A strategy for open access to society publications

Jim Pitman

January 28, 2004


Scholarly societies are typically non-profit organizations which seek a stable and sustainable balance between two basic long term objectives: on the one hand to promote and develop a particular field of knowledge by widespread dissemination of high quality publications in that field, and on the other hand to sustain themselves as organizations and support various other beneficial activities besides publication. As chair of the publications committee of one such society, the Institute of Mathematical Statistics (IMS), I am looking for ways to achieve this balance, and seeking advice from others with the same intent.

The Budapest Open Access Initiative and the Berlin Declaration promote the development of open access electronic journals and digital repositories, free to anyone with internet access. While open access journals maximize dissemination, it is not evident how a scholarly society can sustain their production along with its other activities. It may be possible in some fields to support open access journals by authors or research funding bodies covering the costs with article fees, following the economic model of BioMed Central and the Public Library of Science . But this model does not provide credible competition to commercial journals in fields like mathematics and statistics, where article charges would drive many authors to commercial alternatives, with undesirable consequences for the journal in particular and the field as a whole. One successful means of providing open access in mathematics has been the zero-cash-flow model exemplified by the The Electronic Journal of Combinatorics , Electronic Journal of Probability , and Geometry and Topology . But this model is incapable of generating the cash flow typically required by a society to fund other activities, so is sustainable only as part of a society's publication portfolio.

In seeking the desired balance between open access and revenue flow for a society, SPARC (The Scholarly Publishing and Academic Resources Coalition) has recently proposed a hybrid model, whereby each article in a journal would be made electronically available only to subscribers, unless its author chooses to pay, in which case it would be freely available to all. But in subjects with a well developed open access eprint repository, such as arXiv in mathematics and physics, there is little incentive for authors to pay, since they can self-archive a final or near-final version of their paper on the digital repository for free, with a link to the final version, and that alone largely achieves their interest in the wide dissemination of results. And in a field like statistics, with no tradition of self-archiving , it is widely agreed that one of the best means of knowledge dissemination would be to encourage self-archiving by development of an open access eprint repository in the field.


In view of these considerations, I propose that IMS and other mathematical and statistical societies, such as Bernoulli Society for Mathemathical Statistics and Probability and American Statistical Association , adopt the following policies.

1) Attempt to maintain current subscription revenue from both electronic and print versions of journals, while increasing production efficiency.

2) Encourage and assist authors to place their preprints, up to and including accepted versions of papers, and possibly even final copyedited versions, on arXiv, with links to an offical journal site.

3) Allow authors to place final versions of their papers (exactly as published in the journal) on their own homepage or in an institutional digital repository, with a link to the official journal site (this is already IMS policy).

4) Provide open access to final version at the official journal site only if the author pays a suitable one time article charge. That charge should decline over time since publication in the journal, possibly to zero (moving window).

5) Educate authors that wherever they publish an article, they should place a final accepted version in an open access eprint repository.

6) Educate authors, editors and referees that they should not support journals or publishers which do not allow this practice.

7) Ask members and journal supporters to persuade libraries in financial difficulties to cancel commercial or other journals with regressive copyright policies, and reject big bundling deals , before cancelling journals run by societies that support open access eprint repositories.

8) Attempt to persuade other societies to follow the same strategy.


Policies 2) and 3) provide nearly all the reward of open access to content via free search engines like Google , the Front for the Mathematics ArXiv , OSTI E-print network , MPRESS and OAIster while the proper indexing and organization of the journal content in the version available only by subscription should provide enough added value for libraries not to cancel their subscriptions. In the case of statistics, some initial effort will be necessary to create a statistics section of arXiv, but this should increase the accessibility of the subject to scholars in the many fields where statistics is applied.

Each society maintains control over the charge stucture in 4). That structure can be adjusted adaptively over time to control the society's revenue stream.

Policy 7) involves an alliance of societies and libraries, represented by SPARC (The Scholarly Publishing and Academic Resources Coalition) which supprts alternatives to high priced commercial publication. This alliance serves the interests of both parties, as it puts pressure on commercial publishers to allow self-archiving. If authors routinely self-archive their content, with support from libraries and professional societies, this will effectively place limits on what commercial publishers can charge for subscriptions to professional journals.


Thanks to Chris Burdzy for his philosophy of Bottled water which underlies the strategy proposed above, and Laurent Siebenmann for the following succint expression of this philosphy in his letter math research on tap :

(a) Mathematical research articles should be available to the public as freely as ''tap water'' -- because every mathematician must have access to all known results in his field to be efficient and competitive.

(b) The role of the commercial and non-commercial publishers should be to produce and distribute an upgraded version of the product analogous to ''bottled water''.

(c) The mathematical societies/foundations, the universities, and the government should make it mandatory or at least standard practice, for mathematicians they support in any way, to deposit the raw research in a permanent ''tapwater reservoir'' like arXiv .
The strategy outlined above is intended to provide practical steps for a professional society to support this philosphy, while continuing to sustain itself as an organization. More broadly, the academic community should look for opportunities to assist foundations, universities, and governments to follow suit. One current opportunity is provided by the UK Parliament Science and Technology Committee: Scientific Publications , which is inviting suggestions of what actions government, academic institutions and publishers should be taking to promote a competitive market in scientific publications. Evidence should submitted by February 12, 2004.

About this document ...

A strategy for open access to society publications

This document was generated using the LaTeX2HTML translator Version 99.2beta6 (1.42)

Copyright © 1993, 1994, 1995, 1996, Nikos Drakos, Computer Based Learning Unit, University of Leeds.
Copyright © 1997, 1998, 1999, Ross Moore, Mathematics Department, Macquarie University, Sydney.

The command line arguments were:
latex2html -split +0 -no_navigation strategy.tex

The translation was initiated by Jim Pitman on 2004-01-28

Jim Pitman 2004-01-28