Friday, October 26, 2007

THANKS to Nicole's FOCS blogs and her positive outlook

I want to extend a warm thanks to Nicole Immorlica for Guest blogging from FOCS. So I will:
(Would have done this yesterday but if I had delayed getting the WOLFRAM PRIZE information out there I would have gotten at least 10 more emails telling me to post it.)

The one thing that struck me the most about Nicoles FOCS blogs is the positive outlook usually missing from discussions about theory. If you look at this blog, other blogs, or just talk to people in theory, you usually read or hear negative comments like these:
  1. FOCS is biased
  2. STOC is too expensive
  3. Theory is underfunded
  4. Australian actresses are plagiarizing my quantum mechanics lectures to sell printers.
  5. Back in the good old days people worked on important problems. Now they just get incremental results.
  6. Bring back Lance!
Some of these complaints are certainly valid. But hearing or reading such negativism can be monotonous, and blind us to some positive developments. Hence I appreciate Nicole's positive outlook on FOCS and Theory in general. I hope it lasts her entire career.


  1. Are you kidding Bill?

    Theory community is too sweet talking. Check out your comments from the last two years. If you take out, anonymous comments, mostly you would get what Nicole wrote for FOCS. Sweet talk. Mostly obvious.

    Why so? Why the theory community is so intimidated. Why sweet talking is rewarded more true talking?

    This blog is a good example. Any truth talking happens under the "anonymous" name. Another example is this FOCS coverage. Take out the slides from the business meeting and the Knuth award, there is absolutely no coverage. Every word of this coverage can be copied for the next FOCS.

    I was there. Things do not matter to me. But I wrote what I heard as opinions from others. I was not even there when coffee was missing. I was not even there for the breakfast. I heard it from the other folks. I did not sit in the back. I heard it from the other folks. The lunch was sufficient in quantity for me. I heard it from other folks who went hungry. The hotel rooms was like many other hotel rooms I have stayed in. But I have heard from other people who told me that it is the best room they have stayed in. People told me that the breakfast on the seventh floor was awesome.

    How come I heard so many things but these things are not gotten reported in the coverage?

    Because you love sweet talking. Because we all love sweet talking. Well why do not you simply disable the anonymous option from your comments. I guarantee you that all you will hear is sweet talking. It is typically easy to get people talk sweet. It is hard to get people talk truth.

    If you love sweet talking. Take it, here is my share:)

    Bill Gasarch is more generous than all other Bills:) The theory community always look positive:) The theory community is the ideal model of collaboration:) This was the best FOCS ever:)

    Bill, the truth is that things are not black or white. They are in the different shades of gray. A community where people are more free to act in different shades of gray than freely speak in those shades is not really an ideal community.

    Claire, most of those feedback were not mine. I was simply a messenger. I had to because people who did not attend the conference deserves to know how it was different/same to other conferences. The information is always in delta.

    I read a very enligheting essay today on this topic. It is linked by clicking on my name.

  2. Kamal, well said!
    BTW, I'm also using the alias anonymous... so as you said this must be truth talking.

  3. Kamal: reporting the shocking truth about a shortage of pastries during the FOCS coffee break is of course a valiant thing to do. Never mind that you might hurt the feelings of the organizers, and perhaps not reflect the true feelings of the attendees who thought it was actually just fine (as, apparently, you did yourself), it's certainly very honorable of you not to shy away from the "truth." Keep it coming! Next please do a thorough poll of what people thought of the quality of the air conditioning during the breaks. But God forbid you actually say something interesting about the contents of talks! (Which Nicole actually did, and rather well I think.)

  4. The solution is to have anonymous guest posts.

  5. people, people...

    Let us not direct our bile and petty nastiness inward but rather outward .

    A quick poll:

    What is your least favorite area of computer science and mathematics?

    The one that you find most boring and pointless, the area whose talks you always skip, the area whose job candidates you always fight.

    I nominate the theory of programming languages. Grammar on steroids, dressed up with the jargon of category thery. God help us all.

  6. To anonymous 3 (heck, I hate to respond to anonymous).

    Yes the first day of guest blogging was great.

    It would be okay to skip the more materilaistic things like local arrangements. But if talked about it, then it should be shown in the gray color not in the all white color as done.

    Regarding air-quality, since I am sensitive to indoor air-quality I did notice it was slightly bad:) May be because the space was too crowded. That was only my on feedback and not the feedback I heard from others.

