Résumé: This editorial analyzes from a variety of perspectives the controversial issue of single-blind versus double-blind reviewing. In single-blind reviewing, the reviewer is unknown to the author, but the identity of the author is known to the reviewer. Double-blind reviewing is more symmetric: The identity of the author and the reviewer are not revealed to each other. We first examine the significant scholarly literature regarding blind reviewing. We then list six benefits claimed for double-blind reviewing and 21 possible costs. To compare these benefits and costs, we propose a double-blind policy for TODS that attempts to minimize the costs while retaining the core benefit of fairness that double-blind reviewing provides, and evaluate that policy against each of the listed benefits and costs. Following that is a general discussion considering several questions: What does this have to do with TODS, does bias exist in computer science, and what is the appropriate decision procedure? We explore the “knobs” a policy design can manipulate to fine-tune a double-blind review policy. This editorial ends with a specific decision.

Langue: Anglais