FOS Newsletter, 10/12/01
Oct 12, 2001 10:14 PDT
Welcome to the Free Online Scholarship (FOS) Newsletter
October 12, 2001
Journal editors resign to protest publisher's policies
Forty editors of the _Machine Learning Journal_ (MLJ) have resigned from
the editorial board and published their reasons in a public letter dated
October 8. The MLJ editors were frustrated by the reluctance of Kluwer,
their publisher, to adapt the journal to the digital age. They asked
Kluwer to lower the subscription price and provide free online access to
the articles. Without these changes, the subscription price limited access
to the very researchers whom the journal ought to serve.
Quoting the public letter: "While these [subscription] fees provide access
for institutions and individuals who can afford them, we feel that they
also 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
Kluwer agreed to lower the individual subscription price (to $120) but
would not lower the institutional price (at $1,050) or provide free online
access to the articles.
Leslie Pack Kaelbling resigned as one of MLJ's action editors and began
looking for a publisher willing to host a journal on machine learning more
in keeping with her vision of wide and free online access. She struck a
remarkable deal with MIT Press. She would launch a new journal, the
_Journal of Machine Language Research_ (JMLR) which would provide free
online access to all its articles and publish them online as soon as they
were accepted. Quarterly, MIT would publish a print edition at a
reasonable subscription price. MIT brought in the Scholarly Publishing and
Academic Resources Coalition (SPARC) to use its international network of
member libraries to guarantee an adequate subscription base for the new
journal. Finally, JMLR would leave copyrights in the hands of
authors. MIT would only have the right of first print publication and the
right of first refusal on anthologies of JMLR articles. MIT agreed, in
effect, not to own the journal or its contents, but only to publish the
print edition. No money changes hands between JMLR and MIT.
MIT can agree to these terms in part because JMLR editors keep costs down
by providing online- and print-ready copy in PDF format. MIT is also
willing to experiment with new ways of doing business in the digital age.
Once JMLR was in the cards, Leslie invited all the MLJ editors to join her
at the new journal, without necessarily resigning from MLJ. All but a
handful chose to resign and join her. Some are editors at both
journals. While the 40 resignations have taken place over the past nine
months, the 40 agreed only recently to publish a joint, signed open
letter. Their purpose was to describe their grievance with Kluwer and to
explain to the world that JMLR is not the raw newcomer that it might
otherwise appear to be. Hiring and tenure committees should understand
that JMLR is the leading journal in the field of machine learning, even if
its citation history and impact factor have not had time to reflect the
eminence and experience of its editorial board.
Thanks to Leslie Pack Kaelbling for sharing these details with me in an
interview on October 9.
Public letter of resignation (October 8, 2001)
[old journal] Machine Learning (aka Machine Learning Journal)
[new journal] Journal of Machine Learning Research
SPARC home page
* Postscript. This should remind you of November 1999 when the entire
editorial board of the _Journal of Logic Programming_ (published by
Elsevier) resigned and created the _Theory and Practice of Logic
Programming_ (published by Cambridge). See FOSN for 5/11/01.
It should also remind you of the 1998 decision by Michael Rosenzweig and
the rest of his editorial board to resign from _Evolutionary Ecology_,
which Rosenzweig had launched in 1986, in order to create _Evolutionary
Ecology Research_. Are there are other, similar stories that belong on
this short list?
Will FOS do harm? More harm than good?
In the October 12 _Chronicle of Higher Education_, John Ewing argues
against a thoughtless rush into FOS. His most specific reason for caution
is that small independent publishers have the thinnest profit margins and
will be the first to fail in competition with FOS. If they fail,
publishing will be dominated even more than now by a narrow band of
profitable giants charging high prices. Richard Kaser made a similar
argument in his September 18 contribution to the _Nature_ debate on FOS
(see FOSN for 9/21/01). Arthur Smith points out in our discussion forum
that he made a similar argument in 1998. Here are some thoughts on Ewing's
version of the argument.