    If you like to go by post to post, you could see various shades of grays all missing.

    Post 1 was a good starting post though nothing to do with this focs in particular. But nevertheless a great ice-breaker.

    Post 2 was great. That's what I was expecting for many more sessions.

    Post 3 was great too. Personally since I am working on the subject of "women in technical leadership", I appreciated this kind of debate initiated there.

    Post 4 again had nothing to do with this FOCS in particular. Nevertheless a great advice.

    Post 5,6, and 7: We should post the business meeting slides on the web. That slide deck would have provided more information that these posts. These posts were a compromise to posting the slide deck. Of course we would need Mihai picture too besides the slide deck. In fact there are some important information missing there which affect our community's finances from the long term perspective. That information was missing from the slide deck too, though cliare quickly said it verbally. I missed capturing those information though -- nevertheless those numbers affect our finances. I may write more about those some other day.

    Post 8: Nothing from the two days of talk but a materialistic post, which misrepresented the materialistic content towards all white. I expected ten posts of the post 2 type. A student among us won the best student award. That student needed more exposure than even Nancy. People wanted to know who got the best student paper and best paper award as soon as possible.

    This last post is not the short coming of Nicole. She is a dear friend of mine and a close colleague. This last post reflects on us as the theory community. Why a junior person has to write all white only? Why a statement from Bill mentioning "positive outlook" is needed. Reporting is not positive or negative, reporting is everything what's happened.

  7. You could use something like for conference coverage where only people attending the conference can contribute.

    In fact, you could even use it on each talk so that the speaker can get feedback in real-time -- thus allowing him/her to quickly correct misunderstandings, etc.

  8. So, who won the awards at FOCS?

  9. I was one of those people who found the coffee and pastries insufficient. But these were minor things and overall the organization was the best I have seen over the past few years. I think that people complaining about pastries is a great reflection on how smooth the conference was.

    Technically, I found that talks of my interests (algorithms related) were fewer than I expected. I am not complaining too hard about that as it gave me lots of time to talk to people without feeling guilty about missing talks which I had planned to attend.

    Best student paper award went to Per Austrin (KTH), "Towards sharp inapproximability for any 2-CSP"
    and best paper award went to Dan Boneh, Craig Gentry and Mike Hamburg for the paper "Identity Based Encryption Without Pairings".

  10. Kamal:

    Why be so critical? As a consequence, Nicole might decide that she never wants to blog again. I really like the idea of having an "official blogger". My idea was that reading the blog would help the people who were not attending FOCS but wished they were, and make them feel better connected. That was also the reason for filming the tutorials. The movies may be amateurish, and the blog may be imperfect, but it's a first step in the right direction (in my opinion.)

    On the day before the tutorials, had you been at the hotel at 7pm, you would have seen 4 members of the organizing team wondering helplessly about the conference proceedings' whereabouts. Had they ever been delivered to the hotel? No one knew! At 9pm, after the others had left, you would have seen me exploring the backrooms of the hotel with one of the hotel staff before finally giving up. At 10pm, I got a call from the security guard: "I found them!". In a remote basement in the hotel, there was a pallet supporting 35 boxes, each box containing 10 proceedings. A few minutes later, you would have seen three hotel staff members and myself forming a line and passing boxes from each person to their neighbor, assembly line style, so as to unload the boxes into the board room.

    Because the security guard took my problem seriously and continued searching after I had given up, and because they dealt with the problem immediately to the best of their ability, I liked the hotel very much. They were clearly not used to hosting conferences, but they went out of their way to try and make things work for us.

  11. I'm sure that Nicole's blogging was helpful to most of us who weren't able to go to FOCS, but Kamal's general point stands: collegiality is taken to an extreme in our field. Why criticize someone for being "critical" when they're just speaking their mind (not to mention the minds of several others, such as myself, who're not as brave)?

  12. Claire and others, I was not the first one to write about materialistic things. It was the reporter. If nobody said anything about the food stuff etc, I would have remained quite too. These materialistic things were not very important to me either. What was important to me is to reflect the current culture of our community. The culture was reflected when all white picture is presented whereas there was some of the gray shades too.

    Claire, I agreed with your above comment about the hotel staff, as I wrote earlier, hotel room service was good but their conference infrastructure was poor. They probably would not have been able to handle if we had parallel sessions.

