Thursday, February 25, 2010

STOC Budget Questions

I hate to follow the interesting conversations on FOCS/STOC/SODA (please keep commenting) with the mundane, but Lance forced me asked me to be General Chair for STOC, which means looking over things like the budget. Registration fees will probably end up being around $500 for early registration, $250-$275 for students. The numbers are still being played with.

Here's a bunch of questions that arise. I'm happy to hear input.
  1. Should all PC members' expenses for the PC meeting be paid for? That works out to, roughly, $80-100 on the registration per attendee. Hotels and airfare add up, and keep in mind the way the ACM forces us to do the budget you need to budget over 100% of the nominal cost to deal with contingencies.

    In many other areas, it's assumed you'll pay your own way to the PC meeting. For the networking conferences I've PC'ed, they cover meals, and usually have a very nice dinner after the work is done. For the theory conferences I've helped manage, I've usually aimed to cover everyone's meals and hotel (though the dinner is less nice than for the networking conferences...), and to cover anyone who couldn't fund their own travel. That works out to more like $40 per attendee.
  2. Do we really need morning and afternoon coffee breaks? The afternoon coffee break every day adds something like $25 per.
  3. When you look at fixed cost, every student who attends is actually a loss, that has to be covered from elsewhere. Is this the right way to go? (I like to think that the corporate sponsorships, from Microsoft/Google/IBM/+others, should be first thought of as going to reduce the cost of student attendance, so I think this is still the way to go.)
  4. At what point do registration fees become a noticeable concern?


Anonymous said...

First of all, $500 is really a lot! I think this is an important reason for having not so many people coming to STOC — or other conferences — without having a paper. For Europeans, if you add travel and hotel, they have to pay much more than $1000! In this way, you cannot attend many conferences in a year and you have to restrict to conferences where you have a paper.

My second remark is more precise: You write that "the afternoon coffee break every day adds something like $25 per." STOC is a 3-day conference, thus each afternoon coffee break costs more than $8 per day: with that price, I can have a nice coffee break at Starbucks or whatever. I cannot understand how you manage to have such expensive coffee breaks. At the same time, I do think coffee breaks are important because of the discussions it stimulates between us.

Anonymous said...

I don't understand why PC don't pay their own way. They are getting money mainly from NSF so it is a matter of if their NSF grants pays or if someone else's (through registration). If a PC member can not pay, i.e. does not have their own funds, then the conference can pay for this small number of people.

Also, there should be a student reg fee for academic postdocs who do not have their own travel funds. (or if they do, they typically are alotted a couple thousand dollars rather than an NSF grant).

Glencora Borradaile said...

As a postdoc, the registration fees felt like a problem. For me (and several other postdocs I know), I had a fixed travel budget - having to pay the non-student fee meant the difference between attending 2 conferences instead of 3.

However, now that I have a much less tight travel budget, paying $500 for registration really isn't a problem. Even if I didn't have a travel budget, I would be willing to pay the conference fee out of pocket for my professional development - much less painful on a professor's salary than a postdoc's.

Would it be possible to extend the student registration fee to a trainee's registration fee (thereby including postdocs)? Would it be possible for faculty (perhaps restricted to those with papers at the conference) to apply for funds to attend if they have none? Perhaps such funds could be directed towards typically underrepresented universities ...

Anonymous said...

Why is $500 alot? If you are a postdoc without your own funds, then asking someone for a large amount of money can be awkward, so it would be a lot. But if you are a prof with a 100,000+ NSF award, $500 is nothing.

Probably, there is a big divide between people for whom such a ref fee/conference expenses are negligible compared to their grants, and others who don't have money and anything would be a lot.

So for people to make comments like "$500 is alot" doesn't make sense. It is for some people and it isn't for others. The only solution would be to have a socialist-style system of paying for conferences, where people pay what they can. Alternatively, one could just get the money directly from NSF (chopping off a little from each grant) since that is where the money comes from anyway.

If it is not your money, then $500 is not alot. If you have to pay out of pocket, and you do not work at microsoft, then it is a lot of money.

GASARCH said...

PC's don't have to pay seems absurd since
thats alot of money and they would probably go anyway. Maybe a small break on some items, but they should certainly pay.

At CCC we have sometimes NOT had a banquet which saves some money.

Also, catered lunchs are not needed.

A Conference without Coffee! Perhaps you can cut down on coffee,but coffee is essential.

We really want to encourage students to go so good to keep the price down.

My 2 cents.

bill g.

Michael Mitzenmacher said...

Just to be clear, when I'm talking about PC expenses, I mean for the (face-to-face) PC meeting, not for the conference. PC members generally pay registration when they attend the conference, just like everyone else. The question is whether they should have to pay to travel to the meeting. The reasoning why they shouldn't have to is that it essentially "costs" them money after they've already volunteered their time and effort by serving on the PC; that arguably doesn't seem right.

Anonymous said...

It doesn't "cost them money". They use their NSF grants. And being on a PC is good for your resume/professional development and will lead to more NSF grants.

Why should some poor postdoc who may end paying at least some of the expenses out of his/her own pocket subsidize the PC expenses when the PC are likely sitting on a huge pile of money?

Anonymous said...

As a regular STOC/FOCS-goer and occasional PC member, my opinions:

1. Yes to covering the PC's expenses for the PC meeting, no to covering their expenses to attend the conference. Consider also not having a physical PC meeting.

2. Yes to both morning and afternoon coffee breaks.

3. Yes, it's okay for the students to be a loss.

4. Registration fees are always pretty noticeable. Once it exceeds $400 I start to get disappointed.

Anonymous said...

Addendum to the previous comment:

5. Please stop making paper proceedings!!

Paul Beame said...

