tag:blogger.com,1999:blog-8890204.post4499381382670567102..comments2024-03-10T05:26:42.148-04:00Comments on My Biased Coin: Collective Coordination of Conferences?Michael Mitzenmacherhttp://www.blogger.com/profile/06738274256402616703noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-8890204.post-44070794114248171022012-01-30T04:20:43.063-05:002012-01-30T04:20:43.063-05:00Sarah, I don't understand the complaint about ...Sarah, I don't understand the complaint about the timing of Usenix Security vs Crypto.<br /><br />The two conferences are incredibly far apart in topics. The overlap (in terms of number of papers that would make sense to submit to both conferences, or that stand a decent chance of being admitted at both conferences) is in my experience awfully small -- nearly zero. Given that, I can't imagine why would expect the program chairs of Crypto and Usenix Security to coordinate their schedules.<br /><br />Crypto long ago shifted away from practical papers, towards favoring theoretical work. As a result, any paper that Usenix Security would accept, almost surely would have a very difficult time getting accepted at Crypto.<br /><br />And the same goes for the other direction as well. Any paper that Crypto would accept, almost surely would have a very difficult time getting accepted at Usenix Security. Usenix Security has for some time said that it is not interested in pure crypto papers. Look at the Usenix Security call for papers; it comes right and says it: "Note that the USENIX Security Symposium is primarily a systems security conference. Papers whose contributions are primarily new cryptographic algorithms or protocols [..] may not be appropriate for this conference."<br /><br />I understand the complaint about scheduling, but I don't think Usenix Security vs Crypto is a very good example.Puzzlednoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-34863419626468964582012-01-26T22:40:33.896-05:002012-01-26T22:40:33.896-05:00Daniel: Registration and hotel booking for confer...Daniel: Registration and hotel booking for conferences does not typically open until a couple of weeks after final versions are due, usually about 2 months in advance since conference programs are only finalized after final versions are in. The vast majority of attendees register and book their travel 3-6 weeks ahead of the conference (though SODA in Kyoto might be an exception). Paper revisions don't justify another 2 months between acceptance and final versions.Paul Beamenoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-83573325578514489082012-01-26T19:22:07.082-05:002012-01-26T19:22:07.082-05:00Why not co-locate ITCS with SODA?Why not co-locate ITCS with SODA?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-34811443720696341322012-01-26T16:36:54.437-05:002012-01-26T16:36:54.437-05:00Paul: many people attend a conference only if they...Paul: many people attend a conference only if they have a paper accepted (for various reasons). So some time is needed between acceptance and the conference to allow these people to comfortably organize the trip, find funding etc. Perhaps 4 months are too much for this, but not that much.<br /><br />Also, people need some time to prepare the final version: other deadlines, holidays, coordination between the coauthors can easily make this last a few weeks.Danielnoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-11378408226114977292012-01-26T03:03:11.083-05:002012-01-26T03:03:11.083-05:00It's very often the case that "weaker&quo...It's very often the case that "weaker" conferences accommodate deadlines of "stronger" conferences, and so you have STOC deadline, then notification, and then you have ICALP deadline, notification, and then you have ESA deadline, notification, WAOA deadline, ...etc. PC of these committees do consult about the dates.<br /><br />The fact that SoCG deadline is not well coordinated with the STOC dates is most likely because SoCG community doesn't want to have its deadline after STOC notification, which is within their right, though it's not a perfect solution.<br /><br />SPAA and PODC: now SPAA is weaker and one would expect that it's SPAA who could think about changing their dates, or to propose to collocate with PODC. But I think this would mean some loss of independence for the community and they don't want to do it.<br /><br />ITCS and SODA: the clash in the deadlines and the conference dates its caused by the decisions done by ITCS 3 years ago to run things in almost the same time as SODA; SODA has established dates for many years and there is no reason they should do anything.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-85731850665003769442012-01-25T20:40:47.801-05:002012-01-25T20:40:47.801-05:00Another example of missed coordination is STOC and...Another example of missed coordination is STOC and SoCG. It does no one any good that the deadlines clash, and while the communities are slightly different, they're not so different that some coordination can't happen.Suresh Venkatasubramanianhttps://www.blogger.com/profile/15898357513326041822noreply@blogger.comtag:blogger.com,1999:blog-8890204.post-86321613070448716142012-01-25T20:32:57.973-05:002012-01-25T20:32:57.973-05:00A non-theory example: the USENIX Security and Cryp...A non-theory example: the USENIX Security and Crypto deadlines are 16 hours apart! Normally they are a week apart and the conferences are back-to-back, so I wonder who decided to make things even more tricky this year.Sarahnoreply@blogger.comtag:blogger.com,1999:blog-8890204.post-57776190210175918602012-01-25T19:32:29.443-05:002012-01-25T19:32:29.443-05:00There is coordination that goes on in a decentrali...There is coordination that goes on in a decentralized way. We have a bunch of set patterns that get tweaked each year. People tend to know which conferences to check on, etc. Notification and submission deadlines of consecutive major conferences often are coordinated explicitly between, say FOCS and SODA. ICTS is relatively new and so it is not part of the standard pattern.<br /><br />One issue is that all our standard sequencing is becoming outdated as the time between submission of final version and conference date can be shortened due to electronic publication. The problem is that the review cycles no longer match this. There is no need for a long window between acceptance and submission of final versions. (ICALP gives 3 weeks for example).<br /><br />We have 4 month gaps from acceptance to publication: SODA is in January though the reviewing is done by mid-September. FOCS is in October but the reviewing needs to be done by June to have decisions prior to SODA submissions. STOC is in May but the decisions are made in January. Part of this is that the reviewing work for these big conferences is designed to be done at times when PC members have breaks from classes but even if that weren't the case it seems that the entire system of theory conference dates tends to be too tightly coupled for individual conferences to make major changes.Paul Beamenoreply@blogger.com