    I started writing from the sixth post when the reporter started the coverage of the business meeting. The coverage started with some of the least important things, like local arrangements. But missed what Paul Beam suggested and what Sanjeev Arora suggested. These were the innovations which everybody wanted to know irrespective of whether they were at FOCS or not.

    Regarding whether Nicole would blog again or not, she knows me well. I criticize systems not people. I think Nicole did a great job within the current expectations of the community. Any other typical person, especially a junior one, would probably have done a similar job though not necessarily of this high quality.

    What short term outcome I want from my commentary here is to have even better reporting in future. Is not it positive thinking to expect some improvements? My definition of better is to bring at least half of the utility of attending a conference to non-attendees. That is report what is the same/different about the conference without any evaluation or fear of anything. A reporter is a messenger. Even in old ages, when the kings of neighboring kingdoms used to fight, even then they had rules to make sure that messengers convey the message fearlessly without any bias.

    Why our theory community system pretends to be so much egalitarian. It is actually not as the anonymous comments show. Nobody actually have the problem with my comments here. Many people though have the problem with me putting my names. They have contacted me and tried to explain to me why it is bad to put my name. Not because they are concerned about me. They contacted me because according to them, by putting my name, I disturb the current balance of the community. I understand their position. An anonymous comment can be taken lightly but a named comment is much more serious. But they do not understand my position. I actually want to disturb the balance. WHy? Because I think the current balance does not reflect the distribution of power which is enabled by this medium, which we call the Internet.

    Recently I had a chat with an industry leader of web 2.0. According to the leader, ability of anybody to have a say in anything and be able to tell the say to the world freely is going to change all our societies. I totally agree with the leader. Any community small or big will see this change.

    Look at this blog. Suppose an action X is not desirable in our community. People will write, may be in anonymous label, but they will write X is not desirable. Perhaps somebody, seeing that many agree that X is not desirable, may even write in his/her own name that X is not desirable.

    We never had this ability. A grad-student could not have said X is not desirable if it would hurt some others. But now the grad-student can. This is very big change.

    If I compare today's grad student with the ones in my time, I think todays grad students stand taller than me and my fellow grad students. All because of this revolutionery change which is changing everything faster than global warming.

  13. Kamal, I don't understand your complaints, about Nicole or the community.

    Nicole was not a reporter, she was volunteering to write some light-hearted stuff about the conference. Other than best paper awards (which should have been mentioned), what did she miss? Did you expect her to post summaries of every talk? I would have liked to hear more about funding, but Nicole has never written a grant before and it is unrealistic to expect that she would write on that. And who the hell cares about the quality of the pastries?!

    As for the community, I'm not even sure what your complaint is but I think you are upset because people are afraid to complain non-anonymously. Many people who do not yet have tenure are understandably reluctant. I have tenure, but this argument is not important enough to make personal, nor should it matter what my name is.

  14. As a consumer--not a creator--of complexity research, I will just offer my opinion that we are all living in a "Golden Age" of complexity research.

    Here "Golden Age" is to be interpreted as Dirac defined it: an age in which ordinary people can make extraordinary contributions.

    Of course, the people who are living in a Golden Age sometimes find it hard to recognize the excitement and the opportunities of the age. For example, consider the peer-reviewed articles on quantum density functional theory (DFT):


    DFT articles in Inspec:
    (("density functional" or
    "quantum density") WN KY)

    9 for 1965-1969
    33 for 1970-1974
    262 for 1975-1979
    719 for 1980-1984
    1747 for 1985-1989
    3557 for 1990-1994
    7819 for 1995-1999
    14695 for 2000-2004
    9035 for last two years


    Golly ... forty years of sustained growth compounded at (about) 20% per year.

    The point being, that humanity's ability to simulate complex systems—classical, quantum, and financial—has been increasing exponentially for two generations.

    This "Moore's Law of Simulation" shows no signs of leveling off, and IMHO this exponential growth is likely to continue for at least another 40 years ... and in so doing, thoroughly revolutionize all of science and technology.

    The reason is, in many of the most important practical cases, humanity's understanding of how our algorithms work is exceedingly sketchy. Density functional theory is a good example ... there is no really satisfactory understanding why DFT simulations work so well, or how to systematically make them better.

    There are dozens of other cases where this is true, and these cases are of immense practical and humanitarian importance.

