CHAOSScon - D&I workshop

A Time for Action: Ethics, Privacy, & Consent in Metrics that Matter for D&I

Diversity drives performance. 

However, diversity is a huge problem, particularly in open source vs. proprietary software.

Mozilla has done of work in this area.

Problem areas:

- *Gatekeeping*. For example, people with stronger English skills get more opportunities, only translating things they don’t want to do themselves. Women not included in a lot of core groups. Even in groups intended to empower diversity. Are they enabling others or enjoying their positions?

- *Lack of Strategy to Address Toxic Behaviour*: Sometimes watered down versions of codes of conduct; sometimes once-vibrant projects dwindle. This means significant financial loss to your project; despite the “rockstar” pumping out code and being awesome. Staff unsure of how to actually deal with bad actors, particularly in key roles.

- *Disconnect in lived experiences* For example, people believe that there is far less safety in communities than men/women contributors.

- *In broader ecosystem, everyone hitting the same walls.* Lots of people having the same problems, not a lot of projects reaching “across the aisle” to collaborate on solutions with one another.

CHAOSS D*I working group working across projects to deliver something meaningful.

7 goal metrics: communication, contribution, events, governance, leadership, project places, recognition.

Factors: gender identity, sexual orientation, age, location, socioeconomic status, tenure, race/ethnicity, first lanauge, English competency, dis/ability, caregiver (child/eldercare), identifies as “underrepresented”

Various challenges; for example: introverts overrepresented because they are around computers more. :)

Ethics, Privacy, & Consent

- Core set of interests/values: This is what brings us all together. (e.g. we care about metrics, open source communities.)
- My contributions: This is what I contirbute.
- Specific interest/focus areas: I’m part of CHOASS because I’m a data scientist. I’m part because I can bring/get value.
- Unique about me: Canadian mother of 3 children, used to be a software developer, made up job title, a widow…

The further out, the less alike we are, and harder for us to relate to one another.

If you start measuring those things I’ve shared about myself in relation to contributions… why are you doing that? How is me being a parent or Canadian relevant? Why is it your business?

“It’s because we want you to be more successful. It’s because you have less time in the day, we want to present you with the right type of tasks.”

===

Design thinking / ideal project. Empathy first… understand people you’re trying to design for. Listen. Repeat back what you heard. Step back from solution. What is the problem we’re trying to solve?

Exercise: Find partner who you don’t know / don’t know well. And ask them about “Data about me”

- contribution data
- interest areas
- demographics

Not to share personal information with stranger, but what it feels like to share this data with an open source project like this.

Person interviewing: listen to person you’re interviewing; ask “why” a lot. “Why do you think that is / why is that uncomfortable?” esp. if struggling.

“Emma, tell me about your contributions.” 

“I’ve been a software developer.. you’ll probably find some data on GitHub… some of them are on a wiki… mailing lists… helped people on twitter… etc.”

What’s the core of you, what makes you similar to everyone else?

What are your interests? “Always been interested in contributing to open source.. always been interested in writing curriculum (open badges)”

Unique about me? “Ok, so… you have my email, going to search about demographics… things that make me nervous that I don’t want you to know about. Or don’t want it to weigh into a good contributor.”
“What makes you nervous?”
“I’m worried someone will judge because I’m in my 40s. Find a thread where I lost my cool one day.”
“Why does that matter?”
“Reputation in open source is really important.”
“Why is it important?”
“…they tell me it is?”

Purpose of Interview is to put yourselves in the shoes of a marginalized person whose data is going to get combed through. What does it feel like to have your age, gender, all publicly transparent and available?

Goal is to understand the role of empathy in metrics gathering.

===

Feedback:

- Not as bad as we thought. :D
- We in this room feel comfortable sharing details about ourselves, but we’ve all been around 5-10 years. Not at risk.
- Lots of white people in the room. We don’t have many PoC, who represent the types of diversity we’re trying to attract to open source. These people have the power over the data gathered, and have blind spots.
- There are people who intentionally share some demographic data (for example, intentionally being a feminine-sounding vs. neutral name)
- Blind spots. I’m a white guy with a certain perspective, might think things are anonymized enough, but the wider diverse community would not necessarily feel safe.
- What’s intention of using the data in a clear, representable format? 50-page Terms of Use documents often get clicked through. We need some sort of clear, up-front statement about what data is used for.
- One idea: “Creative Commons” for ethics. Make it easy to spot “at a glance” how data gathered will be used. (And some kind of compliance measure.)
- Very GitHub-focused. Half of my contributions to this community are in Slack, or day-to-day support. Lots of blindspots.
- Imposter syndrome angle. I have no academic background, assume I’m less knowledgeable than others.
- Some people are very good at surfacing their accomplishments; others are scared to show anything because it might not be good.
- I have very little coding contributions. Will people look at it and say “that’s lesser, because they don’t have the kind of profile I think they should have.”

===

More feedback welcome @ https://github.com/chaoss/wg-diversity-inclusion