Friday, January 09, 2009

Conferences Abroad

Samir Khuller guest posts about having conferences in exotic locations.

During the SODA Business Meeting I was a bit alarmed at some of the suggestions made for where the conference should be held in future years—some of the locations that even received substantial support were Paris and Buenos Aires. I have nothing against these locations, in fact Paris is one of my favorite cities to visit and Buenos Aires is high up on the places I want to visit. (I also want to visit Chile and Antarctica!) I wanted to make a few remarks about exotic locations.

  1. The cost of sending students to conferences is already non-trivial. At least with NYC the travel costs are low for many students between DC and Boston. For faculty who are trying their best to stretch limited travel funds to pay for several students, moving the conferences to exotic locations makes life quite difficult. I would like to see major US conferences held in places where there is a substantial presence of students—Boston, Bay Area, NY, DC to name a few. We should move the conferences around, so that each student has an opportunity to attend a major conference in their area during their studies. If someone made a convincing argument about the benefit to Argentina theory grad students that would be an argument for Buenos Aires.
  2. At a conference I spend 90% of my time indoors. For the 2 hours I spend outdoors each evening, it almost does not matter where one is. There are nice cities in the US to visit and NY, DC and San Francisco have emerged as popular locations, drawing a big audience. If we moved the conference or even alternated between the east and west coasts, that would be terrific.
  3. There are visa issues for students on F1 visas making it tricky for them to attend meetings outside the US. For some students getting a visa involves them visiting their home country first, since they cannot get a re-entry visa to the US from some random country. There are already a lot of European theory conferences (SWAT, ICALP, ESA, STACS) to name a few. There are fewer conferences in N. America (STOC, FOCS, SODA). We had one STOC in Greece and one FOCS in Rome already. The other theory meetings such as PODC, PODS, LICS, SPAA already move around between Europe and N. America, why are we then taking our major conference and trying to move it to locations that make it harder for our students to attend?
Finally, if people suggested cities after checking prices at the major hotels that would be useful. While Paris has some reasonably priced small hotels, it could easily be that the large hotels are really expensive. We could then cast a more informed vote. The SIAM Disc. Math Conf. in contrast has extremely low registration fees and is usually held at a Univ campus during the summers.

Thanks to Moses Charikar and Sudipto Guha for comments.

43 comments:

  1. There are visa issues for EVERY conference held in United States. I would sugguest all conferences held in China, since we have the most number of grad students.

    ReplyDelete
  2. I am all for having conferences primarily in or near major US cities. Attendance (local graduate students) is, as you've pointed out, a major issue. Another major issue you've left out is the difficulty in getting to a location. If it takes more than a few hours to get there -- a flight with a stopover -- I'm much less likely to go (unless I have to). I'll do a non-stop coast to coast but that's my preferred limit.

    I wish our community would do more fixed-location conferences (or, if people really can't deal with that, going back and forth every other year between two locations). Consistency in location leads to consistency in administration and quickly to a local maximum in conference niceness. If you want to make sure to have variable locations, have one conference fixed on the east coast (during the summer) and one on the west (during the winter)...

    ReplyDelete
  3. Remember that US is not the entire world!! The exact same problems that you mention for holding conferences in "exotic locations" exists for conferences held in the US from an outsider's point of view. For a European faculty, for example, it is expensive to send students to the US, and there are serious visa issues that are quite problematic for international students intending to travel to the US for a conference.

    So I don't think your arguments are valid, unless you see the US as the only place in the world where theory exists, which is absurd. I also don't see why you associate STOC/FOCS/SODA to the US or North America. Aren't they supposed to attract an international audience? Otherwise they should get names of the sort "American Symposium on Theory of Computing". While it's understandable for "European Symposium on Algorithms" to be held in Europe, an international conference should ideally move around each year to keep its international audience, giving everyone equal opportunities. This is remarkably seen in, say, information theoretic conferences like ISIT (see http://mybiasedcoin.blogspot.com/2008/07/isit-this-week.html) which move around quite a bit each year.

