AAAS's Response: Too Little, Too Late Harnad, S. (2001) AAAS's Response: Too Little, Too Late. Science dEbate.

dEbate response to: The Editors (2001) Is a Government Archive the Best Option? Science 291: 2318b-2319b


Stevan Harnad
Intelligence/Agents/Multimedia Group
Department of Electronics and Computer Science
University of Southampton
Highfield, Southampton
The goal (so we are all clear on what we are talking anout here) is to free all of the refereed scientific and scholarly literature online for everyone, forever, from the obsolete and unnecessary access- and impact-blocking tolls of the Gutenberg era. This anomalous literature has in any case always been an author give-away, written for research impact, not for income from the sale of the text. In the PostGutenberg Galaxy it has become possible to liberate this literature at last.

The release of the contents of Science after a delay of 12-months is too little, too late.

True. But this is not the beginning of the conversation, which was already well underway with the Bachrach et al. Science Policy Forum of 1998 followed by the 1999 original NIH proposal by Harold Varmus, on which the former Editor of Science had already written an editorial an editorial rather similar to the one I am replying to here, in a reply rather similar to the one I made to the prior editorial:

A conversation requires some give and take... We will now proceed to a consideration of these faster and more effective alternatives, while asking only how the question of "risk to nonprofit scholarly publishing" got into this: First of all, we are only speaking of refereed journals. Second, if the only way to free their contents online were to restructure journal publishing in some way, would the benefits to research and researchers from freeing the refereed journal literature necessarily be outweighed by the (putative) difficulties the changes might create for the journal publishers?

There are problems, however, with the Roberts et al. proposal: Although a free online version of the entire refereed corpus would undoubtedly be beneficial to the world scientific and scholarly community, this proposal does ask both publishers and authors to give something up in exchange: Publishers are asked to give up their contents online, and authors are asked to give up those publishers who decline to do so. If these sacrifices are necessary in order to gain the benefit of a free refereed literature, then we can weigh them, along with the likelihood that the parties involved will be willing or even able to make the sacrifices. But are the sacrifices really necessary?

The answer is that they are not, for there is an alternative way to free the entire refereed literature without asking anyone to give up anything, and that is through author self-archiving. The strategy has already been tested and demonstrated to work by physicists. They have already freed 30-40% of their literature in this way. All that is needed is for physicists to accelerate their own rate of self-archiving (which, at its current linear growth rate, would take another decade to free 100% of its refereed literature) and to extend it to all the other disciplines.

Physics self-archiving began as centralized (in the Los Alamos Archive and its 14 mirror sites worldwide). What can now accelerate and extend the self-archiving initiative to all the other disciplines is the Open Archives Initiative (OAI), which has designed a standard for metadata tagging and harvesting that makes distributed interoperable Archives possible at the individual University and Research Institution level; interoperability means it can all be harvested into a global "virtual" archive, its full contents seamlessly searchable and accessible for free from any researcher's desktop. Institutions can now create OAI-compliant Eprint Archives using free, open-source software: The responsibility and the incentive and the initiative for self-archiving can then be distributed worldwide at the University level, where its cost per paper becomes negligible, whereas its benefits in terms of increased accessibility, visibility and research impact are appreciable (not to mention its eventual potential to relieve the institutional libraries' serials crisis).

Yes, and that is precisely the problem -- both with the status quo, and with waiting for journal publishers to take charge of freeing the refereed literature online. Because the problem is not that it is not integrated but that it is not free! And why should tax dollars be used to integrate a scattered set of toll-gated sites when researchers can both free and integrate the entire refereed literature (over 20K refereed journals) by self-archiving their own portion of it in their own institution's registered, OAI-compliant Eprint Archives? Correct. So there is not much incentive for journals to give away their contents at this time. Author/Institution self-archiving, however, may eventually have the secondary effect of forcing publishers to restructure themselves, and scale down to providing only the essentials (quality control and certification [QC/C] through refereeing), which only account for 10% of journal costs. The rest (on-paper version, on-line PDF, other "added values") can be sold as optional add-ons as long as there is a market for them.

Currently, both the essentials (QC/C) and the add-ons are "wrapped" into the same product, with the result that the refereed papers are held hostage to the add-ons, which are kept behind a financial fire-wall and paid for by Subscription, Site-License, or Pay-Per-View (S/L/P) tolls.

It is from these S/L/P access-barriers that refereed research must be freed, and the arithmetic is already clear: If and when the availability of the free online version of refereed papers causes publisher S/L/P revenues to shrink (and institutional S/L/P savings to grow) to the point where there is no longer enough money to pay the essential 10% QC/C costs out of the S/L/P revenues, then they can be covered by the institutions out of 10% of their annual windfall S/L/P savings in the form of per-paper fees paid to journals for the QC/C service for their own authors' papers.

In other words, there is no benefit whatsoever to research and researchers in maintaining "an economically important source of online traffic for journals" when there is clearly an alternative that can provide the QC/C and free online access too. Journals that are not interested in downsizing to this new PostGutenberg niche can elect to pull out, in which case their editorial boards, referees, authors and titles can migrate to new journal publishers that are happy with the new niche.

