Thursday, June 4, 2009

Conceptual Contributions Conference

At STOC, there was an announcement of a new conference dedicated to conceptual contributions, titled ICS (Innovations in Computer Science). Michael asks for more discussion on this conference.

Personally, I am enthusiastic about ICS. It seems like one of the best ideas in decades for improving the quality of STOC/FOCS.

9 comments:

Warren said...

Didn't you argue vigorously against this sort of thing a year ago? Have you changed your mind?

Anonymous said...

I think you missed his point...

Anonymous said...

No, he has not. Just read what he says more carefully. He sees _ICS_ as a mean to improve _STOC/FOCS_. Now, try to infer the true meaning out of it.

Anonymous said...

Whow, Snarky.

Regardless, I disagree with you. Conferences are to a large extent what you make of them. If good people come and present interesting results, then it would be interesting, if not, then not. We only live once, and the scientific process doesn't live or die on who-got-into-what-conference. Everyone were talking about Braverman's solution of the L-N conjecture, regardless of not being published yet. A similar thing holds for the new direct-sum result (also Braverman, jesus, you people). I've seen people with 2 STOc-FOCS publications get good jobs, and people with five not being able to find a good job. I empirically seems (to me) that it's more about what you do then where you published..

In other words, the quality of the conference will be a function of what papers are presented in it.

After having said that, I have to say that I agree with you about "conceptual results" and the conceptual menifesto. In STOC '08 we were sitting in talks, and every time that someone gave a sucky talk, we would say "oh, that must be a conceptual paper". It was the analogue of "he's a really nice guy",talking about someone stupid and ugly.

I think the conceptual menifesto is misguided. The refereeing process is broken, but "conceptual results" are not even the ones that suffer most. Think of two solutions, A and B, to a reasonably difficult but non-famous problem. Suppose that A is an elegant and simple solution, that can be easily described in one or two pages. And that B is a convoluted solution with lots of appendices. Which of them has a better chance of getting into STOC-FOCS? Exactly.

Anonymous said...

Think of two solutions, A and B, to a reasonably difficult but non-famous problem. Suppose that A is an elegant and simple solution, that can be easily described in one or two pages. And that B is a convoluted solution with lots of appendices. Which of them has a better chance of getting into STOC-FOCS? Exactly.

We had a recent similar experience. We introduced a new technique that allows us to substantially sharpen known results obtained using existing methods. All one has to do is re-run the old proof, together with the new insight, and voila! sharper bounds come out. We gave numerous examples where this technique works as described.

One of the comments from the referees was "technical depth is weak, none of the proofs are sophisticated" since all the proofs were known before.

xiaowei said...

let us not forget how this new ICS conference came about. it came about because theoretical computer scientists in mainland china, most notably amongst them Andy Yao, needed to find a good way to attract great potential to come to china. I believe that the underlying incentive is keep the young chinese students motivated in TCS and one way to bring life to such grand aims is something new like ICS. Being a westernized chinese, i wonder what might happen if the TCS community in China outperforms the West, will it follow the usual pattern that china has chosen ?
"thanks, guyz , now we dont need you "visitors" anymore ... "

anonimyster said...

xiaowei:

First of all, this is not formally recognized as the reason for the creation of ICS. It is, of course, but no one admits it.

Second of all, I'm guessing Chinese scientists will play well. From my experience, scientists, at least in less-political fields like TCS and math, almost always show a stronger alliance with science than with the country's habits.

Thirdly, and lastly, a China where good science flourishes will no longer be the China we know. I personally believe that in the current Chinese environment, science is not likely to succeed. Academia is highly political and controlled by the Communist Party. It is also underfunded, and run by geezers that hate new talent. It will take many years to fix this, and even that will only happen in a much free-er China than we know now. And that's before I started talking about the low level of teaching. The fact that more than a billion people learn mainly by way of rote, makes me sad every time.uis

xiaowei said...

anonimyster:

Everything else being said about the origin and cause for the conference is embellishment talk.

Well, I know for fact that Andy Yao has received plenty of funding to support his aims not only from Tsinghua but also from the Communist Party.

Being an expat, I am ashamed to say with confidence if china ever flourished in tcs that the people responsible for it [The West/U.S.] will be ignored and the current friendly climate towards the west will cool in exponential time.

China sad as it sounds buys what it needs at good inflated rates.

The political system is deep ingrained in mentality of the people. it will take generations for china to turn into something better. all else is show and all is merely shines and is not gold.

Anonymous said...

Being an expat, I am ashamed to say with confidence if china ever flourished in tcs that the people responsible for it [The West/U.S.] will be ignored and the current friendly climate towards the west will cool in exponential time.

Even if that is indeed the case, I do not see any reason for not accepting their generosity right now; particularly, given the current finances of theory, we should take all the help we can get. Later, if the source of funding in China dries up, that's fine too -- they are indeed being very generous right now, and it is not like they owe us anything in the future.