[Olpc-open] Women in Open Source projects article

[Chris Ball]

Hi,

   > Free Software Magazine has published an article about the gender
   > gap in open source software development vs. commercial software
   > development and some things that projects can do to be more
   > inviting to female contributors.

   > http://www.freesoftwaremagazine.com/columns/ten_easy_ways_attract_women_your_free_software_project

If I could summarize my response below in one paragraph:  this article
is exactly the problem, because it is promoting a whole series of
stereotypically sexist attitudes about women and their intelligence.
Here are some of the issues I have with it:

   > • Use forums instead of mailing lists

This suggestion doesn't make sense.  I agree with the author (and the
many people who have said this before him) that women are turned off by
an ultra-aggressive alpha geek style of conversation, but the solution
isn't to "use forums", it's to stop using and encouraging these
destructive behaviors.  We might retitle the suggestion to "When
considering flaming someone, don't."  If you need an example of how
moving to a web forum doesn't make the slightest bit of difference to
how respectful a conversation is, consider the average comments in a
Slashdot thread.

A point the author doesn't make, which I find compelling, is that it's
not just women who are turned off by aggressive conversation modes,
it's anyone who doesn't want to take part in the alpha geek mentality.
By making an environment that is more welcoming to women, we make an
environment that is more welcoming to everyone except a loud subset
that we're currently optimizing for.

   > • Avatars create a face-to-face-like feeling that encourages “more
   > human” behavior

Advocating avatars because they're "more human"?  Are you talking about
women or children here?  Women read more books than men¹.  Women are
perfectly capable of and comfortable with engaging in purely written
communication, most likely moreso than men.  This insinuation of a
childish need for needing cartoons to create a "face-to-face-like
feeling" seems extremely insulting.

   > • When possible, wikis instead of version control archives

I'd love it if I could find more solid reasoning behind this than "wikis
are friendly and version control is complicated and women like friendly
things and don't know how to do complicated things so we should use a
wiki for version control even though that doesn't make any sense", but
I'm not seeing it in the article.  Ugh.

   > • When possible, high level languages

Ditto.  For example, the FLOSSPOLS report linked below contains a study
that compares the programming ability of men and women taking college
programming courses.  It finds that women perceived their programming
ability to be far lower than men perceived their ability, yet programming
examinations showed ability levels to be equal between men and women at
the end of the course.  The problem isn't that women aren't smart enough
for low-level languages, it's that we boast about how great we are at
coding so much that we manage to convince women that they must not be
as smart as us.

   > • Replace pecking-orders with affirmation processes (thank you's)

"Women are more likely to want to discuss or seek approval for their
changes, owing largely to confidence issues" is not a respectful thing
to say, and does not treat the women involved as equals.  Who doesn't
have confidence issues when joining a new group of people and submitting
their first proposal or patch?  We'd do well to thank our volunteers
better regardless of their gender.

Oh, wow, I just noticed the photo of the sewing machine:

"Programming, like sewing, is largely a "tacit" skill, which is best
learned by doing and by watching others."

Yes, because a comparison to sewing, which is apparently something the
author thinks women seem to learn how to do *really well*, is totally
appropriate.  Good job.  Full marks.

I think I just ran out of words.  Please don't listen to anything this
document has to say.  It actually uses good sources, even while it
draws insulting conclusions from them, and the sources are worth
reading.  If you had to choose one thing to read about the (very real)
reasons behind there being fewer women in free software, I'd recommend
the FLOSSPOLS report², which is a large-scale academic study of the
reasons women don't contribute more to free software projects, and is
truly enlightening.

So, it's all well and good to complain about the recommendations given,
but what would a replacement set of recommendations be?  As well as the
formal recommendations given in FLOSSPOLS, Val Henson's "HOWTO Encourage
Women in Linux"³ contains a perfectly good set of advice.  Note that one
of the recommendations is (3.13) "Don't assume that all women like
cooking, sewing, and babies".  Note that this article attempts to make
learning how to join a free software project more like learning how to
sew.

- Chris.

¹:  http://www.npr.org/templates/story/story.php?storyId=14175229
²:  http://flosspols.org/deliverables/FLOSSPOLS-D16-Gender_Integrated_Report_of_Findings.pdf
³:  http://tldp.org/HOWTO/Encourage-Women-Linux-HOWTO/
-- 
Chris Ball   <cjb at laptop.org>