    ReplyDelete
  4. First of all I would like to point out that Samir is probably keenly aware of the fact that the US is not the world, after all according to his webpage he got his undergrad degree in India.

    I can't exactly speak to the meeting that Samir attended, but I suspect that rather then having potential hosts from Paris and Buenos Ares coming forth to point their name in the ring for where to hold the conference, it was people who aren't located in those places making the case for them, which makes a difference.

    Also sadly you have to take into account that for established conferences, at some point they inevitably stop target the broadest possible community and instead focus on the community defined by regular attendees.

    And to Michael's comment, while repetition allows for refinement, most conferences are organized by academic faculty and it's a non-trivial burden to them so you're probably not going to find too many volunteers to permanently host a major conference. Of course that begs the question is there a potential solution to that problem, and I can think of at least one. Instead of having conferences organized by active faculty, is to shift to a culture where conferences are organized by the semi-retired members of the community.

    ReplyDelete
  5. STOC/FOCS in Greece/Rome had record attendance, and while there are conferences in Europe as you mention, they are not comparable to STOC/FOCS/SODA in strength and quality. And I definitely worry about visas for foreign grad students.

    That said, despite my excitement about foreign locales, what Michael says about staying in the same place or alternating between a few sounds very reasonable, and NYC/SF/Vancouver (a foreign location)/New Orleans seem like a good choices. Definitely not any major city in the US works. FCRC in San Diego is horrible - there is absolutely nothing to eat near that hotel that's not a chain restaurant, you need a car even for most of the chain restaurants, and there is no decent coffeeshop (Starbucks does not qualify). These are the minimum requirements for me - good cheap eats that aren't chains, good public transit, and decent espresso. The SODA in Baltimore had a a similar problem to FCRC, and Las Vegas just offends my sense of humanity (why do we keep suggesting it?).

    But let's be honest, it's clear people are looking for SODA to be in a place where they can combine it with a vacation. Is this so bad? Work hard, play hard...

    ReplyDelete
  6. Anon 3:

    The US still sends most people to the conferences. While it may make sense for reasons other than economic to occasionally have a FOCS/STOC/SODA in Europe or elsewhere, from an economic/game theoretic perspective (both in terms of maximizing attendance and minimizing cost), we'd probably be better off just having US people subsidize Europeans to come rather than paying to travel to Europe ourselves. (For those who LIKE traveling outside the US for conferences, there's still ESA/STACS/etc.)

    Anon 5: Most repeated conference I know of are taken care of faculty who let their staff do much of the prep work. (I'm thinking of DCC (compression), Allerton, and now the San Diego information theory conference.) The faculty mostly handle the inviting and small amounts of administration.

    I would think, for example, that MIT could easily manage among its 20 or so theory faculty to divide the faculty-level administrative tasks with low overhead and derive substantial benefits from hosting a conference regularly. There would have to be buy-in from the department that the benefit to the faculty is substantial enough to devote administrative resources to it. Again, this is what Allerton (and others) have already done, and it seems a workable model. I agree that you need a faculty concentration to make this work, suggesting that Boston, New York, and San Francisco/Bay Area are the natural places to hold repeated CS conferences. (These places already often have specialized mini-conferences -- theory days -- is it really so hard to imagine they could manage annual conferences?)

    ReplyDelete
  7. If you're thinking about travel costs, you're right that the cost of travel to NYC is low for a large number of people. But NYC hotels are very expensive, and the "travel costs" should really be thought of as the cost of hotel plus the cost of air/train/bus/gasoline/other transportation.

    From a travel point of view, then, it makes sense to head a hundred miles south, to Philadelphia. (But I am biased, since I'm from Philadelphia and currently reside here.)

    However, not being a CS person (I'm a mathematician), I can't intelligently comment on whether it would be possible to find local organizers.

    ReplyDelete
  8. Didn't they say that they spent 100,000 on SIAM so that SIAM could organize SODA? If you gave someone a postdoc at a good university and their only responsibility was (besides research) to organize a conference, this would seem like a good deal for them and $100,000 could probably pay for them.

