Formatting support for ACM conferences

When submitting a paper to an ACM conference you must format your paper in accordance with the ACM specification for conference proceedings. You can find this at (Section 1.1 contains the specification).

ACM has produced a couple of LaTeX, Word Perfect, and Microsoft Word style files that meet this specification. You can find them at Unfortunately, there are a number of things wrong with these templates; their multiple problems have been described by Norman Ramsey in a SIGPLAN Notices article ("LaTeX support for proceedings", Norman Ramsey, SIGPLAN Notices 37(4) (Apr 2002), pp1-3). Please do not use these style files if you can at all avoid it; they look pretty bad. (In particular, the ALL-UPPER-CASE SECTION TITLES are pretty garish.)

Happily, Mike Sperber has produced a LaTeX class file that does not have these deficiencies, and meets the ACM spec. You can find it here:

Word Perfect and Microsoft Word users can use the style files found at the ACM link above -- they are not as attractive or as well engineered as Sperber's design, but will meet the spec. If you are preparing your paper with some system other than LaTeX, then you are on your own with respect to the ACM formatting specification.

Using the new style

The new class file defines all the same LaTeX macros as the ACM template, so please read the ACM guide at first.

Mike's class file has the following differences from the ACM one

Here are few other points about the style.

Font magic and paper size

Like the ACM template, this class file uses Times Roman type-1 fonts as required by ACM. However, this means that typewriter font, often used for typesetting program code, comes out rather faint and spidery. I much prefer the (LaTeX default) Computer Modern Typewriter font, both on screen and on paper.

If you wish to use the CM font, it is easily arranged -- just add the command

However, CM is typically defined as a native TeX/Metafont font, not a PostScript type-1 font, and ACM requires type-1 PostScript fonts, so you (may) also have to change your dvips command:
  dvips -t letterSize -P cmz  <...whatever you have normally...>
The -t letterSize tells dvips to format for US letter size, which is what ACM wants, and the -P cmz somehow solves the type-1-font problem. (This is copied from the ACM FAQ on this topic; see the whole FAQ topic for more information. See also the dvips documentation. Also note that, on a standard Linux system, the font configuration file referenced by -P cmz lives in /usr/share/texmf/dvips/config/config.cmz and /usr/share/texmf/dvips/bluesky/config.cmz.) Incidentally, doing this also means that when you convert to PDF you get decent looking output.

Don't forget the -t letterSize argument, especially if you are in a country that uses A4 paper by default!

A plea for proper style

The single most common, blatant error in academic writing is the use of citations as "nouns." Citations are, in fact, annotations , that is, parenthetical things tacked onto the sentence structure. Think of them as being like footnotes. If you delete all the footnotes and citations from your paper, you should still have a coherent, grammatically correct paper. Do not write
The result of [Mouse94] is fundamentally flawed -- indeed, willfully wrong-headed -- as we had previously shown in [PrincipiaInformatica92].
Instead, write
Mouse' result [Mouse94] is fundamentally flawed -- indeed, willfully wrong-headed -- as we had previously shown as early as 1992 [PrincipiaInformatica92].
This sometimes requires you to change your sentence structure around. But it is never correct to use a citation, such as [Harper92] or [Peyton-Jones85], as a part of a sentence. Authors commonly do this. It is not correct. People who know better grimace when they stumble over this kind of error. Don't be one of the unfortunate souls whose published oeuvre is blighted in this unseemly manner.


I hope that this page takes care of most of the issues involved in typesetting your paper. If you have any questions about the new style, ask me in the first instance (kfisher at Mike Sperber has done enough! (I should also thank Simon Peyton-Jones and Olin Shivers, who wrote the first and second versions of this page for ICFP 2002 and 2003.)
Kathleen Fisher