Monday, January 28, 2008

On Scheduling Conferences

One travels to SODA or STOC or FOCS on a Saturday and returns Tuesday night or Wednesday AM, and in the process, loses the weekend to the conf; include the monday holiday as well in the case of SODA 08 due to MLK day, and include friday evening too if you attend the workshops that precede the confs.

Why do we do it?

Why do we volunteer and take weekend off from our families, our personal and social lives, and do what is most certainly an important part of our work, either as a univ professor or as a researcher in an industrial lab, during the *official* non-working times? It can't any longer be for cheap saturday night stayovers. Avoid conflicts with teaching because our teaching responsibilities triumph our research?


Blogger Luca said...

Hear, hear!

6:08 PM  
Anonymous Anonymous said...

Alas the vote at the SODA business meeting to have SODA start on Mon instead of Sun failed by a wide margin...

At least the paper proceedings should be gone next year

6:26 PM  
Blogger Michael Mitzenmacher said...

There really is no good time. Missing class is certainly a problem with holding the conference during the week. (I've been to conferences that run M-W or Tu-Th; lots of people fly in for the day to give their talk, then fly back to get back to class.) With respect to families, if you have 2 working parents, then having the conference over the weekend might be easier than during the week. (One parent having to get the kids to school and get to work and handle the kids after work is a real challenge; on the weekend, there's no school and no work, and it actually can be easier for one parent.)

This is why, frankly, more and more my choice is not to travel, unless I have to. While I enjoy conferences, if I'm not giving a talk, or not otherwise required, I'm probably not going. I don't volunteer to take weekends (or weekdays) off from my family; it's occasionally a job requirement. Travel isn't unusual for many jobs, so I cope, and try to minimize it where possible.

7:57 PM  
Blogger metoo said...

In business world, if doing X was part of the job description, then if X was done during the weekend, the company would have to find a way to compensate for it. Research is a different market I guess, where we voluntarily work during the weekend including travel out. (Well, most of us in research do not have a "working time" of the day, we work nearly all waking time.)

12:31 PM  
Anonymous Anonymous said...

The standard dates shifted in the early 1990's from a typical M-W to typical Sun-T. The biggest reason was not the missed classes; it was cost. At that point, travel dollars per NSF funded researcher had shrunk by about half in buying power of what they had been in the early 1980's. Most airlines had Saturday night stay-over fares at substantially less than Sunday fares. A number of people were already staying the extra day (with its extra hotel cost) to avoid the extra airfare expense.

In some markets that airfare difference has disappeared but it is still is a major factor in many markets.

1:10 PM  
Blogger D. Sivakumar said...

To anonymous:

That is not surprising since the people who voted were the ones who attended a conference beginning on the Sunday.

I, for one, wished to be with family on the Sunday + MLK day, so had to miss a SODA in my neighborhood, despite the many delightful topics it touched.

I agree there's no easy solution. My personal preference would be a Tue-Thu conference, with Monday and Friday for travel/sightseeing. I think most of us would be (should be?) able to set aside one week in a year for 'work-related travel' (away from all teaching/family commitments).

12:14 PM  
Anonymous Anonymous said...

> I think most of us would be
> (should be?) able to set aside one
> week in a year for 'work-related
> travel' (away from all
> teaching/family commitments).

It's obviously a claim from a research lab person. For those in academia, we can easily set aside one week or even more weeks a year *outside* the semester. Furthermore, remember that it's not just one week: you want to come to (a subset of) STOC, FOCS, SODA, WWW, EC, CC, ... so you need a few weeks. And all of these weeks should have priority over teaching? Then the students would have loooong vacations.

6:10 PM  
Blogger metoo said...

It is hard to spend a week at every conference of one's interest for not only university professors but also labs researchers. But I like the idea that everyone feels like they should be at the top (two?) conferences in their area, no matter whether they have a paper or have to give a talk. (Networking community feels that way about SIGCOMM where very few get to give talks.) This should be just as important as teaching or whatever research labs people do. :) So, explicitly acknowledging this to be part of one's job description, and going to a week long (not weekend long) conference will be interesting.

10:05 AM  
Blogger Mohammad Mahdian said...


10:28 AM  
Anonymous Anonymous said...

I disagree; in fact, I don't even think professors or most industrial researchers should recognize the whole idea of "official non-working times". We've got incredibly flexible jobs, where we can set our own hours, take time out from work during the day more or less whenever needed (as long as we are there for a few meetings/classes), etc.

If we start demanding the right not to work during "non-working hours", then eventually we're going to be forced to work during "working hours". There's no way around it, since the alternative is a "skip work when you feel like it" policy, and nobody has that except self-employed freelancers.

I'd much rather retain the freedom to set my own hours. The cost is that you'll sometimes find that other researchers choose to set their hours at a time that's inconvenient for you. This is the price of freedom.

12:57 AM  

Post a Comment

<< Home