Stop doing the same thing but expecting different results ----------------------------------------------------------------------------------- As Python developers we put considerable effort into optimization. From the Zen of Python and PEP-8 to virtual environments and testing we are always tinkering, trying to make things better, and striving to remove anti-patterns from our code and our development processes. The Python community is known for accepting that like good tests, agile methodologies, and virtual environments, diversity and inclusion are good things. Yet for some reason we have not been as good at applying this spirit of optimization to the problem of increasing diversity and inclusion. While there have been various attempts to improve things, the amount of diversity and inclusion in tech companies and communities has remained largely flat. This talk addresses the question of why improvement in diversity and inclusion has been so difficult, illuminated by my personal experience working in different genders. This talk argues that just as there is no single easy way to write good code there is no single easy way to increase diversity. Rather than looking for sweeping solutions to increase diversity and inclusion we would be better served by acknowledging and reducing the many things we do that actively work against diversity. These anti-patterns for diversity are deeply ingrained and their effects tend to be invisible to those not on the receiving end. In particular there are anti-patterns which are due to refusing to see or acknowledge the problem, including uncritically invoking meritocracy; anti-patterns which are more avoiding the issues, as in blaming the pipeline or society; ones that arise from double standards and a lack of defined processes; others which stem from emphasizing culture fit and privileging free speech; and those that involve gaslighting, victim blaming, and willfully ignoring the lived experiences of those marginalized. While possible solutions to these will be alluded to in passing, there are no neat one size fits all prescriptions. Instead the point of this talk is raise awareness of these anti-patterns and to remind us we are all susceptible to them, even inside the Python community, regardless of gender, race, sexuality, or any other characteristic. Correcting them, in our communities and in ourselves, is up to all of us.