    I think the whole vote on location was disorganized because the community (including people not attending that particular SODA) should have agreed upon goals (student attendance, low cost, etc) and then the places should be chosen with those goals in mind. Also note that the Virgin Islands was nominated and there do not seem to be a high number of grad students there, so that definitely has nothing to do with helping student attendance. That was purely just a "good vacation" place and I do not think everyone would agree on this criteria as one of the goals.

    Since SODA is always held in January during the winter break of many schools, it does not make sense why it is never held at a university, which would be much cheaper.

    ReplyDelete
  9. I believe the other top contender was the Virgin Islands, not Buenos Aires. The US Virgin Islands is part of the US, so there are no visa issues. Also, the top contender in the continental US was San Francisco, and I would like to point out that NYC<->SFO is actually a more expensive flight in January than NYC<->Virgin Islands (I have verified this on orbitz.com, airport code for the Virgin Islands: STT).

    ReplyDelete
  10. But let's be honest, it's clear people are looking for SODA to be in a place where they can combine it with a vacation. Is this so bad? Work hard, play hard...


    Yes, it's definitely bad. First, it's sleazy to try to arrange things so as to get a nice vacation paid for by grants. A lot of people outside of academia are convinced that conferences are really just junkets for professors, and this sort of thing reinforces their belief (and ultimately makes it a little more difficult to get travel funding).

    There's also the issue that many people don't want or can't afford such a vacation. Requiring them to invest the travel time and money anyway creates a real burden. I'm worried about excluding people, not matter how many other people love the vacation aspect.

    Las Vegas just offends my sense of humanity (why do we keep suggesting it?).


    One good reason is that it is the most travel-accessible US city by a large margin (as measured by air fare, convenience of flight schedules from many locations, availability and price of hotel rooms and food, etc.). Plus it's a favored vacation destination for many people, although I don't consider that a compelling reason. Pretty much the only serious drawback is that there's not a strong local research community, but that's not a huge issue.

    I share your feeling of horror (I find Las Vegas garish, tacky, and more than a little embarrassing). However, the location really has a lot going for it.

    ReplyDelete
  11. Hear, hear. I am a CS graduate student whose advisor typically does not have any money to pay for travel, and it is getting harder every year to scrounge up $2000 if I have a publication that requires spending that much money just to go present it.

    In this one area, I find myself in envy of the mathematicians, for example, whose publication record is evaluated solely on journals. This means they can send their paper to the most prestigious journal they think will accept, and are free choose to attend conferences based on matters like money/proximity or other factors, without (directly) harming their career if they don't attend. But we have a more difficult decision, needing to weigh the fact that the best conference for our paper, in terms of impressing a hiring committee, may also drain a large percentage of our annual salary if it is located in some exotic setting chosen by professors making >$100,000/year.

    Not that I mind the idea of researchers who make such salaries traveling to great places so often; I hope to do it one day myself. But it is not fair to then turn around to grad students making less than $20,000/year and tell them, "If you want to impress us enough to get hired into our departments, you better figure out a way to get here also." I understand the historical reason why our field does this (http://weblog.fortnow.com/2004/07/why-are-cs-conferences-so-important.html) but I don't have to like it.

    I'm American so this is an issue if I have a good candidate paper for ICALP, say, but the same applies to a student in Europe or Asia who has a good candidate paper for STOC or FOCS or SODA. From what I hear, the visa problems make it even harder. Similarly, my non-US colleagues studying in the US lately have a hard time getting back into the US after attending a European conference, like they might have lied about the conference to go to a Taliban training camp or something. Anyway, it's really dumb that these are the issues I think about when I'd rather be proving theorems, and it could be fixed if we all agreed, starting tomorrow, to place all the prestige emphasis on journals and take it off of conferences. </rant>

    ReplyDelete
  12. The isolationist attitude of "let the Europeans go to ICALP and ESA" is ridiculous.

    Where is our allegiance? To our grants and tenure cases, or to science? The former may be the more common answer, but it is also the deplorable one.

    If our allegiance is to theory, then we should welcome the European students (of which there are many), and give them the opportunity to attend big conferences and see the cool stuff that's going on. We should recognize the great potential of Europe and help channel it towards the kind of work we're interested in.