But this endstate is unlikely to be reached either by publisher voluntary downsizing now, or by author migration to new journals (although it would be splendid if it could). It will only be reached under pressure from the natural force of author/institution self-archiving.

How? Why? Is there any evidence whatsoever for this among the 150,000 papers already archived by the physicists? And how can free online access to already refereed papers have a retroactive effect on the refereeing? Backwards causation? Where did the notion of "monopoly supply" come into a free online literature? If the worry is about the robustness of the archive, mirroring, distributedness, backup and other means exist to make it as robust as one likes.

And have the editors of Science not noticed that journals occasionally slip on deadlines too? In any case, whatever it is that makes the current journal literature reliable and robust, we can be sure that it is not eo ipso the fact that S/L/P meters are running.

That is correct. And that is why the self-archiving initiative neither demands nor depends on journal publishers doing anything like that. The only essential value is the peer review, discussed above. The rest are options, and can and should be sold wherever and whenever there is a market for them. Indeed it would. And for that reason it is unreasonable to ask or expect AAAS to do so at this time. All AAAS need do is to refrain from any attempt to prevent their own authors from self-archiving their Science papers. The rest of the cards can fall where they may.

But if Science authors were asked forthrightly whether they wish to continue subisdizing AAS's "other activities" at the cost of their own lost research impact (because of the S/L/P toll-barriers), is there any doubt at all what their reply would be?

If, however, researchers were ever asked, clearly and directly, whether they would like to see the accessibility and impact of their research continue to be held hostage to all these other "good works," is there any doubt whatsoever as to what their reply would be?

Note that the above rationale, and all the other ones offered here, good and bad, are understandable ones for journals' declining to comply with the PubMed Central proposal. But they are strained and ineffectual when it comes to self-archiving, and its possible consequences.

The earlier formula -- that the 10% QC/C costs can always be paid out of the 100 S/L/P savings -- covers all refereed journals, great and small.

This is so strained and far-fetched that it does not warrant a reply. Ditto. This sort of convoluted scare-mongering is unworthy of AAAS. Voluntarily freeing their contents online after a 12-month delay is certainly a welcome step from AAAS, but it is too little, too late. It is quite understandable that neither Science nor any other publisher should find any reason for pre-emptively freeing its contents online at this time. All they need to do for now is to let their authors take matters into their own hands, by self-archiving their refereed papers. The rest will take care of itself.


Harnad, S. (1990) Scholarly Skywriting and the Prepublication Continuum of Scientific Inquiry. Psychological Science 1: 342 - 343 (reprinted in Current Contents 45: 9-13, November 11 1991).

Harnad, S. (1991) Post-Gutenberg Galaxy: The Fourth Revolution in the Means of Production of Knowledge. Public-Access Computer Systems Review 2 (1): 39 - 53

Harnad, S. (1994) A Subversive Proposal. In: Ann Okerson & James O'Donnell (Eds.) Scholarly Journals at the Crossroads: A Subversive Proposal for Electronic Publishing. Washington, DC., Association of Research Libraries, June 1995.

Harnad, S. (1998a) For Whom the Gate Tolls? Free the Online-Only Refereed Literature. American Scientist Forum.

Harnad, S. (1998b) On-Line Journals and Financial Fire-Walls. Nature 395(6698): 127-128

Harnad, S. (1998/2000) The invisible hand of peer review. Nature [online] (5 Nov. 1998) Longer version in Exploit Interactive 5 (2000):

Harnad, S. (1999a) Free at Last: The Future of Peer-Reviewed Journals. D-Lib Magazine 5(12) December 1999

Harnad, S. (1999b) Advancing Science By Self-Archiving Refereed Research. Science dEbates [online] 31 July 1999.

Harnad, S. (2000a) E-Knowledge: Freeing the Refereed Journal Corpus Online. Computer Law & Security Report 16(2) 78-87. [Rebuttal to Bloom Editorial in Science and Relman Editorial in New England Journal of Medicine]

Harnad, S. (2000b) Ingelfinger Over-Ruled: The Role of the Web in the Future of Refereed Medical Journal Publishing. The Lancet Perspectives 256 (December Supplement): s16.

Harnad, S. & Carr, L. (2000) Integrating, Navigating and Analyzing Eprint Archives Through Open Citation Linking (the OpCit Project). Current Science 79(5): 629-638.

Harnad, S., Varian, H. & Parks, R. (2000) Academic publishing in the online era: What Will Be For-Fee And What Will Be For-Free? Culture Machine 2 (Online Journal)

Odlyzko, A.M. (1998) The economics of electronic journals. In: Ekman R. and Quandt, R. (Eds) Technology and Scholarly Communication. Univ. Calif. Press, 1998.

Odlyzko, A.M. (1999a) Competition and cooperation: Libraries and publishers in the transition to electronic scholarly journals, A. M. Odlyzko. Journal of Electronic Publishing 4(4)

Odlyzko, A.M. (1999b) The rapid evolution of scholarly communication," to appear in the proceedings of the 1999 PEAK conference.