Advocating for HCI in Computer Science Departments

Note: This post generated a fair bit of discussion on Facebook and I’m adding (as block quotes) a few of the comments since this was first posted.

Recently, I’ve had a number of conversations where I’ve been asked whether I feel like I belong in a traditional Computer Science department. Given that my whole academic career has been centered on Computer Science (from declaring it as my major upon entry into undergrad to my current faculty position at U of M), I can definitely say that it is my community of practice. I get my funding from NSF CISE, I publish almost exclusively in ACM conferences, I teach only Computer Science courses — where else would I belong? But, I also realize that it is a salient question for many colleagues in my field, as HCI1 researchers are more and more likely to find their home in iSchools, Design, Psychology, Cognitive Science, and Human Factors programs. In this post, I want to articulate why I am of the opinion that Computer Science department and colleges/schools of Computing should be doing more to recruit and retain Computer Scientists trained in HCI.

Meaningful Evaluations – Most Computer Scientists develop algorithms and systems with the idea of improving upon some existing baseline. While some metrics of success may be straight-forward to articulate (e.g., time to run, prediction accuracy), many evaluations may require consideration of a richer and more nuanced set of factors. In the end, the most meaningful metrics look at how a system or algorithm performs in service of people. HCI is the branch of Computer Science where researchers have the training to plan and perform evaluations that lead to more meaningful, ecologically-valid comparisons. More importantly, Computer Scientists trained in HCI also have the fundamental understanding of the development process that allows them to carry out studies that lead not only to summative comparisons, but also to specific implications for how systems and algorithms can be improved in order to perform better in real-world evaluations. For example, an HCI evaluation may reveal the relative consequences of different types of algorithm accuracy errors, leading to an understanding of the kinds of improvements to accuracy that have most significant effects.

I agree with all the above, and would add two things: 1) some HCI work is more oriented around systems building, and that is more naturally centered in a CS department. Sometimes it is very technical application of existing techniques and sometimes it is new techniques – but this kind of work readily live in multiple places. 2) I think the best traditional technical work happens when it is done with close work with stakeholders. Even theoretical work can be stronger when considering the real needs and problems of users. Developing that understanding is best done in close proximity to them – thus integrating HCI into CS can help that process.

-Prof. Ben Bederson, University of Maryland

Diversity – Computer Science has a well-documented diversity problem, which has been growing in recent years. It has also been documented that both women and minorities are more interested in applying STEM to culturally-relevant broader impact contexts. Thus, it’s not surprising that of the few women and minority Computer Scientists who do make it through grad school, many choose focus on HCI. While CS departments should be making an extra effort to recruit and retain qualified women and minority candidates, my personal experience with faculty hiring is the opposite. I frequently see that given two candidates with similar training and publication records (in terms of venues and quantity), women are more likely to end up in iSchools while men are more likely to stay in Computer Science. If this trend is real, it is bad for CS because it is a lost opportunity to increase diversity. The problem is two-fold. First, a CS department may not make an offer to a qualified candidate due to a narrow definition of what constitutes technical work or a contribution to Computer Science. Second, even when an offer is made, the candidate may perceive a department culture that is unfriendly towards HCI work and elect to accept a competing iSchool offer (despite the fact that iSchools may pay less and have a higher teaching load).2 The New York Times recently ran an article showing that as more women go into specific field, salaries in that field decrease. We don’t know the mechanism by which this happens, but my worry is that this is the pattern that is emerging here. Of course, there could be a number of alternative explanations, but I think that it can’t hurt to collectively keep an eye on this.

Learner-Centered Teaching – There are many CS courses that do not require any specific area specialization to teach, most notably introductory classes to CS, computing courses for non-majors, and freshmen seminars. There are many reasons why HCI faculty may be particularly well-suited to achieve excellent outcomes in such courses: (1) there is a clear transfer of skills between designing user-centered systems and designing learner-centered curricula, (2) HCI research can be presented in these courses as a counter-example to popular misperceptions of Computer Science as an asocial, solitary pursuit, and (3) HCI faculty may provide stereotype-threat-breaking role models, given the larger proportion of women and minority faculty in HCI. In my opinion, Computer Science departments with a strong undergraduate education mandate should be actively seeking to recruit Computer Science HCI faculty.