    There are many causes why most of the European theory research is currently so weak, but the poor quality of conferences like ESA, SWAT, ICALP, etc seems to be one of the leading ones.

    ReplyDelete
  13. You must have been joking, right? How is Paris an "exotic location"? It's amazing how US people seem to believe they are the entire world.

    ReplyDelete
  14. Let me clarify one thing - I certainly
    do not think that the US is the entire
    world! However, I do think of SODA
    as primarily a N. American Conf. being started by ACM and SIAM.

    ESA is an EATCS meeting and
    I have never heard of them voting to move to the USA (I have also attended a couple
    of ESA's and could not attend several,
    even when I had papers, since I was
    paying for my student co-author to go
    and could not afford both of us going
    to the conference.)

    I like the SIAM Disc. Math model - I went to the one in Toronto. They had
    student dorms where I could stay for
    $30/night. I used Freq. Flyer miles
    to get a free ticket, and the Reg.
    Fee was at most $100 if I recall
    correctly. If they can do it, we
    can do this as well.

    I like to "party hard" as well :-)
    and love traveling, but if I am
    going alone (without my family) to a conference I usually dont have time
    to do anything else. At ESA a few
    years back (Majorca) I flew in early on Mon morning and left on Fri morning
    (the last day of the conference). No
    time to look up Nadal and play some
    tennis with him.....

    There are other good regional conferences as well -- in India
    there is FST&TCS, which I was able
    to attend in 2007 since I was going
    home to visit family anyway. I thought
    the talks were very nice, perhaps
    the quality of the papers was slightly
    below SODA level, but the meeting
    was well attended with 125-130 people there. I heard several nice invited talks and got to meet several researchers who I normally dont run into frequently.

    Samir Khuller

    ReplyDelete
  15. Holding the conference in the same place(s) (say NY/SF). Many pros --
    - familiarity, easier organization, cheaper (though I'd argue with it).

    Holding the conference somehwere new -- travel costs, organization, etc.

    So what about a cycle?
    NY, SF, XX, NY, SF, XX

    where XX would presumably be elsewhere -- Europe? Asia? etc.

    That way we get the best of all worlds.

    -ro

    ReplyDelete
  16. Anon 15:

    If it wasn't clear, I certainly view the "cycle approach" [NY/SF/XX] as perfectly viable. Though I'd slightly prefer Boston to NY. :)

    ReplyDelete
  17. It's much easier to find a cheap (non-conference) hotel room in NY than in Boston. Also, in winter, Boston is clearly worse than NY. (Even flying into Boston is slightly more expensive than NY.)

    ReplyDelete
  18. A good conference location is one satisfying the following conditions:
    1. Inexpensive and convenient public transportation between airport and site
    2. Inexpensive lodging and food
    3. Easily accessible by plane for a large number of attendants
    4. Should ideally be located near a CS department hosting it.
    5. ...

    Portland, Oregon would be a good city. SF, NY, and Boston would cost you more (unnecessarily) for 1 and 2.

    ReplyDelete
  19. SF, NY, and Boston would cost you more (unnecessarily) for 1 and 2.

    Did you mean for 2 and something else? For 1, all of SF, NY, and Boston have cheap public transportation to and from the airport and central city.

    ReplyDelete
  20. Yes, thanks. I was thinking of cost of living in general, which would drive up the conference cost for both participants and hosts. Why pick an expensive city?

    ReplyDelete
  21. I like the idea of holding STOC/SODA/FOCS outside the US once every few years, it helps to better integrate the theory community inside and outside the US.

    How about asking for some European conferences (e.g., ESA) to be held in the US in those years?

    ReplyDelete
  22. FYI, I have it on good authority that the ICALP steering committee is seriously considering bringing ICALP to North America.

    Several folks at SODA said that they would only vote for SODA to go to Europe if ESA comes to the US. Seems like a fair compromise to me.

    ReplyDelete
  23. Unless we're willing to rename SODA as ASODA, I don't see how to convince ESA to come to the US.

