Vis Panel: Building a Research Group in Visualization

This is part 1. I’ll write about the questions in the next post.

Hamish Carr has organized a panel about how to scale up a research group from a few students/colleagues to a larger size. As he pointed out, building a research group is an administrative task. On top of that, visualization is necessarily multidisciplinary. However, the sad truth is none of us gets formal training in administration or in conducting successful multidisciplinary research. So this is a way of getting everyone in the room to talk and think about it. Panelists: Hamish Carr, Sheelagh Carpendale, Min Chen, Tom Ertl, Helwig Hauser, Chris Johnson, Stephen North.

Starting statements.

Sheelagh Carpendale. It’s not about equipment or space. It’s about the people. Some research styles styles: Collaboration, competition, mentoring, individualism, “scaffolding” (individuals + knowledge structure around them).

Min Chen. Research in the UK. There’s a research assessment exercise that determines funding, and it’s based on size and rating. This changes things – you have to scale up. His group is about 25% faculty. On meetings, student and faculty are supposed to be treated the same. Actively seeks external collaboration and accountability.

Tom Ertl. Provide enough grappa, but that might not work for everyone :). Non-linear paths in life are not a bad thing. In fact, broader backgrounds makes interdisciplinary research easier: rech out. Circumstances do help – be prepared. Marketing is important, and a learned skill. Write good papers and proposals! Understand your funding structure. You need breadth, but some critical mass to actually do research anywhere (this forces your group size up). Attract good talent by showing respect – in general, you’re not as smart as the other people. Success is a combination of performance and perception. His group style: 80% PhD students are “home-grown” – you know what you get. Research style: laissez-faire. Small teams within are good – ~3 people.

Helwig Hauser. He mentioned mostly being influenced by working with Eddie Groller (MEISTER!) in the build-up of a vis group at TU Wien. A lot of work, but not a huge deal – just do it. Two ways to go about it. Either you are a Vis enthusiast and reach out to other groups, in which way you have freedom but not money, or you are requested help in Vis, in which you have money but not freedom. Remember to compromise. Are you unreplaceable? If you’re a young researcher, mix proactive research (deadlines? join in. partners? join in. but remember to say no) with proactive research (scratch your itches, work on one grand idea). Remember to explicitly mix the two. Write down your vision. Hire with care. If you are not having fun, rethink all of it.

We’re starting to run late, so the pace will pick up.

Chris Johnson. “We’re hiring postdocs!” If you grow, you will need staff: software developers, administrative staff. Most people stop here, but you can become a research center. Here you don’t know everyone as well anymore – can’t fit everyone around a table! Currently 135 people, 60 PhD students. SCI initially grew “by accident” – getting all three grants while expecting only one. The big goal is to do more interesting science than you could ever do by yourself. Only way to do this is to surround yourself with smarter people than you. Get used to it. Ask the big guys, and they say the same thing: it’s people people people. However, “We strictly follow the no asshole rule”. We think long and hard about hiring – specially when getting started. The 30 minutes became an hour. Watch out for Pauli’s rule for faculty hiring. Remember the critical mass: prioritize.

Stephen North. InfoVis research at AT&T (Stephen wrote graphviz). Much smaller group. AT&T is too big, likes to do things incrementally. If you are in industry, make yourself necessary – write software. Culture at AT&T. Put together statisticians, algorithm designers, software writers, and interesting data. Read John Tukey. Good relations with the DB group, and Network group (good data!). Find good executive support within the company: espousers are important in the food chain.

Hiring: Carpendale is also hiring post-docs and faculty at Calgary, Canada. So is Hauser, at Bergen, Norway.


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 )

Google+ photo

You are commenting using your Google+ 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