Andrew Begel and Alexander Serebrenik

andrewSerebrenik_20042291_WI_PO_VH_0694Today’s Identity Function is a dual interview with Andrew Begel (left) and Alexander Serebrenik (right). Andrew Begel is a Senior Researcher in the Ability group at Microsoft Research. He studies software engineers to understand how communication, collaboration and coordination behaviors impact their effectiveness in collocated and distributed development. He then builds software tools that incentivize problem-mitigating behaviors. Alexander Serebrenik is an Associate Professor of software evolution at Eindhoven University of Technology in the Netherlands. His research covers a wide range of topics, from source code analysis, to collaborative and human aspects of software engineering.

Tell me about yourselves.

Alexander: I grew up in Moscow, Russia, studied at Hebrew University in Jerusalem, Israel, and got my Ph.D. from Katholieke Universiteit in Leuven, Belgium. I also used to be a postdoc in France before accepting a faculty position in the Netherlands. I still live in Belgium and cross the border twice a day. When it comes to personal interests, I like art (“flemish primitives,” anyone?) and opera. I’m happily married and a proud “daddy” of two beautiful Yorkshire terriers.

Andrew: I grew up in New York, went to college at MIT, and got my Ph.D. in 2005 from University of California, Berkeley. I currently live in Seattle, WA with my fiancé and two dogs. I love cooking, eating, seeing Broadway musicals, history, traveling to international cities, and watching funny YouTube videos that my fiancé finds for me.

And how do you two know each other?

Alexander: In 2012 together with two coauthors I submitted a paper to a magazine and Andrew was appointed to shepherd it (and did a great job with it).

It seems like you’re both interested in the human side of software engineering. Can you tell me a bit about this area?

Alexander: The idea is very simple: Software is not only produced for humans, but also by humans. This means that personal stories of developers, their fears and aspirations, and their daily lives and work affect both what software is produced and how it is produced.

Andrew: Developers are people, first and foremost. In groups, teams, and organizations, they behave the same as anybody with any other job. For example, we have looked at why software developers change teams within a company. Not surprisingly, many developers report leaving their team because they want new challenges, or they don’t like their manager. Those who have been at the company longer more often say that their decisions about leaving or joining are about liking the manager or the team they work with. That totally makes sense. It doesn’t really matter what you do, as long as you like the people you see every day at work.

And what are you currently working on?

Alexander: My most recent projects are related to diversity in development teams and understanding emotion in developers’ communication. When it comes to diversity, I would like to understand how different minority groups operate in development teams and communities, what obstacles they encounter, and how we can support them. On the emotional side, I’m looking at what makes developers angry: For instance, if they are angry about code they are working on, this code might deserve special attention, while if they are angry about each other, a very different kind of intervention might be needed.

Andrew: Lately, I’ve pivoted my research to focus on helping tech companies work more effectively with autistic software engineers, and more generally to focus on accessibility technologies for people with disabilities.

How do you design and carry out these kinds of studies? What have you found so far?

Andrew: The studies we do take advantage of standard methodologies from HCI for things like user studies and usability studies as well as for surveys, interviews, and observations of software developers. I usually use mixed methods analyses to understand what’s happening in a small population that I might have interviewed, and then follow up with quantitative analyses of larger samples from surveys. I’ve learned a lot about how software engineers communicate (or don’t communicate) when working between teams, and how professional (read: unemotional) they can be when working on code, even when that code was written by other people.

Alexander: So far my studies have usually combined analysis of data from large scale software data repositories such as Github and StackOverflow with listening to what developers say in surveys and interviews. When it comes to emotions, I’m analyzing their reflections in written communication. When it comes to minority groups, we’ve been mostly focusing on gender and more recently on developers with English as their second language. We have observed that more gender diverse teams on Github are more productive than less gender diverse ones, and that in more traditional environments such as mailing lists, women and men show similar contribution patterns as opposed to more gamified communities such as StackOverflow, where women’s levels of participation are significantly lower than men’s.

Do you have any hypotheses for why these things might be true?