    As someone who's organizing a major theory conference in two years, I can tell you that it's a thankless job, and if I had been at SODA, I would have continued to support letting SIAM do the organizing.

    It's worth pointing out that SODA had an approximately fixed east coast/west-coast plan for a while. It's not NY/SF, but the intent was the same (ensuring that people from both sides, and Asia/Europe) find it easier to attend.

    I agree with Samir's point about having no time to do "sightseeing" during a conference: since SODA is more often than not during the start of spring semester (or winter quarter), it's not like we can take extra days off and cavort around. I don't think there's any harm in thinking about nicer locations, but it's not realistic to choose locales like the Virgin Islands, or even Puerto Rico.

    ReplyDelete
  24. When you're a grad student in New Zealand, virtually every conference is in an "exotic" and expensive location... :-(

    ReplyDelete
  25. Contrary to Nicole's comment, neither STOC/FOCS in Greece/Rome had remotely close to record attendance. The records for both conferences were much higher - over 500 for both (see some old STOC numbers). Both of the European FOCS & STOC conferences did not do well financially - I believe that STOC lost money.

    One issue that used to be the case (but may no longer be true) is that USA->Europe->USA airfares were substantially greater than Europe->USA->Europe airfare for the same flights.

    Being held in January when at least a few North American academics do not have classes means that moving SODA to alternative locations may create less conflicts than moving FOCS/STOC. Paris in January? Why choose a location whose residents would prefer to be elsewhere at that time of year? Either skiing or sun at that time of year seems far preferable in choosing a location.

    ReplyDelete
  26. I feel the pain of the person in New Zealand. However, that is the reason many regional theory conferences started - FST&TCS (India), ESA (Europe), ISAAC (Japan/HK/Korea...), CATS (Australia/N. Zealand). I view
    SODA as the N.America regional meeting. Where should our students go if SODA were to move far away?

    In the 80's there was FOCS and STOC in the US, and ICALP in Europe. SODA is about 20 years old and ESA is 17 years old so both have had time to establish themselves.

    I too have the impression that
    STOC in Crete lost a lot of money
    (confirming what Paul said).
    BTW, FOCS 1991 was in Puerto Rico
    and my impression is that the attendance was low (much much lower than FOCS 1990 and FOCS 1992). I was not there, so am not 100% sure.

    ReplyDelete
  27. FYI, I have it on good authority that the ICALP steering committee is seriously considering bringing ICALP to North America.

    I can confirm this. The EATCS Council is seriously considering this possibility as well as the option of having an ICALP in Far East Asia in the next few years.

    There are many causes why most of the European theory research is currently so weak, but the poor quality of conferences like ESA, SWAT, ICALP, etc seems to be one of the leading ones.

    Mihai, do you really think that writing "European theory research is so weak" gives a fair picture of theory research in Europe? For what it is worth, I believe that there are many very strong researchers in TCS in Europe, just like basically anywhere else in the world. We are lucky to live an era in which high-quality research is being carried out throughout the world. We have Europeans working in North America (you, for instance), Asians working everywhere, Americans working in Europe etc. This is what makes research a truly international enterprise.

    Also, several of the readers of this blog served on the PC for ICALP 2008. It would be interesting to hear their opinion as to how weak the programme for the conference was.

    ReplyDelete
  28. Luca asked:

    Mihai, do you really think that writing "European theory research is so weak" gives a fair picture of theory research in Europe?"

    Unfortunately the answer is yes -- CS theory in Europe is not comparable in any way with that of North America. Of course there is some good work going on in Europe, but not nearly as much as you would expect by its size, economic strength, or past accomplishments.

    Such a statement is certainly a matter of judgment, but the effect is so strong that it is hard to not see: you may try to count various things (citations, papers in top forums) or just think of your favorite results from the last 10, 20, or 30 years, and see where they came from in terms of Geography.

    Interestingly, many of the important North-America results are by European nationals (British, Greek, Hungarian, Italian, and so on). This is not just a matter of standard of living: why didn't all these stars (as well as those from India or other places) choose to work in Germany, Britain or France instead of the US?