Priced journals cannot compete with free journals, when the two sets are
roughly equal in significance and quality. If FOS journals gain the
readership and recognition to wipe out priced journals from small,
independent publishers, then they will a foothold to threaten the journals
from the profitable giant publishers as well, even if the giants have a
thicker armor of savings to postpone the inevitable. If the big publishers
eventually fail or retreat from the journal market, then it's simply not
true that FOS will cause big publishers to dominate the journal
market. The worst-case scenario is not the dominance of giant publishers,
but a temporary period in which the giants coexist with free journals.
But in fact, there is no evidence that FOS would hurt small, independent
publishers before large ones. In a market of expensive, inexpensive, and
free journals, again assuming comparable significance and quality, there is
good reason to believe that libraries will drop the expensive journals
first and retain the affordable and free ones as long as possible. If so,
then big publishers will suffer first, not last.
Is it fair to assume that free and priced journals can be equals in
significance and quality? Ewing doesn't argue to the contrary but others
have. The short reply to this objection is that significance and quality
depend on the editors and authors, not on the marketing, medium, or
imprint. Not only can free journals have editors and authors comparable to
those at the best print journals, they can have the very same editors and
authors. It's true that it takes time for prestige and reputation to catch
up with quality, but the lag time is getting shorter as librarians work
together to boost inexpensive new journals and as a new generation of
academics understands that the medium is not the message.
Ewing doesn't pretend to know the future and I don't either. If his
predictions and mine are both taken with a grain of salt, then it remains
the case that his scenario is a risk that we can avoid only with
caution. I accept this. The only problem lies in his implication that
some FOS advocates, or journals, or publishers, or scholars are
thoughtlessly rushing fundamental change. Some calls for FOS may be less
cognizant of the obstacles than others, and some FOS projects may
fail. But let's be clear: no one is rushing the change of the conditions
of journal competition. Rushing deep changes of this kind is
impossible. Richard Kaser, in his version of the argument, worries that we
will "trade in" a journal system that works acceptably for one that may
not. Both arguments assume that FOS could replace the current journal
system suddenly, or before we adequately understand what is happening. But
this is far-fetched. FOS is emerging gradually, one journal or archive at
a time. The slow pace of change provides all the time for measurement and
reflection that caution requires. We certainly have time to monitor the
effects of FOS as it grows. Those who worry about harmful consequences can
help the cause of scholarly communication by looking to see whether feared
forms of harm are materializing. By all means warn us of risks that others
didn't see or heed. But don't pretend that we can't monitor our own
experiment and make mid-course corrections.
Finally, Ewing argues that free online journals lack a good business
model. Volunteer labor and government support may both
disappear. True. But there are many other business models than these to
examine, including author fees, university funds, print sales, and
endowments. It's true that none of these has yet proved itself for free
online journals over the long term. But Ewing's argument assumes that the
currently prevailing business plan works. It doesn't. Currently, authors
of journal articles donate their professional labor and intellectual
property to the dissemination system. In this system, publishers stand
between authors and readers, charge for access, and keep the
money. Subscription fees limit readers' access to literature and limit
authors' access to readers. This obstructs both research and
education. It might be acceptable if publishers had to stand between
authors and readers in order to disseminate the literature and if they
charged reasonable subscription prices. But neither is true. The internet
makes most functions of traditional journal publishers unnecessary. To
continue to pay for these functions at the expense of reader access and
author impact is perverse. Moreover, in the last decade journal
subscription prices have risen faster than inflation, faster even than
health care prices. This causes libraries to cancel important journals
every year, which only aggravates the harm to research and education. It
is precisely the failure of the present business model that has stimulated
the current, healthy experimentation with other business models.
John H. Ewing, No Free Lunches: We Should Resist the Push to Rush Research
Richard Kaser, When allegory replaces rational thought, science had better
Arthur Smith, August 28, 1998 posting to the _AmSci_ forum
Discussion thread on Ewing's article in the September98Forum
(Particularly good on Ewing's argument that FOS journals lack "frills".)
What do you think? Post your thoughts to the FOS discussion forum.
(Anyone may read; only subscribers may post; subscription is free.)
