Guidelines for authors regarding double-blind reviewing

A reviewer's attitude toward a submission may be affected, even unconsciously, by the identity of the author. We want reviewers to be able to approach each submission without such involuntary reactions as ``Peyton Jones; he writes a good paper'' or ``Who are these people? I have never heard of them.'' For this reason, we ask that you omit your names from your submission, and that you avoid revealing your identity through citation.

Different conferences handle double-blind review differently. For ICFP, most of the burden is on the reviewers; the burden placed on you, the author, is light:

• Your job is not to make your identity undiscoverable but simply to make it possible for our reviewers to evaluate your submission without having to know who you are.

• Our job is to evaluate your submission and write reviews without trying to discover who you are. For example, Google may know who has written what, but we will resist the temptation to ask!

Your part should be simple: omit authors' names from your title page (or list them as ``omitted for submission''), and when you cite your own work, please make it difficult to figure out exactly who you are, even if a reviewer might be able to make a good guess. For example, if your name is Smith and you have worked on amphibious type systems, instead of saying ``We extend our earlier work on statically typed toads (Smith 2004),'' you might say ``We extend Smith's (2004) earlier work on statically typed toads.'' If anonymizing your submission isn't easy, please ask the program chair for help.

Our part is to arrange that reviews of your submission are written by reviewers who do not know your identity. Once all reviews are written, your identity will be revealed to the program committee. This procedure makes it possible for the committee to understand the entire context of your work as well as to perform such necessary tasks as checking for dual submissions.

If you are curious about details, here are the organizers' thoughts:

• As an author, you are always obligated to discuss all relevant related work, including and especially your own. The desire for anonymity does not remove or reduce this obligation. Just discuss your work as if it's important work that someone else has done.

• While your submission should not be obviously written by you, there is no need to worry about exposure outside the submission itself. Feel free to put your submission on your web page, give talks about the work, and whatever else you normally do to disseminate it.

• It is quite common for a certain kind of submission to be supported by a voluminous technical report, which may contain many pages of rules and proofs. Because even the location of such a technical report may provide a big clue as to your identity, we are instead providing a way for you to include the technical report with your submission itself, by placing both your submission and the accompanying technical report in a zip file. This way, a reviewer with an appetite for rules and proofs can see that your submission is backed by substantial nutritional value—all without having to know your identity.

  The bundling mechanism is not limited to technical reports. If you find it appropriate, you may bundle your paper with source code or even web pages—just please remove evidence of your identity. (find and sed are your friends.)

• We recognize that work focused on particular languages and systems is less likely to be anonymous. If your submission is about the Xyzzy language, and if it is well known that only a few people can speak Xyzzy, then some reviewers may guess your identity. That's OK—in many cases there will still be an element of doubt. There is no need to resort to extreme measures such as renaming your language or system.

  For example, suppose that you have previously worked on the Fungol system, and now you have developed a static type-checking algorithm for it. It is perfectably acceptable for you to say ``Building on our experience implementing the Fungol system, we realized that static type checking would be easy.'' We want you to be able to demonstrate that you are an expert in the Fungol system. We recognize that if Fungol is well known and there are not many Fungol developers, your identity may well be revealed to some reviewers. When you cite those earlier Fungol papers, however, please leave your names out of the citation. That way you preserve the element of doubt, at least for those reviewers who aren't certain who's who in Fungol.

• You may feel that in order to place your result in the context of a larger project, you must not only cite many of your own papers, but also make it clear whose they are. If you must, go ahead; it is better for you reveal your identity than to weaken your submission. But if you can, please try to preserve some element of doubt.

Our guiding principle is this: while anonymity helps make us more confident that the reviewing process is objective and fair, anonymity must not force an author to weaken a submission. So please try to hide your identity, but not to the extent that it warps your submission—if you try to hide your identity and fail, there is no penalty.

If you are not sure how to apply these guidelines, please make contact with the program chair. (And if you do, please do not mention the title of your submission—if the chair needs to know it, he will ask for it.)