    Maybe here would be a good forum to discuss why this is so and how it can be amended?

    ReplyDelete
  29. Isn't the perception about the poor level of theory in Europe due at least in part to the fact that not the same areas of theory developed as in the US ?
    I can imagine my European colleagues musing about the sorry state of lambda-calculus and linear logic in the US...

    ReplyDelete
  30. There is another factor to consider before moving conferences outside of the US - the jet lag. It is hard to concentrate on a talk or a discussion when all you can think of is a nice comfy bed. And the issue is not symmetric: going East (as in US ->Europe) tends to be tougher than going West.

    Piotr

    ReplyDelete
  31. why didn't all these stars (as well as those from India or other places) choose to work in Germany, Britain or France instead of the US?

    Actually, I think that the most striking example is Italy. There are plenty of very strong Italians in TCS, but almost all of them work in the US. This may have something to do with the salary system in this country: the salaries at the beginning are extremely low (below the poverty line), and then just "automatically" grow with time. This is attractive only if you are old. As a result, the country is strong only in the fields that went out of fashion 30 years ago...

    ReplyDelete
  32. Why do we pay so much attention to the venue where a result is published, and less so to the utility of the result itself?

    If the main point of the conference is to inform people of ongoing current research on a topic and what the new ideas are, then if funding is tight its perfectly fine to publish your work in whatever conference happens to be easier to get to: ESA/ICALP for Europeans,
    STOC/FOCS/SODA for US/Canada etc,
    LATIN for S. Americans, ISAAC for E. Asia. For those with more travel money, there are more options to submit and they have the luxury of interacting with people all over the world.

    The other option would be to have ONE big meeting (like the Math. Programming Symp) that we have once a year, and move it around. The meeting should be large enough that anyone wanting to present their work should get the opportunity to do so (with some basic filtering so that the results chosen are reasonably interesting). The conference could have many parallel sessions - MPS does have that feature. It could last one week. Instead we have actually splintered our community by creating a large number of new meetings -- SODA, APPROX/RANDOM, SoCG, PODC, ESA, SPAA, WAOA, WADS, SWAT, CAAN etc. These are all conferences on algorithms.
    The number of edges in the interaction graph are a lot more when we have one large clique, rather than several little cliques.

    ReplyDelete
  33. Reading the comments of many of you: perhaps Las Vegas is the most reasonable location. Flying there is rather cheap, lots of flights, accommodation is inexpensive (for example, for the next week I found even a few good hotels below $100), and since many of you don't care about the location ;-) we should do it every year in Vegas. Still, whenever Las Vegas is brought as one one of the locations, most of the potential participants complain!

    As for Europe: I don't believe one should do SODA/STOC/FOCS in Europe very often, but doing it once every few years wouldn't hurt anybody and could help the community to reach over to Europe. Same can be said about Asia. Even if most of the community is in North America. Note that SoCG, SPAA, PODC, WWW, SIGMOD, etc, all have non-North American locations every few years.

    As for STOC in Greece: if they have some losses, then it's mostly because they had huge local overhead (eg hotel registration was arranged through some Greece travel agent - made it more expensive). However, they did it together with ICALP and SPAA, and they ended up with very good attendance.

    As for FOCS in Italy: it has very low attendance. It was considered to be very/over expensive.

    However, both these conference (STOC/FOCS) have typically very low attendance from Europe; SODA is different and there is larger interest in algorithms in Europe.

    For those saying that one could do it cheaper on the campus or using help of academics: many conferences do so, but then there is often a problem how to find volunteers to organize it.

    ReplyDelete
  34. Is it a problem with pride/status/prestige that academics don't want to be associated with the sin city? Might there not be a chance that this conference location could inspire new research problems in game theory?

    One vote for Vegas!

    ReplyDelete
  35. With the SODA model of paying SIAM to organize the conference, Las Vegas should work well and there are non-stop flights from lots of places. The exact timing is critical for getting those cheap rates, though. There are few local people in the field. For FOCS/STOC Larry Larmore has organized things twice in Vegas but how many times can we ask him to do this (given some unwillingness to pay the extra $75-$100 per person for professional organizers)?

