SIGGRAPH also has no stated page limits, and every year there’s a 2, 3 or 4 page paper. The median is 8 pages. I can only think of one 12-pager off the top of my head, and I don’t think there’s been any inflation.
However, there is a strong culture around judging papers as (contributions/length). I suspect this is a critical feature.
John Regehrsays:
I agree with all this. As an author, page limits are stupid and limiting. As a reviewer, as long as I am given license to stop reading when I get bored (as the SPLASH call makes explicit) there’s no problem with long papers.
Indeed I’ll often print a PhD thesis instead of the paper version because I appreciate the extra details and the longer format makes it easier to find the parts I’m interested in.
Marcelsays:
Daniel
Very good point. Some journals like PRL restrict also the number of figures… That’s a real pain.
Best,
Marcel
Sureshsays:
The issue with copy editing charges is an important one for the conference. Even with electronic proceedings that can be downloaded. proceedings prep cost is a large fraction of the overall registration fee.
One solution of course is to skip copy editing (or at least the professional version)
Itmansays:
I totally agree that conference paper limit of 10 pages (sometimes little bit more) is somewhat ridiculous. It is also true that short papers are often harder to read, because they DO OMIT ESSENTIAL details. Mostly, it is hard to read a paper not because it is long, but because it is hard to understand. I can read say 30-60 pages an hour, but I can hardly review a single 10 page paper in several hours. Most time is spent on understanding, not reading. On the other hand, I heard very bitter complaints of other reviewers in regard to the size of conference papers.
Yet, with larger page limits, authors have fewer incentives to write concisely. So, the size of the paper is a difficult trade-off! I would not recommend abolishing it completely, but, at least, to be more flexible about it. In particular, if proceedings are supposed to be fully online.
I’m afraid, I very much disagree. We do not need more information, but better information.
Granted, loosening the page limit will have no effect on extreme cases: bad writers continue to write bad papers regardless the page limit, and good writers, like artists, choose the right canvas size for the work of art.
But what about average writers?
For them, the page limit serves an important purpose they don’t even know about: if they care about their work, they will make an effort to cram as much information into their paper while getting rid of redundancy in passages written before. Loosing the page limit for the average writer will set them free to float into the direction of the bad writer, as they have not incentive to gradually improve their skills of filling up a limited amount of pages.
Plenty of evidence can be found that properly defined constraints (even extreme ones) increase quality. Also, limited resources are one of the driving forces of creativity (see the recent http://xkcd.com/1045 or, as Sergey Bring puts it: “scarcity bring clarity”). Unlimited resources, however, make people fat an lazy.
However, I’m OK with loosening the page limit iff other constraints are introduced: for instance, how about every researcher in the world is eligible to publish only one paper of unlimited length per time period (say, 6 or 12 months). The average researcher would then make sure that his/her one shot per year counts.
Finally, I must apologize: I didn’t have time to write a short comment, so I wrote a long one instead.
[coined by Mark Twain or Blaise Pascal, dependent on who you ask]
SIGGRAPH also has no stated page limits, and every year there’s a 2, 3 or 4 page paper. The median is 8 pages. I can only think of one 12-pager off the top of my head, and I don’t think there’s been any inflation.
However, there is a strong culture around judging papers as (contributions/length). I suspect this is a critical feature.
I agree with all this. As an author, page limits are stupid and limiting. As a reviewer, as long as I am given license to stop reading when I get bored (as the SPLASH call makes explicit) there’s no problem with long papers.
Indeed I’ll often print a PhD thesis instead of the paper version because I appreciate the extra details and the longer format makes it easier to find the parts I’m interested in.
Daniel
Very good point. Some journals like PRL restrict also the number of figures… That’s a real pain.
Best,
Marcel
The issue with copy editing charges is an important one for the conference. Even with electronic proceedings that can be downloaded. proceedings prep cost is a large fraction of the overall registration fee.
One solution of course is to skip copy editing (or at least the professional version)
I totally agree that conference paper limit of 10 pages (sometimes little bit more) is somewhat ridiculous. It is also true that short papers are often harder to read, because they DO OMIT ESSENTIAL details. Mostly, it is hard to read a paper not because it is long, but because it is hard to understand. I can read say 30-60 pages an hour, but I can hardly review a single 10 page paper in several hours. Most time is spent on understanding, not reading. On the other hand, I heard very bitter complaints of other reviewers in regard to the size of conference papers.
Yet, with larger page limits, authors have fewer incentives to write concisely. So, the size of the paper is a difficult trade-off! I would not recommend abolishing it completely, but, at least, to be more flexible about it. In particular, if proceedings are supposed to be fully online.
I’m afraid, I very much disagree. We do not need more information, but better information.
Granted, loosening the page limit will have no effect on extreme cases: bad writers continue to write bad papers regardless the page limit, and good writers, like artists, choose the right canvas size for the work of art.
But what about average writers?
For them, the page limit serves an important purpose they don’t even know about: if they care about their work, they will make an effort to cram as much information into their paper while getting rid of redundancy in passages written before. Loosing the page limit for the average writer will set them free to float into the direction of the bad writer, as they have not incentive to gradually improve their skills of filling up a limited amount of pages.
Plenty of evidence can be found that properly defined constraints (even extreme ones) increase quality. Also, limited resources are one of the driving forces of creativity (see the recent http://xkcd.com/1045 or, as Sergey Bring puts it: “scarcity bring clarity”). Unlimited resources, however, make people fat an lazy.
However, I’m OK with loosening the page limit iff other constraints are introduced: for instance, how about every researcher in the world is eligible to publish only one paper of unlimited length per time period (say, 6 or 12 months). The average researcher would then make sure that his/her one shot per year counts.
Finally, I must apologize: I didn’t have time to write a short comment, so I wrote a long one instead.
[coined by Mark Twain or Blaise Pascal, dependent on who you ask]
@Martin
I thought that the quote was from Victor Hugo? 😉