Daily Life in an Ivory Basement

a.k.a. "why we should care:"

From the Comp Sci dept WIMS mentor:

• "In terms of design teams and designing products, there's evidence from other industries that if you have just a male team, you could have a flawed product. Let's look at air bags, for example. Only 8 percent of mechanical engineers are female, and most of the teams working on air bags were predominantly male. When air bags were invented based on the male body as the norm, they ended up being potentially deadly to women and children. That's also happened with heart valves and voice-recognition systems; they were geared toward the male." (J. Margolis, in an interview at http://news.cnet.com/2008-1082-833090.html) Of course, a similar concern holds for teams with no minority representation.
• Is there something in the genetic makeup of women and minorities that explains the lack of interest? Not likely. This was the argument that used to be made for why so few chose science, medicine, law, etc. decades ago. But these fields have made an effort to address some of the barriers to women and minorities and, as a result, their percentages have increased.

Separately, from Gizmodo via the Interesting People mailing list:

This is awkward. It appears that HP's new webcams, which have facial-
tracking software, can't recognize black faces, as evidenced in the
above video. HP has responded:

We are working with our partners to learn more. The technology we
use is built on standard algorithms that measure the difference in
intensity of contrast between the eyes and the upper cheek and nose. We
believe that the camera might have difficulty "seeing" contrast in
conditions where there is insufficient foreground lighting.

HP Face-Tracking Webcams Don't Recognize Black People - Hp - Gizmodo (21 December 2009)

What would you want to bet that that development team had only white people on it?

--titus

(Some more meanderings on the brouhaha about diversity in the Python world.)

First, I've removed 'python' from the tags and made sure that neither Planet Python nor Advogato feed from this blog otherwise; I suspect by talking about politics and feelings in OSS I'm getting further from my normal target audience, and I don't want to spam the entire community.

Second, because my last two replies directly into a diversity-list-inspired discussion have inspired blowback of the form "why are you such a jerk for not supporting women/minorities in Python?", I want to CMA and point out that I'm completely in favor of increasing the diversity of the Python community, and I'd like to publish a community-wide diversity statement. Heck, I could probably convince you that I've worked reasonably hard over the last three years to bring women into the community. Unfortunately, expressing skepticism about statements like this: "...publishing a diversity statement is one of the only things that works" apparently means "I don't like women in my programming community" to some people, even though that's not what I meant by it at all. (If you're interested, what I meant was that Kirrily Robert's post/talk about how AO3 and Dreamwidth created communities with a more representative balance of men and women by, among other things, posting a diversity statement, DID NOT AUTOMATICALLY IMPLY that such a tactic would do much in the long-established Python community -- even though that was what some people were implying, and what GvR actually stated in an e-mail. See how controversial I am?)

I've also been astonished to watch otherwise sane and reasonable people embroil themselves in this discussion with good intentions but ... poor implementation of those intentions. When one person (who I know somewhat and respect as far as I know them) said that the PSF's discussion of the diversity statement "violated" their expectations, I had to throw up my hands and laugh somewhat ruefully; was that really the least inflammatory choice of words that that person could have used?? (There were many more examples, but that one stuck out in my head as possible the least appropriate word choice I'd seen that day.) The discussions seemed to often boil down to people questioning the precise meaning of specific words, frequently by choosing the most narrow and least positive meaning possible. That is not a productive way to discuss something, folks.

Perhaps the most disappointing aspect of this whole discussion has been the disappearance of the normal assumption between long-term Python contributors that we all mean well to each other.

Perhaps because of this, I've gotten a number of private e-mails from people in the Python community expressing feelings of isolation from the community on this topic, either because they don't feel strongly about the topic or they feel their opinions are not welcome. Given what happened to me after expressing an honest inquisitive opinion about why women didn't feel comfortable coming forward in the Python community, I can hardly blame them; the Thought Police were out in force, and there's no winning when someone implies that you either agree with them completely or you're a bigoted misogynist asshole. (That would be me, incidentally!)

So why do I keep thinking about this so much, anyway? (To give you an idea of how much, I have two other posts in the pipeline that address this issue and I've read more e-mail and had more conversations about this in the last two weeks than any other topic... I'm debating deleting the posts, you'll be happy to know ;) Partly it's because of my general surprise at the inflammatory nature of this discussion in the generally fairly friendly Python community, and partly it's because I liked & respected the guy who (in paraphrase) called me a bigoted misogynist asshole and I still very much respect the guy (GvR) who called me a wanker and asked if I'd actually read Kirrily Robert's post (I had, and I stand by my point of logic). I don't like losing potential friends for stupid reasons, and I think you can probably understand my feelings about the author of the wanker comment...

Today while driving I thought about this topic some more, and settled on something specific that I actually think is worth addressing: namely, could this whole topic have worked its way through the Python community with fewer misunderstandings and in a less inflammatory way? There are a couple of things I would suggest doing next time, if there is a next time for a topic of a similarly divisive nature.

Zeroth, I don't think there's any way the diversity list could have been made more pleasant for me. By nature I'm very much a skeptic, and that was a poor fit for the list. (I'm glad I joined, though; the links and discussions that I read while on that list were indeed eye-opening.) What, then, would I have changed?

First, I would have communicated the diversity statement to the PSF list with a request for a simple up-down vote by the membership, along with an additional request that anyone who thought that it was worth discussing in detail move over to the diversity list itself. There was quite a lot (QUITE a lot) of duplicated discussion on the PSF list.