* On October 10, the International eBook Award Foundation announced the
second annual Frankfurt eBook Awards. The top two prizes in non-fiction
went to Steven Levy's _Crypto_ and Eric Nisenson's biography of Miles
Davis. There is no special category for academic or scholarly
non-fiction. It appears that all the winning ebooks were simultaneously
published in print.
* The California Digital Library (CDL) and Berkeley Electronic Press
(bepress) have become partners. Bepress has developed software tools to
facilitate the creation, editing, and management of free online
journals. Under the new agreement, CDL will make these tools available to
researchers at the University of California. Because the bepress tools
work with eprints.org software, we should soon see some new free online
journals and OAI-compliant archives from California departments and
* Adobe Systems has announced that it will use Info2clear technology to
improve the security of PDF ebooks. The press release doesn't say so, but
this is presumably a response to Dmitri Sklyarov's proof that existing PDF
ebook security can be broken.
* Adobe has also announced an international version of its ebook
reader. It is supposed to be the only ebook reader for books in German,
French, or Spanish that runs on both the Mac and Windows.
* Collection EgoDocuments of Montpellier has announced what it calls the
world's first free, continuously edited, web-passive ebook, an online
edition of _Le Journal du chevalier Marie Daniel Bourrée de Corberon_,
originally published in Paris 1776-1781. "Web-passive" here simply means
non-interactive. The editors plan to make a future edition the book
interactive, drawing on a database to answer user questions about Bourrée
de Corberon's life and travels. The present edition has a very useful set
of links from the text to explanatory notes and images of people or places
mentioned in the text. The boast that the book is "continuously edited"
apparently means that new links of this kind are continuously being added.
(Thanks to Ellen Fernandez-Sacco for bringing this to my attention.)
* OCLC has announced that WorldCat now has more than 500,000 records for
digital resources --out of 48 million records for print and digital
* The American National Standards Institute (ANSI) has approved the Dublin
Core Metadata Element Set.
(The announcement has not yet appeared on the ANSI, NISO, or DMCI sites.)
The approved metadata element set is online here.
* Can free online access be *too* accessible? Federal law requires that
criminal records be publicly available. Before the digital age this meant
that they were available to anyone willing to visit the courthouse and
riffle through paper files. In the digital age it meant free online
access. But recently the federal government decided to remove criminal
records from the internet in the name of privacy. The question whether
free online access is excessive even for public documents has also come up
with voter registration records, bankruptcy records, and the financial
disclosures of public officials. (PS: These are nice examples of
quantitative changes that become qualitative changes. For scholarship, we
welcome the qualitative changes that come with quantitative increases in
speed and ease of access. But for criminal and voting records, is the
qualitative change undesirable or just unsettling? Must we have two
categories from now on, "publicly available with ease" and "publicly
available with difficulty"?)
New on the net
* The National Security Archive (NSA) has launched a free online series of
September 11th Source Books. The NSA is a non-governmental non-profit
organization with a two million page archive of other security documents
dating back to 1985. Its new September 11th Source Books consist of
primary source documents which it has obtained from the government through
the Freedom of Information Act.
* The Open Language Archives Community (OLAC) is a newly created
international organization of institutions and individuals assembling a
free online OAI-compliant archive of resources on language and
linguistics. The archive will officially launch in January 2002.
* The U.S. Department of Energy has launched the Energy Citations Database,
a free online archive of bibliographic citations to federally sponsored
scientific research on energy. In a small but growing number of cases, the
citations include links to free online full-text.
* The government of Australia has launched a free online archive of
Australian science and industry information. This supplements its earlier
archives on Australian agriculture, business, and culture.
* The American Association for the Advancement of Science and Science
magazine have launched a free online archive on the science of aging, the
Science of Aging Knowledge Environment (SAGE KE).
* The Johns Hopkins medical school with support from the Sloan and Wood
Foundations has launched a Center for Civilian Biodefense Studies. In
addition to providing news and conferences, it hosts a small but presumably
growing online archive of published articles on relevant scientific topics.
* Sabawoon Online has launched Afghanpedia, a comprehensive free online
guide to the history and geography of Afghanistan.
