I’ve just finished up a conference submission, which means I’ve spent quite a bit of time forcing my paper into compliance with the publisher’s formatting requirements:

Papers must adhere to the standard ACM conference format: two columns, nine-point font on a ten-point baseline, with columns 20pc (3.33in) wide and 54pc (9in) tall, with a column gutter of 2pc (0.33in).

Furthermore, papers must be

at most 12 pages, including bibliography and figures.

Is there any doubt that these requirements actively harm scholarship? Anyone who has tried to comply with them understands that:

• They force authors to spend a great deal of time eliminating widows and orphans, and otherwise trimming text according to the geometry of the page, regardless of whether that text could better explain the research.

• Including the bibliography in the page limit makes it impossible to properly cite the literature. In practice, this requirement means that papers written by program committee members will be cited (otherwise, the paper stands no chance), but many relevant papers must be omitted.

• A nine-point font is almost unreadable on paper, the narrow margins leave no room for annotations when reading, and a two-column format is difficult to read on an electronic device.

A fair reviewing process requires some limit on the length of submissions, but basing it on a number of US letter-sized pages is archaic. On top of that, all of these requirements are driven by the desire to keep the cost of printing low, at a time when I, for one, don’t even want printed documents.

I once sat on a plane on the way to a conference next to a colleague who wrote up a submission for a different conference in pencil, during the plane ride. The submission was accepted. I’m sure there is some sensible way to fairly enforce a limit on the length of submissions that does not require them to be printed out in a nine-point font.

Some people argue that the best (only?) way to read a paper is to print it out. But of course a paper that has been formatted for print is best read in print! That has no bearing on how we format future papers. Going forward, academia should abandon paper.