A cancelbot is an automated or semi-automated process for sending out third-party cancel messages over Usenet, commonly as a stopgap measure to combat spam.
Contents
History
One of the earliest uses of a cancelbot was by microbiology professor Richard DePew, to remove anonymous postings in science newsgroups. Perhaps the most well known early cancelbot was used in June 1994 by Arnt Gulbrandsen within minutes of the first post of Canter & Siegel's second spam wave, as it was created in response to their "Green Card spam" in April 1994. Usenet spammers have alleged that cancelbots are a tool of the mythical Usenet cabal.
Rationale
Cancelbots must follow community consensus to be able to serve a useful purpose, and historically, technical criteria have been the only acceptable criteria for determining if messages are cancelable, and only a few active cancellers ever obtain the broad community support needed to be effective.
Pseudosites are referenced in cancel headers by legitimate cancelbots to identify the criteria on which a message is being canceled, allowing administrators of Usenet sites to determine via standard "aliasing" mechanisms which criteria that they will accept third-party cancels for.
Currently, the generally accepted criteria (and associated pseudosites) are:
By general convention, special values are given in X-Canceled-By, Message-Id and Path headers when performing third-party cancels. This allows administrators to decide which reasons for third-party cancellation are acceptable for their site: