What Usenix Can Do for Students

Jun 17, 2009 Conferences, Grad School

After the Usenix ATC welcome session tonight there was a brief Students BoF meeting to discuss what students get out of Usenix (both the organization and its conferences). We talked a fair amount about the idea of student conferences run by students, which I think is a very good idea. The main issue seems to be one of transportation – if it is a national conference, then only people with sufficient funding will be able to get to it, but if it is a regional conference, then only regions with a high density of students (ie the North East and California) will be capable of gathering a big enough crowd.

I still think this is an idea worth pursuing, although it probably works best at the regional level which will sadly leave out a lot of people in the middle of the country. I know that many AI students in my department attend NESCAI (the North East Student Colloquium on Artificial Intelligence held at Cornell each year), and find it very useful since it gives them a chance to practice presenting their work and networking with other people in a low stress environment. It was repeated several times during the discussion that the “hallway track” at Usenix can be the most valuable part, but many students miss out on that because it can be a bit intimidating to strike up conversations, especially with faculty or industry researchers. Giving students opportunities to practice that at a conference just among their peers would be very helpful. For the students helping with conference organization, they would be exposed to reviewing and how program committees work, experience which is normally very hard to acquire as a graduate student.  I don’t think that Usenix would have too much trouble finding students to help organize such a venture, and I’d be tempted to volunteer myself.

On a more broader note, I feel like Usenix currently does a great job in these areas:

• Technical research: Usenix ATC provides a forum for the presentation of top quality academic and industrial research. I consider it a great venue for any type of general systems work with strong technical components.
• Mixing industry and academia: In my (relatively limited) experience, Usenix ATC is the conference with closest to an even match between academics and industry professionals. This is good since both sides need the other, but in most other conferences I’ve seen their is a clear majority in one direction or the other.

Other areas that Usenix could expand on to better support students are:

• Graduate student development: offer tutorials or seminars on topics like research methods or personal organization (ie systems like GTD). A professor in my department teaches a research methods course which was incredibly helpful for me, and I know he has given 1 hour talks on the subject at other schools to rave reviews. These are the kinds of things that graduate students currently are learning on the job through trial and error, and it is much better to just have them taught to you upfront. I’m not sure how well this would fit at something like ATC, but it would definitely be ideal for a student conference, and even just lists of online resources could help.
• Insights into academia: this would include things like organizing student run conferences or shadow PCs that allow students to get a better idea of what keeps their advisors busy when they aren’t meeting with us. Learning how to review papers helps us become more critical (in a good way) of all the other papers we read, letting us get more out of them than we would otherwise.
• Realtime research updates: I wish I had a list of blogs written by systems researchers. Usenix could help organize this by at least setting up a list of links to all blog posts written about their conferences (you can start with my notes from HotCloud!). I want to know what other researchers are thinking about, and I also want to be updated whenever people in my area publish new pieces of work (currently I rely on elaborate mechanisms that automatically check the publicaion webpages of the top people in my field to see if they change each day).  Obviously for this to be fully useful, it needs to support more than just Usenix conferences and workshops, and the updates need to be propagated when papers are accepted, not four months later when they are presented.  Usenix’s push into social networks may help with this too, although I’ll admit that I haven’t “friended” Usenix yet, so I’m not sure…

That’s all I can think of for now, and I’m still on east-coast time, so I need to get to sleep.