I disagree with “there is a clear transfer of skills between designing user-centered systems and designing learner-centered curricula.” There’s a potential for transfer, but I’ve met some HCI (and even Ed) researchers who are really awful teachers. And I don’t like the idea of giving all other CS teachers an excuse, e.g., “I don’t do HCI, so I don’t have the background to be a good teacher.”

-Prof. Mark Guzdial, Georgia Tech

I wanted to keep it to these three top reasons, but there are a number of other ways that Computer Science as a whole benefits from the specific skills that Computer Scientists trained in HCI bring to the table, including more interdisciplinary work, broader impact of research, and considerations of critical issues (e.g., bias) at design stage.

If you are currently in a Computer Science department and you agree with my arguments, there are a few clear steps you can take to help your department: (1) advocate for soliciting and hiring Computer Science candidates trained in HCI, (2) confront and question arguments against hiring a candidate that focus on vague concern of “not being technical enough” (especially, when candidate is trained in Computer Science and publishes in computing venues), and (3) articulate narratives of Computer Science that include rather than exclude HCI.

1. I use HCI (Human-Computer Interaction) here as the broader discipline, but this includes many related disciplines including HRI, Game Studies, Social Computing, Mobile & Ubiquitous Computing, etc.
2. I am uncomfortable calling out specific schools and contrasting specific people. I am basing this entirely from my personal experience on the job market 3 years ago. I received 3 offers from iSchools and 3 offers from CS departments. Your mileage may vary.

Purpose, Visibility, and Intersubjectivity in Video-Mediated Communication Technologies

Video-mediated communication may be able to benefit from a number of novel technologies, but designing for a good experience requires considering purpose, visibility, and intersubjectivity for both partners.

Skype, Google Hangouts, Facetime, and ShareTable are all examples of real-time video-mediated communication technologies. Designing, implementing, and deploying novel systems of this sort is a big research priority for me and every semester I get a few entrepreneuring students approaching me with ideas for cool new technology to try in this space: robots, virtual reality, augmented reality, projector-camera systems, and more. Frequently, I ask them to consider a few things first and if you’re new to thinking about computer-mediated communication, these may be helpful for you as well (many of these ideas come from my work with play over videochat).

In this case, let’s assume the “base case” of two people—Alice and Bob—using a potential new technology to communicate with each other (though the questions below can definitely be expanded to consider multi-user interfaces). Consider:

  1. (Purpose) Why is Alice using this technology? Why is Bob? The answer should be specific (e.g., not just “to communicate,” but “to plan a surprise party for Eve together”) and may be different for the two parties. It’s good to come up with at least three such use cases for the next questions.
  2. (Visibility) What does Alice see using this technology? What does Bob see? Consider how Alice is represented in Bob’s space, how Alice can control her view (and then flip it and consider the same things for Bob). Consider if this appropriate for their purposes. For example, maybe Alice is wearing VR goggles and controlling a robot moving through Bob’s room. It’s cool that she can see 360 degree views and control her gaze direction, but what does Bob see? Does he see a robot with a screen that shows Alice’s face encased in VR goggles? Does this achieve level of visibility that is appropriate for their purpose?
  3. (Intersubjectivity) How does Alice/Bob show something the other person? How does Alice/Bob understand what the other person is seeing? The first important case to consider is how Alice/Bob bring attention to themselves and how they know if their partner is actually paying attention to them. If Alice is being projected onto a wall but the camera for the system is on a robot, it will likely be difficult for her to know when Bob is looking at her (i.e., when he’s looking at the wall display it will seem that he’s looking away from the camera). It’s also useful to consider the ability to refer to other objects. Using current videochat this is actually quite hard! If Alice points towards her screen to a book on the shelf behind Bob, Bob would have no idea where she’s pointing (other than generally behind him). Solving this is hard—it’s definitely an open problem in the field—but the technology should at least address it well enough to support the scenarios posed in question 1.

Generally, I find that new idea pitches tend to propose inventions that provide a reasonable experience for Alice but a poor one for Bob. It is important to consider purpose, visibility, and intersubjectivity experience for both of them in order to conceive a system that is actually compelling.