* The proceedings of the June conference in Amsterdam, Change and
Continuity in Scientific Communication, are now online.
* In August a group of individuals and organizations responding to a public
invitation launched the BioMed Archives Consortium in a meeting at MIT
Press. The plan is to provide online access to biomedical research. The
archive plans to support itself by charging institutional subscriptions.
Share your thoughts
* The U.S. government has decided that the internet isn't secure enough for
its critical and classified communications. Richard Clarke, the new
presidential advisor for Cyberspace Security, has proposed the creation of
an alternative network called GovNet. The General Services Administration
is asking the private sector to make suggestions on how to implement GovNet
so that it meets its specs, and even to bid on its construction and
maintenance. Public comments and submissions will be welcome until
Postscript. Here's hoping that this has no FOS implications. Right now
the federal government is one of the most generous providers of free online
scientific and cultural content. As proposed, GovNet will handle only
other sorts of data. But we know that the military experiences
"classification creep" and that some government agencies have already
deleted scientific information from their web sites, thinking it to be
helpful to terrorists (see FOSN for 10/5/01). Any content that migrates
from the public internet to the new GovNet will cease to be accessible to
researchers, teachers, and students.
* If you think you've missed some Requests for Comments (RFC's) for
internet standards and protocols, check out the free quarterly, _RFC
Sourcebook_. The current issue covers July-September, 2001.
In other publications
* In the October 10 _USA Today_, Edward Baig criticizes digital
encyclopedias that force students seeking information to wade through
advertisements or even sign up for commercial services in order to get it.
* The October 9 _Library Journal_ has a brief, two-paragraph report on the
Forum on Publishing Alternatives in Science held at the Johns Hopkins
medical school on October 1. One statistic revealed at the forum: between
1986 and 1999, the consumer price index rose 49%, the cost of health care
rose 111%, and the cost of scholarly journals rose 175%. During the same
period the number of scholarly journals increased only 55%. (PS: Did any
FOSN readers attend this forum? I'd like to see a more detailed report.)
(Free registration required.)
* In the October _EContent_, Sylvia Lacock Marino describes Discussion
Miner, new software to read online discussion groups, summarize them, and
package the summaries to sell to advertisers who want the latest dope on
the zeitgeist. I once read about a financial version that crunched through
stock trading discussion groups in order to summarize investor beliefs in
real time. This was based on the plausible theory that the stock market
goes up and down according to people's beliefs about whether it will go up
and down. Would an academic version be useful only to social scientists
doing field work on academic communities? Or could you get useful
meta-analysis from the distillation of academic chat?
* In an October posting to _GigaLaw_, Bob Pimm reviews current U.S. law to
summarize what rights are held, and not held, by authors of ebooks.
* In another October posting to _GigaLaw, Doug Isenberg analyzes the DMCA's
anti-circumvention clause and some of its lesser known provisions.
* In the September issue of _Ariadne_, Susi Woodhouse describes the UK's
People's Network and the New Opportunities Fund's program to digitize
* Also in the September _Ariadne_, John MacColl summarizes some of the
presentations at a June ACM/IEEE meeting in Roanake on digital libraries.
* Also in the September _Ariadne_, John MacColl, Marieke Napier, and Philip
Hunter summarize each presentation at a July meeting in London to find ways
for JISC to help advance the cause of OAI.
* In September, the JISC/DNER E-Book Working Group posted its paper on
strategy and issues, written by Hazel Woodward and Louise Edwards, to the
* In the August issue of the _Journal of Digital Information_, David Miall
and Teresa Dobson summarize research suggesting that readers are less able
to concentrate and reflect when reading hypertext than when reading text
* In April 2000, Ray Siemens and colleagues undertook a study of the use,
perception, and credibility of electronic journals in Canada. They based
their analysis on published literature in North America and Europe and on
questionnaires they sent to scholars, publishers, and university
administrators. Their reports are now online.