Second, I would have tried to keep the diversity list better informed about the depth of the discussion going on in the PSF list. One problem with mailing lists is that you tend not to realize that everyone else isn't on them; with closed mailing lists, this is even more of a problem! Perhaps because of this at least a few people seemed to think that the PSF members had read the diversity statement generated by the diversity list and collectively shitcanned it without much discussion. Not so, my friends - we had lots of discussions! I can give you screenshots of my mailbox if you doubt me ;)

A third thing I might have tried would have been to gather more background material than a few blog posts for people who wanted to read more. (This is definitely 20-20 hindsight; I didn't realize how much material was easily accessible on things like increasing diversity in the sciences! That's one of the blog posts I have queued up...) Blog posts are almost necessarily advocating something, and to do so they take a biased perspective on that topic. That bias can in turn bias conversations. I'm reasonably certain that the diversity list is ignoring possible options because of the set of background reading material that was sent to the list. Watch This Space?

Oh, and fourth, I probably would have tried to make a stronger effort to make the point that a diversity statement is almost certainly not going to matter in the slightest without substantial backup efforts. (This point was made by several people, but didn't seem to percolate effectively in the face of disagreements about 'the' vs 'a'.) I will post a viciously disappointed post about this if I don't get good turnout for GHOP.

This is all debatable and in hindsight, and I certainly wouldn't have thought that the topic would blow up to this extent in advance.

I will always have more to say on this topic but I think I'd better cut my losses while only half the community thinks I'm a bigoted misogynist asshole wanker... ;)

--titus

p.s. Comments welcome from the Python community, except from that one guy.

p.p.s. Does anyone else on either list find it strange that so many messages were exchanged on both lists without much comment in the Python blogosphere?

Since a few people have asked, here's a rough guide to the diversity discussion. No specifics allowed.

1. diversity list created to (among other things) ponder an official diversity statement for Python. List is closed-archive but open for general subscription.

2. Various diversity list discussions become heated. Some people (including myself) leave list in response. Sigh of relief, back to normal life; is that a good response?

3. A few weeks pass.

4. Diversity list discussion hunts me down on psf-members and tries to pounce. Narrow escape.

5. Proposed diversity statement from diversity list posted to psf-members for discussion and hopefully? approval; diversity discussion engulfs psf-members list like a revenant whale.

6. 1000s of messages pass. Or at least many dozen. People agree, disagree, agree to disagree, disagree on their agreement, and otherwise cause trouble by collectively failing to accept any part of the proposed diversity statement. (Tho it's actually much more complicated than that.) Troubling and unprovable accusations of widespread anti-diverseness in the Python community are softly bowled across the lawn.

7. Diversity discussion from psf-members cross-posted to diversity list. Non-PSF members on diversity list freak out at the idea that the PSF might adopt a diversity statement that did not take into account some of the issues they had discussed. Hurt feelings ensue, including frustration by various people that other people are doing things they don't want them to do, in complete violation of expectations. Troubling and unprovable accusations of fairly specific anti-diverseness in the Python community are left, steaming gently, on the lawn. Closed nature of both lists engenders and amplifies confusion.

8. Still no diversity statement from the PSF on the horizon.

Things have quieted down for the evening.

Personally, it's been the most unpleasant set of interactions to watch and (occasionally) participate in that I've seen in the Python community in a long time; one can only hope that we reach some form of passionate agreement in the future:

Agreement in a group setting is truly a wonderful thing. But we should be wary of agreement that comes without any work, any disagreement, and disruption. We must never mistake quiet civility for passionate agreement.

(See this link for the whole post from which that quote is taken; Godwinning is unintentional but, frankly, a rather ironic endpoint to my meanderings.)

My new theory? It's all a plot instigated by the Perl community to distract the Python community so that Perl 7 can get the jump on Python 4k. It's the only way I can make sense of it all.

--titus

Comments closed, because I just don't care what anybody thinks any more.

Sarah Mei posts about teaching Ruby to high school girls. Good stuff.

While searching for some GHOP info from way back, I ran across this post asking "where are the girls among the GHOP winners?" (The statistics mentioned in the post may have been posted since, although I haven't seen them.) We asked the Python mentors to "rate" the students, and the hands-down winner was someone who had worked closely with several different mentors and performed very well. Perhaps next time we should highlight everyone who did well; there were several women in the group, too.

In general it's tough to raise the visibility of minority groups, though. Do we engage in affirmative action of some sort, and if so, how do we do so without being unfair to others? Or do we simply rank people in a presumably gender- and color-blind way and see what happens? I've talked several times in various venues about trying to run a female-oriented GSoC derivative, like GNOME's Women's Summer Outreach Program, which would at least call attention to one minority in OSS... and if GHOP ever happens again, we could work on getting younger women involved.

--titus

After being on the new Python diversity mailing list for a bit, I've just unsubscribed. While there was an unpleasant personal incident that catalyzed my decision, I also don't think I'm a good fit for the style of discussion taking place. (YMMV ;)

That having been said, I want to give my own little diversity statement:

I'm interested in improving the openness and accessibility of the Python community, and making new people as welcome as possible.

I will not tolerate intolerance.

I would like to better understand why minorities are underrepresented in computer science and programming in general, and even more so in open source communities; most specifically I'm interested in understanding what the barriers are and how we might lower them, and also how we might encourage more people to "test the waters", so to speak.

A willingness to work hard and a desire to participate should be all that's necessary to join and participate in the Python community.

I have been trying to introduce more women to open source, but I'm already dealing with a pretty significant underrepresentation in my undergrad classes: about 10% of my 3rd-year class was female. (Almost all of the rest were white males -- I teach in the midwest!) The diversity list crystallized the already present notion in my mind that we have to "look" younger; I'm not sure how to do that. I helped teach an 8th grade science & math computer lab that was much more even, but they may have been hand-selected to be even. It's an interesting challenge and, nicely enough, education is even part (tho a tangential part) of my job description.

--titus

p.s. Comments are off, sorry. Drop me a note if you write a blog post and want me to reference it.