Originally Printed in the Summer/Fall 2009 Newsletter
Rachel Pottinger is an associate professor in the Computer Science Department at the University of British Columbia (in Vancouver) where she is affiliated with the Data Management and Mining group. She earned her Ph.D. in the data-base group in the Computer Science and Engineering Department at the University of Washington. Her research focuses on managing data that are poorly supported by traditional databases. You can read more about her here.
Q: Explain a bit about your research. What kinds of data sources are you trying to organize and coordinate that are not well supported by current databases?
I’ll give you one example that’s a great source of problems: civil engineering data about building a building. Right now, most people involved in the process use computers: architects use CAD models, contractors use estimating software and scheduling information, etc. However, when they try to communicate, they typically print everything and put it on a table. This means that if the contractor looks at a design and says “if you lower the ceiling by 2 cm, I can save you $50,000,” trying to figure out how this impacts the rest of the systems (e.g., the ductwork and the electrical work) is a big pain. Hence the building process both isn’t as flexible as it could be, you tend towards a local maximum and it’s tedious because people have to redo a lot of things by hand. Civil engineers know this, so they created an XML standard for design data in order to make everything interoperate. After all, we (the computer scientists) told them that XML would solve all their problems. Only it doesn’t. As it turns out there are at least three major XML standards within the scope of the very simple problem that I’ve described above, each of which describes different parts of the data, and each of which is exceedingly difficult for civil engineers to understand. I had a CS masters student look at how to pull out all of the data for some simple queries (not even at the level of the one above), and it took her three months, because trying to figure out how many openings were in a wall took tracking down five different levels of idrefs (i.e., pointers).
So there are problems that need to be solved in dealing with the design data. There are also problems with updating estimates. And if I run out of work there, they also keep all sorts of meeting minutes and other data that could help them figure out whether a project is on track or not. It’s a great source of data. And many fields have such sources. I’m also working on data from disaster management and business intelligence data, and I’ve looked at bio/health informatics data a bit, too.
Q: How did you become interested in these problems? What do you think are the most exciting opportunities for research in this area?
My Ph.D. was on more abstract research. When I graduated, while I had enjoyed the work that I had done, I wasn’t convinced that the work that I was doing was going to be really relevant to people actually trying to use databases. Sure, there were lots of technical reasons why what I did was useful, but given the advances that we’d made in the behind-the-scenes components, I wasn’t sure that we were really solving the problems that prevented people from integrating their data. So I went out and talked to various people across the university and Vancouver. My first two years as a faculty member were largely trying to figure out which were going to be fruitful collaborations and sources of problems while in the short run my students and I continued on the direction of the research that I’d started as a student.
The most exciting opportunity for research in what I’m doing is the chance to talk to people outside of computer science about what they’re doing and how what I’m doing is relevant. I believe that computer science is at a cross-roads right now: we can either define ourselves broadly, and consider cross disciplinary work (e.g., bioinformatics) to be part of computer science and have all sorts of interesting problems to work on, or we can define ourselves narrowly and talk ourselves right out of all of the interesting problems. My research is directly in line with that.
Q: What do you enjoy the most about your career right now?
I work with great people. One of the most important things that I’ve learned is that life is too short to work with people that I don’t like to work with. When I was a grad student, I was once in a meeting with one of my advisors, and after I asked him how he was, he said that he thought I was physically incapable of having a meeting without asking him how he was. I thought that one over and realized that he was right. So I only work with people who I care enough about to ask them how they’re doing. Fortunately, that covers a lot of people. But it’s true for both students and colleagues.
Q: What are the greatest challenges of your career right now?
There’s not enough time to do everything well. It’s hard to let things go when you know that, with an infinite amount of time, you could do things better. I try to keep up with my graduate students as well as I can, but I’d like to do better (I recently told one of my grad students that, and she laughed at the notion that I didn’t spend enough time with my students. It’s true that I probably do spend more time with them than many advisors, but I still wish I could do more). I can keep up better with the things that people depend on me for, though, than I can with the things that I’m the only one depending on me for, like publications from work of masters students who have graduated. Unfortunately, this is an area that I really can’t afford to ignore as a pre-tenure faculty member, so there’s a great deal of guilt in spending time doing lots of things that are unrelated.
Q: How is an academic career in Canada different from one in the US?
I’ll focus on being a faculty member who has research and supervising graduate students as part of her job description, since otherwise it’s too hard to make generalizations. A big one is that grants in Canada tend to be evenly distributed: more people get them, but they’re much smaller than the grants that you would get from NSF. To have more funding generally requires some kind of industrial support. I also don’t have to pay myself my summer salary out of grant money. So it’s easier to get enough money to have a small group, but harder to have a big one.