Ray Siemens, Introduction and overview
Jean-Claude Guédon, Peer review and imprint
Michael Best and Elizabeth Grove-White, Copyright issues
Alan Burk, James Kerr, and Andy Pope, Archiving and Text Fluidity / Version
* Jean-Claude Guédon's historical perspective on the digital revolution in
scholarly publishing, and the commercial counter-revolution, is now
online. He concludes with five reasons to support the Open Archives
Initiative and a call to librarians to play a central role in the
continuing revolution. The paper is a revised and enlarged version of a
talk originally given at ARL's Creating the Digital Future Conference in
Toronto in May.
* The URL I published last week for Audiobooksforfree.com worked when I
wrote my article on it but didn't work the day I mailed the issue. The
site is getting more than two million hits per month and its database
engine cannot handle the traffic. The URL is not dead, just periodically
dormant. The company promises better service in three or four weeks.
* In the last issue I described the self-censorship practiced by several
government agencies and private organizations. In the October 9 _Search
Day_, Chris Sherman reminds these agencies that much of scientific
information they deleted because it might be useful to terrorists is cached
by Google for all to retrieve and use.
Topica.com, the email host for this newsletter, will be down for
maintenance for 24-26 hours starting tomorrow, October 13, at 8:00 am PDT
(4:00 pm GMT). If you visit the site e.g. to search or read back issues,
don't be deterred by the "Under Maintenance" message. Just come back
later. Sorry for the inconvenience.
If you plan to attend one of the following conferences, please share your
observations with us through our discussion forum.
* IT in the Transformation of the Library
Milwaukee, October 11-14
* Collections & Access for the 21st Century Scholar: A Forum to Explore
the Roles of the Research Library
Washington, D.C., October 19-20
* Intellectual Property Rights in the Knowledge-Based Economy
Washington, D.C., October 22
* International Conference on Dublin Core and Metadata Applications 2001
Tokyo, October 22-26
* e-Book Lessons: From Life-Cycle to User Experiences
Waltham, Massachusetts, October 23
* Fourth Meeting of the [NAS] Committee on Intellectual Property Rights
(only parts are open to the public)
Washington, D.C., October 23-24
* Copyright Issues in the Electronic Age
Waltham, Massachusetts, October 29
* Paperless Publishing: Peer Review, Production, and Publication
Washington, D.C., October 30
* The XML Revolution: What Scholarly Publishers Need to know
Waltham, Massachusetts, November 1
* Information in a Networked World: Harnessing the Flow
Washington D.C., November 2-8
* Electronic Book 2001: Authors, Applications, and Accessibility
Washington D.C., November 5-7
* Content Summit 01: Funding opportunities for European digital content on
Zurich, November 7-9
* Internet Librarian 2001
Pasadena, November 6-8
* Setting Standards and Making it Real (on Digital Reference Services)
Orlando, November 12-13
* First Annual Meeting of the Text Encoding Initiative Consortium
Pisa, November 16-17
* ARL Workshop for Publishers: Licensing Electronic Resources to
Libraries: Understanding Your Market
Philadelphia, November 19
* European Forum on Harmful and Illegal Cyber Content
Strasbourg, November 28
* Digital Media Revolution in the Americas
Pasadena, November 29 - December 1
* 4th International Conference of Asian Digital Libraries
Bangalore, December 10-12
The Free Online Scholarship Newsletter is supported by a grant from the
Open Society Institute.
This is the Free Online Scholarship Newsletter (ISSN 1535-7848).
Please feel free to forward any issue of the newsletter to interested
colleagues. If you are reading a forwarded copy of this issue, you may
subscribe by signing up at the FOS home page or the FOS Newsletter page.
FOS home page, general information, subscriptions, editorial position,
FOS Newsletter, subscriptions, back issues
FOS Discussion Forum, subscriptions, postings
Guide to the FOS Movement
Copyright (c) 2001, Peter Suber
** If you receive this newsletter by email, then please delete the "easy
unsubscribe" footer (below) before forwarding it to friends or
colleagues. It contains a code identifying you as the original recipient
of the email. If someone down the forwarding chain clicks on the
unsubscribe link, then you will be unsubscribed. **