There are lots of ways to save money. We did STOC 2006 for a registration fee of $330 with a reception, afternoon breaks (no food, just drinks in the afternoon since everybody is stuffed from lunch, no afternoon break on 3rd day, only 3/4 of the lunches on day 3). We also paid for all PC expenses that people asked for. (We ask people to pay for what they can; e.g. industrial PC members can get expenses covered by their employer but the typical theory grant is way less than that for networking.) Rather than just use the hotel A/V quote we got an outside quote and the hotel dropped their fees by 1/2.

In general, we were very aggressive in our budgeting by lowballing costs and being optimistic about how many people would register late and pay higher fees. Despite all this,
when we were done we ended up with a $10K profit that we didn't want to earn!

Though Boston is a more expensive city than Seattle, Philadelphia is similarly expensive and Philadelphia FOCS earned a very sizable profit at roughly $400 for early registration. My advice is to set registration fees as low as you can. You will probably end up with a surplus even when you are trying your hardest to avoid one!

David Andersen said...

1) As someone who's crossed from the student side to the faculty side, I'm a little shocked at the implied perception from a lot of the commenters that money is free. Yes, I have $100,000+ NSF grants. Yes, I still decided not to proactively send all of my students to SIGCOMM this year because of the combined expense of the conference + travel to Europe. I think many people have no idea how little money $100k is -- most of that goes to paying the tuition + stipend for one student, leaving only about $5--10k for travel.

That said: I agree that PC members should pay their own way, unless they can't. In which case, I suggest that the conference should take a similar view as they do towards presenting authors of papers who can't attend - help 'em out. Keep the dinner, though - it's a valuable opportunity for conversation and a nice "thank you" for contributing to the community.

But I do strongly favor trying to co-locate the PC meeting with something else the PC members are likely to be involved in. Or do what Tom did for SOSP this year -- host a "mini-workshop" along with the PC meeting. Benefits the host site, benefits the PC members, amortizes travel costs.

2) Sure, kill the food at one coffee break. Who cares? You'll lose more people heading to starbucks, but people are flexible. I might keep the coffee and ditch only the food, though.

3) I strongly favor keeping the financial bias towards student participation.

4) The real concern is the total cost, not individual components. I'll try not to rant & rave too much, but as one example, having SIGCOMM outside of the US two years out of three has imposed a financially significant barrier. $2k to attend a conference is a lot aggregated across several students. $300 vs $500 for attendance for a three-day conference? Who cares.

Michael Mitzenmacher said...

Paul --

One problem I found in local arrangements/budgeting is that the ACM insists on a HUGE cushion. If you tell them 300 people will attend, they'll insist you budget for 270. They won't let you count corporate donations in the budget unless you have the check in hand. They even build a 15% surplus into the budget! This makes it hard to
"be very aggressive in our budgeting by lowballing costs and being optimistic about how many people would register late and pay higher fees." We certainly try, but we tend to get pushback every step of the way...

Paul Beame said...

For years SIGACT has arranged for the 15% cushion to be waived for STOC. You should get Lance to do that.

I understand about the checks in hand but you can re-budget as you get them (and confirming e-mails are often enough at budget time). As for the numbers, you can change everything by just changing the ratios between students, late registrants, etc.

Anonymous said...

If you use this year's surplus for next year, and last year's to reduce costs for this year, this could cancel. So, is a fixed % surplus really a problem?

Anonymous said...

$500 is really a lot. As a postdoc, with a small fixed travel budget, I could not afford to submit papers to some conferences because of registration fees. Postdocs also do not get professional expense reimbursement for joining ACM SIGACT.

It is reasonable to ask PC members to pay for their own travel, if possible. Tougher negotiations with the hotel might be another way to push down the costs.

Anonymous said...

"I'll try not to rant & rave too much, but as one example, having SIGCOMM outside of the US two years out of three has imposed a financially significant barrier" said David Andersen.

Now, make the efforts to imagine the problem for people not living in the US. Almost all the big conferences (such as FOCS and STOC) are located in the US. So it is a big deal to go to a lot of conferences. So some efforts from the PC members to decrease the cost are really welcome (I am not saying at all that Michael and the other PC members this year do not make efforts!).

Another point is that you are lucky to have some generous grants from NSF but it is not the case everywhere in the world... I am not complaining about my grants because I can do almost everything I want, but to do so, I have to be very careful on my expenses.

I know this is not the same conference but as an example, organizers of CCC09 managed to have registration fees around $350 (it was €250). I find it more reasonable.

Anonymous said...

Sometimes postdocs are PC members :-)

Anonymous said...

It is unacceptible not to offer reimbursement for travel expenses to the PC meeting. This is the absolute minimum requirement. Often PC members choose to pay themselves for whatever reason, but the decent thing to do as a community is to offer to pay for travel to the meeting itself. PC members dedicate a lot of time to the reviewing process, and travel to the PC meeting is not a fun trip.

Anonymous said...

Re: contributing to the expenses of PC members.

If you want to see junior people in PCs (like post-docs), you should be prepared to cover at least a part of their expenses.

Some academics have lots of funds, but most doesn't have so much. One grant brings little travel money which is difficult to commit to the PC task especially if one expects PC members to come to the PC meeting and to the conference.

Having physical meetings is important IMHO. We have all these claims about a bias in the decisions of PC; the bias would be significantly bigger if the meetings were done electronically.

Anonymous said...

"It doesn't "cost them money". They use their NSF grants. And being on a PC is good for your resume/professional development and will lead to more NSF grants.

Why should some poor postdoc who may end paying at least some of the expenses out of his/her own pocket subsidize the PC expenses when the PC are likely sitting on a huge pile of money?"

Even if it's not 'huge', it's still more money than others have. A little thought for others would go a long way...