Beyond that, it’s fairly hard to generalize. My department is fantastic and very supportive (since my husband and I both found jobs there at the same time, I tell people that I feel like I won the job search lottery), but there are many aspects of it that are due to the department rather than being about Canada. I will also note that when I was on the job search, I met very few U.S. junior faculty women who had kids, and I met many Canadian junior faculty women who did. Our parental leaves are much better for both men and women, and I think that the departments in general are more supportive.
Q: How have your interactions with CRA-W affected your career? In what other activities in the support of women in computing have you been involved?
I was a DMP (now DREU) student as an undergraduate, which is where my interactions with CRA-W really got started. I did my work at my home institution, Duke, as I requested, which in some ways wasn’t really smart of me, it would have been better to meet more people at other institutions, but in other ways was fantastic since it let me make really good local connections that I could use for the rest of my career. I participated in a career mentoring workshop the year before I got my Ph.D. That was very valuable; in addition to the advice there, I made contact with a great group of people, some of whom I’m still in touch with. I set up a mailing list for graduating Ph.D. women in CS to help others benefit from this type of community (PhdjobhuntHers), as well as a mailing list for pretenure women in CS (JrProfessHers). I think that they’ve helped a few people, but I wish there was something that I could do to make them more successful. The lists are very quiet. When I talk with people at conferences, it’s pretty clear that people on the list just aren’t comfortable talking about the problems that they’re having. It makes sense, but it’s frustrating, there are people out there who need help, and I can’t figure out how to get it to them.
I’ve also spoken at two of the Grad Cohort programs and been on two panels for the Distinguished Lecture Series. Beyond that, I’ve also mentored in several official and unofficial capacities, first just students, and now also unofficially other junior faculty members. I’m also on our department’s committee for the Focus of Women in CS where I’ve been working on increasing our participation in Grace Hopper (both to help our current students and to attract more women to apply to UBC for graduate school).
Q: Tell us about your successful solution to the “Two-Body Problem.”
I met my husband, Steve, at orientation of our freshman year of college. Four years later, we went to graduate school together. As graduation approached, it became clear that I wanted a position at a research university, and Steve was interested in a more teaching-focused position. So we got a big map of colleges and universities in the U.S. and Canada and put pins of various colors in the map where there were places that we were interested in. Then we applied to places where there were pins for both of us—I think I applied to 46 places and he applied to 56. When
all was said and done, it was clear that UBC was a great fit for us—they had an opening in data management, and they had a tenure-track teaching position opening, and they wanted to hire us. Plus, it’s a great department, values the things that we value, and we would get to live in Vancouver, which is both beautiful and near where Steve’s parents and my mom lives. We’d specifically not ranked the places that we were applying, since we didn’t want to get our hopes up, but if we had, UBC would have been close to the top.
Q: How do you balance work and family?
Sleep. Oh, wait, no. Um. No, really, sleep. I prioritize the things that are important, and try to concentrate on doing those. Having time with Steve, and our daughter, Naomi is important, as is getting enough sleep, or I get cranky and can’t function (I’ll leave it to you to decide which of those is more important). So I get to bed at a decent hour, and when I’m not doing work, I try to do things that are fun for the family—for example, Steve and I like to get Naomi to help us cook — that means we get to spend time with her as well as eating. We also eat out more and pay people to do things—before Naomi
I remember when I was a grad student we had a guest speaker who talked about doing such things, and I thought that sounded like an awful way to live, working all the time, but it’s usually pretty doable. You end up with many tasks, like answering e-mails, that are simple and easy to do at night, and I don’t have to drive back into work or get all stressed out to do them, and answering e-mails can be fun, too.
I’ve also given myself a goal of submitting one paper (whether it’s a new one, resubmission, or camera-ready) a month to keep myself from freaking out about needing to publish a zillion papers. That helps, too.
I sleep at least 8 hours a day unless something really urgent crops up. Knowing that I’ll want to be up and doing things with Naomi by 7:30am seven days a week helps me put down the work when it’s time for bed. I also couldn’t make life work if Steve and I weren’t equal parents and partners, but we are. Naomi’s daycare is also fantastic, and I couldn’t do it without that, either. A supportive extended family helps a lot, too.
Q: Do you have any advice for young women con-sidering a research career in academia?
Someone (maybe Jan Cuny) once said that when you’re a junior faculty member, you should do the job that you’d do anyway. If you can get tenure doing that, great. If not, you don’t want to be tenured in that job anyway. This is really good advice. There are things that you have to do that you don’t really want to do, but there are things that you have to do, and things that you just feel like you’re supposed to do. The trick is figuring out the difference. That’s where having good mentors is important. It’s great to have as many mentors as you can, and definitely try to get some who are both local and not. You’ll need them for different things. I’ve been exceedingly fortunate in this, and it’s made a huge difference.