Alexander: Our hypotheses tend to come from outside of the software engineering domain. For instance, gamification as implemented in StackOverflow encourages competition, and from behavioral economics we know that women tend to be less effective than men in competitive environments, even if they are able to perform similarly in non-competitive environments.  Moreover, the observed effect has been stronger when women have to compete against men than in single-sex competitive environments. Similarly, theories about advantages and disadvantages of diversity we have used come from organizational science.

What have your experiences been like being LGBTQ in the software engineering community?

Alexander: I’m very open at my work, my colleagues know that I’m married to a man and some of them have even met him. Several students have contacted me in the past with LGBTQ-related questions. One of these questions triggered a recent creation of the LGBTQ group of students and employees; we have received support both from the university and from the community of students and employees.

Andrew: I’ve always been out while working at Microsoft. In grad school at Berkeley, I was in a great place to meet and socialize with other gay people. When teaching, I wore a rainbow pin on my backpack to show that I was out to my students. In the software engineering community, I’ve never shied away from talking about my personal life with my friends and colleagues. I’ve never felt any disapproval or negative reactions from anyone.

It’s great that both of you have had positive experiences in the community, and that you’re serving as role models. Were you out during the job search as well? What was that like?

Andrew: When the time came to apply for jobs, my parents worried that me being out could limit my job opportunities. However, I am convinced that a place that could frown on my personal life would not be a place I wanted to work at anyway.  Location did matter, however. At the time, each state in the USA had differing laws governing gay rights, and honestly, some were really bad. Washington turned out to be a great place to move, and was one of the first states to extend partnership rights to same-gender couples.

Alexander: I was out to close family and friends, but did not really plan to discuss this during job interviews. However, during a job interview in Eindhoven, I was confronted with the question, “why did you decide to apply for a position here?” I decided to tell the truth and said that my boyfriend lived in the same area. When I told this story to my boyfriend (now my husband) he was taken aback and was pretty sure that I would never get the job. Surprisingly, I got the job and still work at the same university.

Was it different for you in different countries?

Alexander: As an undergrad student in Jerusalem I was completely closeted, but I had some experiences that influenced the way I think and teach today. When I was a first year student, I saw a television interview with the head of the Jerusalem Open House and his boyfriend can you imagine my surprise when I recognized one of my TAs? He turned out to be the very first gay person I saw in reality. And yes, we never talked to each other about anything but algebra, and it took me many more years to accept my sexuality and come out, but this was the first step.  This is why I believe that it is extremely important for us LGBTQ teachers to be open, to be out and visible: There may be one of our kind in the audience fearing to come out.

Speaking of which, last year, you mentioned that both of you were organizing meetings for LGBTQ researchers at software engineering conferences. Can you tell me more about that effort?

Alexander: This series of meetings started in September 2016, when my favorite conference was organized in Raleigh, NC. You might remember that in March 2016 House Bill 2 (the transgender bathroom bill) was actively debated in the NC Senate and House of Representatives. The general chairs of the conference put a very nice and welcoming message on the website stating that despite this debate, the conference welcomed everybody. During the conference itself, the mayor of Raleigh reiterated this point and a restroom at the conference venue was explicitly designated as gender-neutral. Encouraged by the general chair, we organized the first LGBT in Software Engineering lunch. Since then we have organized such lunches on multiple occasions all over the world, including in Buenos Aires and Shanghai, and most recently in Gothenburg.

That’s great to hear. Are these held at every major software engineering conference? Are the lunches just opportunities to chat with a supportive community, or do you specifically discuss diversity?

Andrew: We hold them at any software engineering conference we attend. We often channel the discussions towards questions of diversity, challenges, and how to overcome obstacles.

Have any ideas come out of these discussions so far? If so, what?

Alexander: We succeeded in including a special LGBTQ-safety related clause in a conference charter. But I think that the most important thing that has come out of these discussions is the community itself. We are slowly building a group of people that make LGBTQ researchers more visible.

Thanks to both of you for your time!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s