Posted by: jonkatz | September 2, 2009

TCC 2010

As many of you are probably aware, today is the submission deadline for TCC 2010. (On a side note: whose idea was it to make this the week NSF proposals were due, Asiacrypt final versions are due, and the first week of classes?!)

The deadline is 5PM EST. I just submitted a paper (it’s shortly before 3), and it got assigned a submission-ID of 174. Since submission-IDs start at 100 (or 101, I forget), this means there have so far been only 74 papers submitted. This surprised me a bit at first, but is about in line with the previous two years (assuming, as usual, that there will be 20-30 papers submitted at the last minute*). So here’s the controversial question of the day: is TCC more competitive than Asiacrypt (which gets roughly 3 times the number of submissions)? If so, why? If not, then how to explain the perception among many that it is?

*In case you’re wondering, I submitted this paper early since I need to teach at 3:30…

Advertisements

Responses

  1. “So here’s the controversial question of the day: is TCC more competitive than Asiacrypt”

    At least TCC seems less random, the comments I got so far from TCC were always to the point and the accept/reject decision was reproducible. On the other hand, almost all reviews I got so far from AC were… how do I say that diplomatically… not very relevant. I really wonder why, at least this year the AC committee looked quite strong?!? the fact that AC has a much higher

    #committee members/#submission

    might be one reason.
    krzysztof

  2. Krzysztof, do you mean that you get better comments from TCC because *TCC* has more committee members per submission?

  3. AC has a quite strong PC (how do you compare them anyways?), though it’s bigger and thus not everybody is a star. The comments I got from AC were also not so random – some were very good, some were weaker. I’ve got much worse comments from Crypto.

  4. @john: yes, of course I meant the load per committee member is higher at AC. but as Helger points out the AC committee is much bigger, so that difference is actually much smaller than I thought.

    I’ve heard of quite a few papers now which got rejected from AC and then got into Eurocrypt/Crypto (my only submission to AC before this year is such a case), but of no case where this happened to a TCC paper (though there probably are some). maybe this impression is just wrong, another explanation is that AC is actually better than TCC, and in fact competitive with EC/Crypto, a third is that the AC review process is simply not very good, and of course there might be other reasons I’m just missing.
    krzysztof

  5. Forgetting about the status of AC for the moment, I think TCC is “overrated” in some sense — though it really depends on what you mean when you rate a conference.

    It’s fair to say that TCC papers are narrow: for the most part, a TCC paper is of interest to the TCC community (rather than of interest to the “general crypto community”), and at worst TCC papers are of interest to a subset of the TCC community that might be as small as 10 people. This is an observation, not a complaint — I think TCC was created with exactly this vision in mind.

    As a consequence, TCC attracts a dedicated, interested audience. For the TCC community, it has become perhaps the conference to attend. This also means that if your paper appears there, people will notice it; in that sense, it is a conference where you really want your papers to appear.

    On the other hand, it is not that hard to get a paper into TCC. This is true both in terms of sheer numbers (since the acceptance rate is high) as well as in terms of the bar for acceptance. Certainly it’s easier to get a paper into TCC than Crypto, and it seems now this holds true for AC as well.

  6. While I have never had a paper at TCC or AC, I have felt pretty much the way Jon describes about TCC (I really have no idea about AC though).

    I think there are two sets of people in the “TCC vs Crypto/EC” debate that I have noticed over the last few years… and this question I hear invariably at almost every conference I attend: 1) people who care about how a paper at one of these conferences looks at your CV, and 2) people who discuss where do you see papers more “relevant” to your research interests.

    Type-1 is basically more about the “competition aspect” of research and it seems to be higher at Crypto/EC than TCC. It also seems true though that there is less randomization in the selection procedure at TCC than at Crypto/EC.

    Anyway coming back to the discussion, given a person I can typically tell his/her reaction to “Crypto/EC vs TCC”. The TCC-community tends to weigh TCC higher than Crypto/EC and their standard argument is that TCC-PC does a really good job at selecting papers (and so there’s less randomization) and they typically see more of what they’d like to see at TCC than Crypto/EC. So they’re thinking in terms of what conference they’d like to attend if they could attend only one.

    Outside this community, Crypto/EC seems to carry more weight. Or more accurately, people who are equally involved with stuff that is not completely hard-core theoretical crypto (like practical IBE, efficient CCA2 secure encryption, etc), tend to weigh Crypto/EC higher than TCC.

    Omkant


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: