Publishing Results

I’m attending the HCIC (Human Computer Interaction Consortium) conference. During a session today I learnt about a challenge that Systems HCI researchers face when publishing the results of their work. They seem to feel based on their experience that it’s hard to publish more than one paper about any system. Given that developing a system can take 1-2 years to build, this is a problem. Indeed, it is suggested that this is discouraging people from focusing on Systems contributions in HCI.

Since I do not work in this space, I was quite surprised and disappointed. I was surprised, because I similarly feel that I can find people who feel that it’s hard to publish the results of a study more than once (not the same ones obviously). And like systems, I can find people who would tell me that the contributions of a piece of empirical research can not be put into a single paper, but when it is written about more than once people are concerned that it is a repeat of the previous research. Further, I would add that in both instances people were also not focused on publishing the least publishable unit (LPU).

My disappointment is obvious. It seems like we’re focusing on the wrong things if this is going on. Rather than focusing on the contributions that the piece of work makes to the community we’re focused on the unit of research itself (a field study, a system). I know that’s a statement of the obvious, and yet, here we are having conversations about how we perceive the review process to be working, potentially quite pervasively (since it spans a number of areas within the field and multiple people within each).

A related concern that cropped up during the discussion was the question of how systems work is evaluated. I was reminded of Paul Dourish’s Implications for Design paper. In that paper, Paul argues for broadening the criteria by which Ethnography is evaluated, what contributions it might make, why they may not be design (or even make sense to be design recommendations). I wonder whether it’s time for a contribution to the discourse about how systems HCI research might be evaluated. I don’t review Systems HCI papers very often, but I want to hear more about how best we learn from a wide scope of contributions.

Books and Blogs

Two different posts, but both contain the same message.

Ian’s thoughtful post is about writing books that people want to read as opposed to ones that have to be written for tenure. And here’s a post that suggests that writing popular books, the very type that Ian suggests, is not a good thing to do for tenure.

I know people who’ve written text books, technical books, and scholarly books. This is despite being in a field where books are not as prioritized as in other disciplines. All of them have gotten tenure, and I believe all of them have received it post-writing their first book. Particularly the text and technical books. Perhaps it’s because the field doesn’t tend to place as much value in them as they do in other disciplines. Which says a lot for how to write a popular book in the tenure culture.

I’m not sure I agree with him about the blog, although I do think that there’s probably things that are better unsaid on a blog prior to tenure. But that’s also true in person.

What’s in an Academic Name

Pick the name you publish under wisely.

Why do I say that.

In the days of citation counts and impact factors it’s actually relatively important that you can be “found” in publication searches.

I chose Rebecca E. Grinter to publish under. I chose it because it was, and remains my legal name. When I married (the most likely time that the name would change) I was detered by being a resident of one country and a citizen of another, I decided that the pain of changing my surname legally with multiple governments was not worth it (hmm, I wonder whether I could legally be two different people, one in the U.S. and one in the U.K. another interesting experiment with international law…). I’m also rather attached to the name Grinter, not just because I am a keen family historian, but because it’s relatively unusual. I am frequently the only one in the phone book, and a Grinter event (i.e. meeting another one) is quite rare so always fun.

I chose the E to avoid the problem that R. Grinter would create. That would be the other Dr. R. Grinter, or as I know him, Dad. Of course, I get extra publications if you search me as R. Grinter, so I encourage everyone who is doing a citation count of me for any reason to search R. Grinter. But, then of course, there are the times when my “E” gets dropped, so I end up needing to search R. Grinter to find my own citations. Initials, for all their distinctiveness, seem to create their own problems.

And then there are the publications where I am B Grinter. I’m B Grinter because I go colloquially by Beki (there is one person who calls me Rebecca, I know her as Mum). So if someone writes up the results of a workshop and (kindly) puts my name on it, then frequently it ends up as B Grinter, unless I can intervene and switch it over. This also turns out to matter for my H-index. I wish that my nickname started with the same letter as my official publishing name.

I think now I wished I’d started with Beki Grinter as my non-de-plume for academic publications. It took me a while to realise that it doesn’t have to be your legal name… although I dunno why I thought it had to be my legal name.

I think it’s better if its plausibly a name by which you are known. For example, I think it would be a little odd if I switched to publishing under the name Paul Erdös. Although I have some colleagues whose Erdös number would improve. I think what matters more is that it’s distinct and it’s consistent. Distinct helps people find you, and that’s hugely useful (it’s an academic brand if I’m honest). Consistent helps with time. An academic career is built over time, and having the ability to find people’s earlier works if you find their later ones is really useful. There are likely ways to mitigate this, I like how some people move their former surname to their middle name, and others just let people know on their websites what publications belong to them.

But names are not just academic brands, they are personal choices. But I can imagine a variety of reasons to want to change your name, particularly at marriage….

I guess this started out as a reflection on publishing name. Distinct, consistent, and something plausibly connected to the author seem like good criteria for deciding what name you want to publish under.

What do I do in the Office?

There are several reasons to write this post.

First, this poorly maintained blog has nothing about what occupies the majority of my life (there’s a relationship between the two).  And that is, research in Human Computer Interaction (well almost). Second, because I do research, and because I work in the Academy (which is University, for some reason my American colleagues tell me that as a Brit I tend to use that word more than is normal here) I work in an institutional context which make the “office” more interesting.

OK so that’s two reasons.

Taking the second reason first. The academic context, the University, is a really interesting place to work. If it was, it is no longer the “ivory tower” of lore. No, it is a place where much time and energy is spent in outward facing activities, such as working with sponsors (from the military-industrial complex) engaged in connecting research to real-world needs. And I like that, as some might say it has the effect of “keeping it real.”

In fact meetings are a real hallmark of the academic environment. The Academy is an organization in continual flux, imagine if a 1/3 of workers arrived and disappeared each year. Well if they are students then that’s exactly what happens. We’re constantly wrapping up and recruiting new students. That’s a lot of collective work but it also means that the sheer number of people I meet is enormous. Then there’s all the regular meetings.  Teaching, one-on-ones generated by teaching, and then advising meetings. I spent around 10 hours a week in meetings and interactions generated by the standing teaching requirements.

Then there’s actually finding time for writing and reading.

I’m not complaining, please don’t get me wrong. The job is invigorating in ways that are hard to describe. This is an environment that hands those who want it as many challenges to solve as possible, provides unbounded opportunities to meet people and learn new things from them, and so forth. But, I guess where I’m headed is that I think it always surprises people that I can’t meet with them on the same day, and that’s what I’m trying to explain.

So, when I am doing research it is with my fabulous PhD students. I said that my interests are in Human Computer Interaction and that’s a reasonably accurate description. Human Computer Interaction is all about the intersection of Computing and the human world. But, I’ve also always been very open to trying to bring HCI to other areas of Computer Science, so along the way I’ve published in a surprising (to me anyway) number of areas of Computer Science including Software Engineering, Security, Networking and Robotics. OK so the last one is a little bit of a stretch, I’ve published in conferences in Human Robot Interaction, a field that spans HCI and Robotics. But anyway, there it is.

So, that’s a trailer. But not too much content. What I do in the office is try to spread the wealth of HCI to all of CS, via a set of fabulous projects with awesome student led research, and then in the remainder of my time I meet with people to do all the things that I need to get done in order to do being a faculty member. And in a few months we’ll find out whether I’ve been doing the right things to be a faculty member.