Notes from the Field: Pokémon GO!

The obligatory evidence of being accosted by Pokémon during everyday activities :-)

The obligatory evidence of being accosted by Pokémon during everyday activities 🙂

Note: As promised, I’m revisiting this post (roughly 6 months later) to review my thoughts and predictions. Reflections are highlighted like this note.

Those of you who know me, know that I am obsessed with Pokémon (it was even the theme of my Intro to CS class last semester). I’ve blogged about Pokémon before (1 & 2), but now that everybody has bought into my insanity, I feel the need to do it again. Of course, I’m talking about the mobile augmented reality game Pokémon GO!, which has taken the U.S. by storm over the last two weeks.

The general Internet rhetoric regarding the game is a bit extreme (of course): it will kill your children (maybe in the process of arresting them), after robbing them, and stealing all of their data! Or, it will turn your children into monsters who mock the Holocaust or stare at their phones instead of the real world! Also, it may get YOU sued (as a property owner), even if you don’t play it. On the other hand, it could improve people’s health, get them to go outside, make friends, and help them discover and fall in love with their own cities!

I want to share my experience. I’ve been conducting “participatory observation” since the day Pokémon GO! came out: I’ve been playing myself and I’ve been stopping people on the street who are playing and asking them questions. I’ve done this in two cities (4 days in Minneapolis, 4 days in Chicago). Based on this, there are a three points I want to add to the above narrative:

      1. It is probably not doing anything to your children. There may be children playing this game, but I haven’t met any (only anecdotally: one man I met said he DID see one child playing it). All the players I’ve met have been roughly in the 19-35 range. I’m currently doing fieldwork at a middle school and NONE of the children I work with play. My main two main hypotheses for this: the game was aimed at the nostalgia of the older audience (e.g., it only has 1st generation Pokémon), and American children are not allowed to be as independent as Japanese children in exploring outside. Revisiting: Actually, this totally became more popular with the kids later — but it did take a solid 3 weeks post-release. Fail.
      2. It’s more about spatial computing, than the AR or stepcounts. The augmented reality (AR) is exciting at first and can always be turned on to catch a good photo op, but most experienced players turn it off to make catching Pokémon easier. The step count intervention is actually only a small part of the game and nothing new for Pokémon. What IS new and integral to the game is that everything happens in physical space, on a real map that you must move through to have meaningful encounters. This is positive because it creates opportunities for encounters with strangers, gets you moving, and (most importantly for me) leads to a playful exploration known as a dérive. It is negative when it leads to car accident, arrests, injuries, trespasses, disrespectful behavior, and all the other problems mentioned at the beginning of this post. Do the potential benefits exceed the costs? Each player must decide for themselves. Revisiting: I think this one was on the money — people play this as a spatial game not an AR game. Success.
      3. It is undeniably urban. My drive from Minneapolis to Chicago was a Pokémon desert, with very few spots along the way (though definitely enough quirky points of interest that could have qualified). This is because Pokémon GO! relies on user-generated geographic interest points (borrowed from the game Ingress). One big issue with that (pointed out by my colleagues in a recent publication) is that peer generated geographic data is significantly biased towards urban environments. So, while the download numbers may make it seem like the whole country is playing, it’s really just us city-dwellers. Revisiting: Others have picked up on this too, so I think I was right. Success.

And now, for a few predictions… (1) Most people will stop playing soon — it’s too repetitive and battery-draining and nostalgia can only get you so far (Revisiting: Yep, many sources confirm this. Success.); (2) if you buy a few of these devices when they become available, you will be able to turn a profit reselling them (Revisiting: Yep, it you bought it for $35 when it first came out, you can now resell it for over $50 since Amazon and Gamestop are both sold out. Success.); and (3) Pokémon Sun and Moon will be the best selling Pokémon generation ever. (Revisiting: Yep, Sun/Moon was the best selling not just Pokemon but Nintendo game ever. Success.) I’ll check in a few months to see how many of these came true! Revisiting: I’d say that went pretty well! Out of my 6 points/predictions, 5 were successes!

Have you been playing Pokémon GO!? What is your experience with it? Any stories that run counter to the points I’ve made above?

P.S. Team Mystic is the best team.