    The last time FOCS was in Vegas in the Tropicana we were right beside a lingerie show - there was very little space outside the rooms and the models walked from the elevators through the crowd at the break and I still recall the "and now #235, a white lace teddy" blaring out louder than the conference speaker during one of the talks.

    ReplyDelete
  36. Was STOC / FOCS / SODA ever held in Israel?

    ReplyDelete
  37. We should note that LV and New Orleans both have certain undesirable aspects. I was told by several people that New Orleans is not very safe any more (post Katrina) and to stick to the main touristy areas. However, the main touristy area (Bourbon St.) is
    quite tacky to say the least.

    I suspect that the women at the conference would find both locations quite unpleasant. Maybe LV has areas that are away from the
    "shady" areas. From that point of view, SF, NYC and DC are all better options. LA has been suggested several times, but never won a vote. The weather there in Jan is quite nice, but perhaps its expensive there? CA has other nice towns, like Santa Barbara, but
    someone would need to organize a bus service from LAX for attendees to get there easily. If we could get a local organizer, the shuttle service cost can be rolled into reg. fees. After all, people get to Dagstuhl as well!

    Israel: given the density of theory people there, I am surprised that a major conference did not emerge there. ISTCS started many years back, but then appeared to die out. I am not sure why. It had the potential of becoming a major theory conference. The weather in Tel-Aviv is quite nice all year.
    (There was an ESA there recently, and I am sure Eilat has hosted other conferences as well.)

    ReplyDelete
  38. KDD 2008 was held in Las Vegas, so soda/stoc/focs organizers can contact KDD organizers to see how they got around the negative points of having a conference at the sin city.

    ReplyDelete
  39. 1) Conferences should try to
    be at a place where there
    are people interested in
    going LOCALLY who often
    do not get the chance.
    (e.g., COMPLEXITY in Prague
    attracted 12 people from
    Prague).

    2) Complexity does Europe
    every 3 years. This makes
    sense since there is no other conference covering
    the same material and there are complexity theorists in Europe.

    3) IF SODA and ESA cover the same material then there would seem to be no
    reason for either to change
    locales. If they overlap but there are some differences then a swap might be reasonable.

    4) ICALP and STOC are diff enough that a swap might
    be reasonable--- though
    if GREECE STOC did badly
    then we should find out why to avoid what they did wrong (if we can)
    OR to scrap the idea
    (if we can't).

    5) Side note- Co-locating
    is a good idea on a small level- COLT and ML have done it, STOC and COMPLEXITY have done it.
    This is diff then FCRC
    that has other pros and cons.

    ReplyDelete
  40. Some persons are talking about Israel as if nothing is happening at this moment. Don't you see the Children and women of Gaza killed under the tons of American bombs dropped by American F16 jets. I think academics should do something to revert the course of history. Political leaders are too corrupted to do any thing to stop genocide in Gaza. What is happening there if far more important than any other thing. Until now there were more than 4000 wounded and more than 900 killed. How many innocent people should be killed before people begin to act to stop the tragedy. The average American people are mis-informed about what is happening as large medias are largely pro-Israelis, but you academics are more informed than average Americans and you have the duty to do something about it.

    ReplyDelete
  41. " ... I think academics should do something to revert the course of history. Political leaders are too corrupted to do any thing to stop genocide in Gaza."

    Not every politician is corrupt. The UN human rights commission has (justly) condemned this violation
    by an overwhelming vote. With an American veto in the UN security council not much more can be done at the moment.

    ReplyDelete
  42. I can't tell if Paul's second paragraph in #35 is meant to be for or against having STOC/FOCS in Las Vegas.

    ReplyDelete
  43. Regarding boycotting Israel - without debating who's right in that conflict - do you have any idea how many innocent civilians were bombed in Afghanistan?

    http://en.wikipedia.org/wiki/Civilian_casualties_of_the_War_in_Afghanistan_(2001%E2%80%93present)


    Do you think STOC / FOCS should not be held in the US ? I find such reactions to be very hypocritical.

    ReplyDelete