History of the ETAIAlthough the discussion about a European-based journal for artificial intelligence has gone on for a very long time, the debate took a new turn through an evening session at the ECAI conference in Budapest in August, 1996, and at the subsequent meeting of the ECCAI trustees (consisting of representatives of all members societies). The conclusion from those meetings was that there is an active interest in starting a new publication scheme which is based on electronic publication over the Internet, but which differs from conventional journals not only with respect to the medium of transmission (paper vs. Internet), but also with respect to how the submission and refereeing process is organized.
Based on the discussion at those meetings, Erik Sandewall undertook to write a concrete proposal for how the new publication medium could be organized. The ECCAI board discussed an interims proposal in November, 1996, and proposed that the new periodical should be called the Electronic Transactions on Artificial Intelligence. A further detailed proposal, The ECCAI system for specialized research publication, was accepted by the ECCAI board in January, 1997 as the basis for the new enterprise.
In brief, the proposal is to have two periodicals with different profiles. The Electronic Transactions for Artificial Intelligence (ETAI) will be a vehicle for publication within specialized areas, such as terminological languages, automatic deduction, induction-based learning, or reasoning about actions and change -- areas where there is a distinct community of participating researhers, and where many research papers are written for other researchers within the same niche. A review board consisting of expert researchers within each specialized area will guarantee the quality of ETAI publications.
Conversely, the Artificial Intelligence Communications (AICOM) will continue to appear with limited subscriptions from ECCAI member societies. It will publish high-quality scientific articles and other scientific material which is of interest to a broad range of researchers within artificial intelligence.
The ETAI will primarily use electronic publication. It will also introduce another innovation, namely posteriori reviewing - articles are first published, then reviewed and (if they pass) they obtain certification as a quality stamp. The first publication occurs in separate first publication archives, similar to "preprint archives" but with a guarantee of persistence. The details of the proposal are in the memo mentioned above.
A related article, A Neo-Classical Structure for Scientific Publication and Reviewing contains a more detailed explanation and rationale of the reviewing scheme which is proposed to be used for the ETAI (Electronic Transactions on Artificial Intelligence).