    And because Mother Nature is a wonderfully creative program manager, the problems that arise most naturally are indeed among the most wonderful and beautiful.

    This is IMHO a huge opportunity for the computational complexity community ... that is why the present is a Golden Age.

    And of course, many of humanity's problems are so urgent, that they must be solved within the next two generations ... as Apollo system engineer Gene Kranz famously said, "Failure is not an option." So it is very good news that these Golden Age opportunities exist.

    That's my (optimistic) 2¢.

  15. To the tenured professor at comment 13.

    1. Nobody is complaining at least not at this set of comments besides that I am an optimization person and I see a lot of suboptimality.

    2. If this reporting is what you wanted to know ask Hal to put the business meeting slides online. That will save a lot of work and generally be better coverage than anybody could do about the the business meeting.

    3. Besides that there was not much reporting about this particular FOCS which can't be just copied pasted for the next FOCS. Why so? But the fact is that the current equilibrium of our community reflects that. An even happens on the last three days, and reporting conveys that this is the most perfect even possible. You may check my set of comments starting from post 6 that what was missing when the reporting jumped to materialistic things like good food. For an example, Sanjeev's website should be publicized. Sanjeev emphasized this website two or three times talks.

    4. Why your name matters? It is because that put credibility to your comment. Right now, the possibility varies from you being a spammer to really a tenured professor your comment is claiming. You could be a grad student and simply trying to augment wightage to your comment. You could even be a tenured professor outside of our community. Whoever you are it is unfortunate that you do not have your name out with your comment. The comment is not credible unless the commentator choose to invest his/her personal identity to it. The biggest reason for me to feel safe in the flight is that the pilot life is invested in my own safety. The biggest reason for me to accept a comment credible is that the commentator identity is invested in the comment.

  16. "Why your name matters? Because that put credibility to your comment"

    The same argument could be applied to blind refereeing. Isn't it unfair that the accumulated worth of one's previous achievements is completely ignored when a research contribution is being judged?

  17. You realize Kamal that a commenter can sign off as anyone ?
    --A. Turing :-)

  18. #16 you are absolutely right. Identity typically provides a lot of useful information. But sometimes using that information is either considered unfair and sometimes that information can also be abused.

    So the society either wants to restrict the access to identity information, and if that is not possible than prohibit the use of such an information.

    For an example, suppose you are a selector in a univeristy. You would get a lot of statistical information by knowing the race/skin color of a student. But using this information is unfair to students. Because eventually you are picking the individual quality, and using the group quality is unfair to a student who is exception to the group. Sometimes the society allows even encourage the use of such information if it counter acts the unfairness somewhere else. For an example somekind of affirmative action.

    Things are not that different in refereeing a paper. Refreeing a paper and accepting it is just like picking a student to be admitted in the univeristy. In refereeing a paper, double blind is also useful to manage conflict of interests. It is also useful to protect the referees from accusation of referee bias.

    But when a question of free speech comes, we want the best of both the worlds. A free speech which really allows a person to speak freely (equivalent of anonymity) and still allows the person to use invest credibility earned over the years.

    There is no short solution to enable free speech. It requires cultural changes. Which are gradual and painful. But luckily we have the internet, a cheap multiway way universally accessible fast flexible communication medium. This has speeded up everything, including the cultural changes.

  19. #17 you are right. But there could be technical solution to that problem when people need to use it. For now, there is enough trust in our community that nobody is going to sign each other's name.

    Well if somebody does then to some extent we have the protocol of USPS address change. When you fill an address change request at a post-office, the post-office sends a verification at your old address. If you did not set up the address change request you could call the post-office to ignore this request.

    Similarly if somebody signs my name, I am going to object to it. Well as you could see this is not yet a foolproof protocol because I do not get a notification that somebody is signing my name. But that's an easy thing to fix if the situation ever become so bad. USPS notifies you. Similarly blogger service can notify me if somebody uses my name.

    Meanwhile, USPS system suffers from a similar problem. If I know you are on vacation for a month, I can set up the address change request on your behalf. Since you are not around, you would not be able to object and I will start getting your mail.

  20. Is "sweet talk" really a problem, assuming it's genuine? For example, I was genuinely happy with FOCS and did not have any concerns, except for very minor ones (e.g. lunch portions being too small on one particular day). If Nicole thinks that way, as well, and writes it... where is the problem?