Magazine article Information Today

Machine Learning's Editorial Board Divided

Magazine article Information Today

Machine Learning's Editorial Board Divided

Article excerpt

Forty members of the Machine Learning journal's (MLJ) editorial board have resigned, citing policy differences with Kluwer Academic Publishing. (See the October 22 News$reak at http:// -3.htm.) The resigning members, who comprise two-thirds of the board, explained their reasons for leaving the 15-year-old Kluwer-- published journal in an open letter that circulated on the Internet in October. In the letter they offered their support for the Journal of Machine Learning Research (JMLR), a new, primarily online journal.

JMLR was begun in spring 2000 and is based on "a new vision of the journal publication process in which the editorial board and authors retain significant control over the journal's content and distribution." JMLR is a SPARC partner and was specifically founded to provide researchers in the machine learning field with a "high-quality, low-cost alternative" to MLJ. JMLR is peer-reviewed, with the content and format of the Web site entirely controlled by the editorial board.

While highly unusual, the actions of MLJ's board are not without precedent. In 1999, the editor and editorial board of Elsevier Science's Journal of Logic Programming (which is now known as The Journal of Logic and Algebraic Programming) resigned to join Cambridge University Press' Theory and Practice of Logic Programming. These editors also cited philosophical differences in the journal's restrictive distribution policies. The transition to the new journal had been considered successful enough for MLJ's editors to make a similar move.

Policies and Responsiveness

According to the open letter, MLJ originated when "a viable commercial publishing model was in place so that the fledgling MLJ could begin to circulate." But the Internet has changed the distribution model so that articles can easily circulate on it. The articles in JMLR are available for free, without limits and without conditions, at the journal's Web site ( JMLR is also published quarterly in hard copy by MIT Press.

The editorial board members had asked that Kluwer reduce its prices. The company agreed to lower the individual subscription fee to $120 but kept the institutional subscription fee at $1,050. The signers of the open letter contend that these subscription plans provide access to those who can afford them, but the fees "have the effect of limiting contact between the current machine learning community and the potentially much larger community of researchers worldwide whose participation in our field should be the fruit of the modem Internet." In contrast, the hard copy version of JMLR has an institutional rate of $225 for print and electronic access, $202 for electronic access only through MIT Press, and $45 for print and electronic access for individuals.

Kluwer posts articles for free on MLJ's Web site immediately upon acceptance. And a recent change in Kluwer's policies now also allows authors to post their papers on their personal Web sites both prior to and after publication. Leslie Pack Kaelbling, JMLR's founder, contends that the new Kluwer policy of permitting authors to self-- archive their refereed papers online is a reasonable one and would have made the creation of an alternative journal unnecessary. Even though the editorial board members had made it clear they wanted such a policy, it was only after the threat of resignations and the actual founding of JMLR thai the Kluwer policy finally changed.

Robert Holte, MLJ's editor in chief, said: "Until recently, the publisher was not very responsive to requests by the editorial board to change things to better serve the research community. In the meantime, journals such as JAIR (Journal of Artificiai Intelligence Research) had shown thai electronic journals could be created and run successfully entirely by the researchers themselves. The creation of JMLR along the JAIR model was a natural development. Some number of resignations were inevitable in order to give JMLR the commitment and credibility it needed to begin to establish itself. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.