Melissa Draper is a manager, a web geek, an Ubuntu contributor and Pad Thai addict. She was our first guest poster, and is now a recidivist offender.

As a woman involved with several projects that aim to encourage women (and other marginalised groups) to get involved in Open Source, I often come across the concept that nothing needs to actually change within our projects to achieve better gender balance, we just bait and blackmail as many women and girls as is humanly possible to try something such as Ubuntu. The philosophy behind this is that some of these women and girls will stay and conquer the gender divide to much rejoicing.

Basically; If you throw enough mud at the wall, some of it will stick.

To a tinkering community, this may sound logical; If you try enough at something, you’ll get some successes eventually. In fact, the philosophy is thought to be based around the wattle and daub method of construction.

I have an issue with applying this to the Women-as-a-minority issue. My apprehension with applying this philosophy to potential contributors is that humans have remarkable abilities that wet dirt does not. Humans are able to remember things. No elephants or knots of string required. Humans are also able to make decisions about what they do and prioritise good experiences above bad ones.

If I take 50 women and successfully encourage them to join a project because everyone is nice and they will feel like they are frolicking with bunnies, but they actually have a contradictory experience — then this is bad.

I cannot magically take away the bad experience that they had. I cannot turn back time and make the sexist joke or come-on go away. I cannot turn back time and make the negative stereotyping alienation non-existent.

Basic sales and customer service training will tell you (right after “keeping customers is cheaper than getting them” (pdf)) that it only takes one bad experience to turn a customer in to an ex-customer. If it is the first experience, then it is more or less permanent. First impressions really do count.

As an example, it took until this millennium to convince my mother to use garlic in her cooking. She got turned off the taste from eating pizza totally smothered in it at an Italian restaurant in the mid 70s. That single bad experience took 25 years to overcome.

So, back to our 50 women. Lets say that after this bad experience, of these 50 women, 1 stayed. I have then achieved a 2% hitrate.

(A note about customer satisfaction statistics: the “customer to ex-customer” conversion is typically in the order of 80-90%. The worse the bad experience, the higher the conversion. For the purposes of this article, I am using a commonly accepted ballpark figure of women in open source.)

The other 98%? I would not hold my breath waiting for them to come back for a second helping. Once burned, twice shy, as they say. Actively try to lure them back, you say? Good luck. In the real world, you probably don’t even know that (or why) they’ve gone.

Ironically, had I been honest with these women from the start, then it would likely have been different. Had I said that it can be a challenge; that bad things can happen, and why it is a challenge; but that there’s a support base they can fall back on, then I’d not have been ‘tricking’ them. I would not be misrepresenting the project, and they’d not be buying in to it expecting a utopia. I’d not be breaching their trust.

If I can illustrate how the available social support bases and safe spaces such as (<project>_women, <industry>chix for example) have been used in situations in the past, then I’m giving them the power of knowledge to be prepared for reacting to a bad experience. It will ensure that they know what to do, who they can turn to and so forth. It will ensure that they are many times less likely to feel like they’re the only person ever made to feel alien within the project community.

The bad experience would be put down to the challenge they were warned about. The warning made it expected enough that the response to the experience would be much less likely to be the reflexive fleeing. The overall level of hurt after support from the group they were told about would be many times less.

We don’t want to only ever achieve a 2% hit rate. We cannot afford to. If we are to get half of the world’s computer users to use a non-proprietary operating system as is the desire behind things such as Ubuntu’s Bug #1 then we cannot afford to scare away 98% of the women and girls (you know, half of the world’s human population) who are exposed to the alternative. Doing so means we need to achieve a conversion rate of 98% of the male half of the population.

To achieve a useful sticking rate, then we are going to need to do something other than throwing every woman and girl we come across in the general direction of the nearest project in the hope she does not get repelled for the next 25 years by the experience.

Looking for opportunities for change within your project is a pretty good start.