tag:blogger.com,1999:blog-3722233.post8528773158611563236..comments2024-03-28T18:17:00.135-05:00Comments on Computational Complexity: Conferences AgainLance Fortnowhttp://www.blogger.com/profile/06752030912874378610noreply@blogger.comBlogger10125tag:blogger.com,1999:blog-3722233.post-56360630035215518372011-09-21T22:45:07.501-05:002011-09-21T22:45:07.501-05:00will lead to the field finally "growing up&qu...<i>will lead to the field finally "growing up" and moving to journal-publishing system. </i><br /><br />That is just one possible solution to the conference problem and as far as I can tell there is rather low support for it. So the conversation continues until we find a solution we are comfortable with.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3722233.post-81961342312462696702011-09-21T21:17:50.191-05:002011-09-21T21:17:50.191-05:00I agree with that. However, it is not clear to me ...I agree with that. However, it is not clear to me that gradual changes like "proposals for more parallel tracks, poster sessions, rebuttals etc." or introducing new conferences, like ITCS, will lead to the field finally "growing up" and moving to journal-publishing system. It may as well lead to more discussions like "should we make 4 or 5 parallel sessions?", "should we increase the number of accepts by 5%?" etc. (see recent discussions about SODA e.g. on Mitzenmacher's blog).<br /><br />Anyway, I'm not really a member of the TCS community and I don't want to start an argument, I was just surprised (as an outsider) that the conference subject is brought up that often (in various variants, from "why conference X is in Japan rather than in USA", "why are the conferences so expensive" to "why are they so few/so many accepts" etc.), practically every time a major conference is approaching.Michal Kotowskihttps://www.blogger.com/profile/04141417129628388586noreply@blogger.comtag:blogger.com,1999:blog-3722233.post-31270008931180130362011-09-21T17:55:43.656-05:002011-09-21T17:55:43.656-05:00I'm curious what people who make these complai...<i>I'm curious what people who make these complaints that often actually do to *actively* participate in changing the system</i><br /><br />Well, the system is consensus driven, so we have to start by discussing the subject. If you search far back enough in this blog you will find people slowly coming around to the idea of change. <br /><br />Secondly, at business/steering committee meetings you will find people making proposals for more parallel tracks, poster sessions, rebuttals, etc. As you know, some of those have been adopted. <br /><br />The ITCS conference was, to a great extent, borne out of a consensus that was slowly built both in cyberspace and off-line in face to face meetings.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3722233.post-19174860568629782102011-09-21T13:06:39.673-05:002011-09-21T13:06:39.673-05:00I find the frequency with which the "conferen...I find the frequency with which the "conference-centered publishing system is bad" topic pops up annoying and a sign that something is seriously wrong with TCS publishing system. I'm curious what people who make these complaints that often actually do to *actively* participate in changing the system (it's easy to grumble about "FOCS/STOC mania" etc. and at the same go to all these conferences nevertheless, with an air of self-indulgence). I don't mean to offend the authors of this blog or anyone, I'm just curious what people are currently doing to change the trends.Michal Kotowskihttps://www.blogger.com/profile/04141417129628388586noreply@blogger.comtag:blogger.com,1999:blog-3722233.post-28752393087030374752011-09-21T12:59:35.275-05:002011-09-21T12:59:35.275-05:00So on average, for any author and any given paper,...<i>So on average, for any author and any given paper, publishing in a journal or in a conference takes the same amount of time. The difference is that for a conference publication one needs in addition to spend lots of money, and put considerable time and effort to promote the work.</i><br /><br />And in addition the reviews you get from a conference are not nearly as good and thorough as those from a journal.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3722233.post-72375358342690905802011-09-20T02:46:50.096-05:002011-09-20T02:46:50.096-05:00As long as the time from submission to review to p...<i>As long as the time from submission to review to publication remains measured in years, journals will never be able to do what conferences do now. <br /></i><br /><br />Actually, conferences do not help in fast dissemination of papers. The reason is simply that most papers in any given conference are rejected, and thus enter the re-submission phase, which takes up a lot of time.<br /> So on average, for any author and any given paper, publishing in a journal or in a conference takes the same amount of time. The difference is that for a conference publication one needs in addition to spend lots of money, and put considerable time and effort to promote the work.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3722233.post-23032704699686571592011-09-19T13:28:18.748-05:002011-09-19T13:28:18.748-05:00I like Lance's suggestion "we need to get...I like Lance's suggestion "we need to get conferences out of the ranking business in CS so conferences can instead play their role of bringing community together" very much! This alone would eliminate any need of "having a virtual conferences" and many other of "CS conference problems". <br /><br />@Anon (11:09 AM, September 19, 2011): it is big secret also for me where this delay in journals comes from. When I receive an invitation to review a paper, there is a clear deadline (2-3 months or so). If I am busy right now, I decline. If not - I do this in these 2-3 months. I know that most of my colleagues act similarly. So, the handling editor should receive all reviews after at most 5-6 moths (taking into account declinations and re-invitations). So, the decision about most journal papers should be done in 6 moths, at most! Why then it takes 2-3 years(!) until the paper is published? A big secret. On the other hand, look at say JCSS or TCS. Most of their issues are "occupied" by conference publications. Regular papers must wait until they can "land" ... So, again, the "conference problem" ... and that of double publications.Stasyshttp://www.thi.informatik.uni-frankfurt.de/~jukna/noreply@blogger.comtag:blogger.com,1999:blog-3722233.post-2135351945525116902011-09-19T11:19:59.588-05:002011-09-19T11:19:59.588-05:00Geoff- many people who get turned down from Conf A...Geoff- many people who get turned down from Conf A<br />then submit to Conf B.<br /><br />Anon- Alas. One day there will be a technology<br />so that people can ... (looking for a word)<br />POST there papers on a ... (looking for a word)<br />web-like-knowledge-cener (maybe `web' for short)<br />so that works in-development can see the light of day. But until then we must live with the tyranny of high prestige conferences.<br /><br />Other thought- There really do seem to be more<br />high quality papers (SODA turning down many good papers is NOT a rare event.) I would support<br />allowing more papers into conferences in general,which will NOT dilute quality but instead<br />let more papers that ought to be out there, get<br />out there.GASARCHhttps://www.blogger.com/profile/06134382469361359081noreply@blogger.comtag:blogger.com,1999:blog-3722233.post-85932472969772278462011-09-19T11:09:10.015-05:002011-09-19T11:09:10.015-05:00As long as the time from submission to review to p...As long as the time from submission to review to publication remains measured in years, journals will never be able to do what conferences do now. Which is too bad, because I really do think that our insistence on high-quality conference proceedings hurts us as a field by preventing in-development work from ever seeing the light of day.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3722233.post-68657936761733418652011-09-19T10:40:59.403-05:002011-09-19T10:40:59.403-05:00What happens to those SODA rejected papers that we...What happens to those SODA rejected papers that were good?Geoff Knauthhttps://www.blogger.com/profile/12025560607512616605noreply@blogger.com