Confessions of a researchaholic

Most paper committees I have served have purely electronic review processes. These are relatively easy. Those with in-person meetings (e.g. SIGGRAPH) are more challenging as they involve live human interactions in real-time.
Below are some of my personal experiences to make the process more fun and effective.

Emotion

The most important and yet difficult task is to remain neutral, no matter what happens. It could be quite some experience to see your paper getting rejected and immediately you have to discuss a paper you reviewed.

I have a very simple strategy that works superbly well for me so far: I just assume all my papers are (or will be) rejected, even if they have very high ratings. (Anything could happen, and has happened before.) By assuming the worst case scenario, I can never be disappointed. I also do my best NOT to track my papers; I did not even look at the status on the spreadsheet when I am outside the room. Then it is easier for me to remain cool.

It also helps if you naturally care less.
One possibility is to not have any submission, but this is not common for people who are still productive.
Another possibility is to have enough prior papers so that you care less.
The paper chairs like to recruit more senior people not only for experiences but also for this “care less” factor.

Other things being equal, it is usually better to be positive than negative. My rule of thumb is to accept if unsure. This is better for humanity; a good paper wrongly rejected will not be read by anyone, while a bad paper wrongly accepted will likely be ignored by future research anyway. This is also better for myself; I do not want to leave a reputation for being a paper killer.

Workload

It is a lot of work to review 20-something papers. You will look bad if you do not seem to know what each paper is about, especially during the plenary sessions or breakout discussions. So make sure you put in enough efforts.

I also keep enough dark chocolate around to maintain my brain function at the end of the (long) day. (OK this is probably some lame excuses; I overdose cocoa no matter what.)

I am probably lucky (or maybe I am good at requesting papers; dunno yet) in that I usually get good assignments (high quality submission fitting my interests and expertise well), so I usually know each paper well. For those few that are outside my expertise, I just admit it to other committee members and reviewers. Nobody knows everything, so honesty is often the best policy.
Plus, I guess many of you have seen reviewers who clearly have no idea what the f*** they are talking about, so try not to be such jerks.

Participation

For those of you who think the committee members have some edges in getting their papers in, you might be disappointed; as far as I can tell, no such advantages exist, and the system has been well designed so that it is very difficult to game.

However, the committee members do have advantages in organizing the paper sessions, which is also the most fun part of the committee service in my opinion. You can influence where and when papers (including yours) go and which sessions you chair. People who do not show up might find papers (they authored or reviewed) going to a strangely titled session with a motley collection of seemingly unrelated papers, or find themselves chairing sessions that are too early for many people to wake up or too late in the last day for many people to remain behind.

I stumbled upon these advices on advising by David Patterson (here) and Jeff Ullman (here) today. I found both insanely useful, especially after having a few years of advising experiences. The advices differ in the way that Patterson is mainly on systems and Ullman mainly on theory. But they also share similarities such as the importance of identifying problems and collaborative work.
(I would really love to hear from someone in graphics and HCI, fields with more emphasis on human side applications. I am thinking about asking my former school advisers/professors to write such articles.)

I reflected upon my (unusual) style of individual work and asynchronous communication. This has been quite effective so far, but I never stop thinking it could go wrong, at least for some students. So what I did is to make sure there is always at least one co-adviser who can help with the normal human side of needs, like in person meetings and emotional support.
Exposure to industry labs or even startups, fortunately, should be the benefit of working with me, given where I came from and whom I know.

I encourage all my past, current, and future students to read these articles, and let me know if you have any comments.