example of a woman working in a given field. (In PL and software engineering, those names include Grace Hopper, Barbara Liskov, and Margaret Hamilton.) When people unfamiliar with the area only hear the same famous names over and over, it can give them the false impression that those are the only women who have ever worked in the area. The existence of a list of not-necessarily-superstar women in a field as a reference can help ameliorate that problem.

The downsides I’m thinking of are more subtle than “lists give would-be harassers a list of targets”, although that’s a potential downside, too. The less well-known downsides include the following:

  • Lists like this tend to spread around via certain social media platforms, and therefore they might only be seen by a small subset of the community they’re supposed to serve — the subset that happens to be active on those platforms. So, they may end up merely helping the women who are already visible become more visible (perhaps even more than they’d like), while not doing much to help those who could use the visibility more.

  • In my experience, more visible senior and mid-career women tend not to add themselves to such lists, for a variety of reasons: they’re busy with more important things; they don’t have anything to gain personally by being on the list (they’re already getting all the PC and speaking invitations that they want); they think the list isn’t intended for them; or some other reason. Then, someone who doesn’t already know that those senior and mid-career women exist may look at the list, see that it’s mostly names of junior people, and mistakenly conclude that there are no senior or mid-career women in the community.

  • Related to the above points, the list may just fail to achieve any kind of traction in the community. Then, people who aren’t already part of the community may see how short the list is and think, “Wow, there are that few women? How disappointing — maybe this community has even fewer women than I’d thought!”, when in fact the list undercounts the women who are there. In that sense, a short list may be even worse than no list.

All of these downsides are mitigated by publicizing the list widely and getting it to grow. How big should we expect such a list to get? For comparison, there’s a public opt-in list of women active in machine learning with over a thousand names on it. Admittedly, that list isn’t specific to researchers, and even if it were, the ML research community is bigger than PL/SE research. Nevertheless, I feel like we easily ought to be able to accumulate, say, a hundred names of women active in PL/SE research. (There are more than that, but I’d be happy to see a hundred names as a start.)

I’d particularly love to see more senior women add their names to the list. I’ve noticed a phenomenon where someone who’s outside the research community, if asked to name women PL researchers, will list people in two categories: (a) the really famous, really senior superstars, who they know because they’re famous, and (b) early-career women, who they know because we hang out on Twitter or on the non-academic speaking circuit. A third category — mid-career women — tend to go unnoticed, and that irritates me. I believe that research-community-created lists like this have the potential to help fix that problem, but they also have the potential to exacerbate it if mid-career women’s names never get added. It’s my hope that by advertising the list in places that aren’t Twitter, I can prevent that.

Comments