On Sun, 3 Sep 2000, Andrew Odlyzko wrote: > 1. I think Joseph is right that Stevan overemphasizes the role > of conventional peer review. This is one of the areas where Stevan > and I do not see eye-to-eye.
Fortunately, it doesn't matter whether we see eye to eye on either the current causal role of peer review or its future. The objective here is to free the current research literature, such as it is, both pre- and post-refereeing, online. Self-archiving it all will do this. If the peer review has indeed been superfluous all along, that can and will eventually come out in the wash; but it has no bearing on the (agreed) desirability of self-archiving both preprints and postprints NOW. To put it another way: Self-archiving has FACE-VALIDITY, now, as a matter of FACT. The causal role of peer review in the current quality of the literature, and the future course it will take online, are a matter of conjecture, and one upon which nothing we can and should do now hinges! So why are we even discussing it? Why is the face-valid desideratum of self-archiving the current literature so as to free it being linked in any way to speculations about peer review? For it is (and has been) all too easy for opponents of freeing the literature to dismiss self-archiving as "unfiltered, chaotic, etc." IN CONTRAST TO the filtered, orderly refereed literature we have now. This groundless objection is immediately neutralized if we point out that it is that VERY SAME LITERATURE that is being freed! But if it is made to look (for purely speculative reasons) as if peer review is indeed being tampered with or abandoned as a precondition or component of self-archiving, the horses are again lulled into Zeno's Paralysis and Adypsia. So can we save the hypotheses until the current literature is freed, and start worrying about peer review reform only afterward? > Starting with my original paper > on this subject, "Tragic loss or good riddance ..." (available, > along with all my other papers referenced later, at > <http://www.research.att.com/~amo/doc/eworld.html>), I have been > predicting that as we move to a continuum of publication, we will > also evolve towards a continuum of peer review. So have I! But these untested hypotheses about how to IMPROVE the literature have no bearing whatever on the face-validity of the objective of freeing the current literature, such as it is! On the contrary, by needlessly injecting (debatable) points of inference and conjecture, they weaken what would otherwise be only incontrovertible facts! 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). http://www.ecs.soton.ac.uk/~harnad/Papers/Harnad/harnad90.skywriting.html > Scholars will be > increasingly relying on a variety of cues to judge the validity of > communications from fellow scholars, not just the formal editorial > and peer review system. No doubt. But do we need to wait until this is proven to be the case to everyone's satisfaction before we can take the simple step of self-archiving the CURRENT literature, such as it is, to set it free? > However, that system is slow to evolve, > since the inertia of the scholarly community is enormous, as > I discussed in "The slow evolution of electronic publishing." This conflates two factors: The slowness of publishers to change their ways (often for economic reasons) and the slowness of peers to change their ways (peer review, and reliance on it). Fortunately, the subversive proposal requires them to change NOTHING, in either respect. Self-archiving is simply a (subversive) supplement to the rest, which proceeds apace (journal publication, peer review). Authors do not give up their journals; journals do not give up their peer review practices. Readers simply get it all for free online. (Publishers may eventually have to downsize, but that too is a future possibility, nothing that needs to be done now). Harnad, S. (1995) Universal FTP Archives for Esoteric Science and Scholarship: 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. http://www.arl.org/sc/subversive/ > The incentives for scholars, especially established scholars > with tenure, who already have well-developed methods for reaching > their closest colleagues, to change their ways are not all that > great. Hence we still have a thriving traditional journal system, > (although an increasing fraction of those journals are available > electronically, but as "shovelware," just electronic versions > of the paper editions). (And for a price: S/L/P). But note that scholars are not being asked to give anything up; they are merely being asked to self-archive, on top of it all. (At least that is all they are being asked to do by me!) They are not being asked to submit to different journals, online journals, non-refereed journals, non-journals. They are merely being asked to self-archive what they do submit (pre and post). The enhancements can and will come later. The only enhancement needed now is free online access for all (to this special literature). > In spite of the slow pace of evolution of basic scholarly articles, > I wholeheartedly endorse Stevan's (and Paul Ginsparg's) efforts > to stimulate usage of eprint archives. I think Joseph is wrong > about the need for some special kind of maturity on the part of > a scholarly area to adopt such archives. I have seen many fields > successfully switch to the use of such archives, and I am not > aware of a single instance of any area that has then abandoned > it, nor even of any area where a substantial number of people > regret the move. Much more common are comments of the form > "I don't know how we ever lived without this." I agree. And, a priori, the only relevant question was: Is there any field that would NOT benefit from having its literature freed? If the answer is no, then there are no relevant field differences. > 2. In the general evolution of electronic communication, there will > be considerable pain for publishers and librarians. Why for librarians? Can they not do something better with their money than what is now being spent in the "serials crisis"? Perhaps you are thinking of their time rather than their money. But that too is hypothetical. I doubt that you will find a single librarian who would prefer to perpetuate the serials crisis and prevent the online freeing of the journal literature because of worries about what they will do in the world of free journals. (Perhaps I am wrong about this?) With journal publishers it is another story, and I regret that it must be so, but it is: there is now a conflict of interest between the old, Gutenberg way and the new PostGutenberg way, and current publishers' revenue streams can only be protected at the expense of what is both optimal and attainable for research and researchers. That conflict of interest has to be resolved, and I think it is fairly clear that the resolution must be in favour of research, not of protecting publishers' current revenue streams. But downsizing does not mean extinction. Moreover, although (and I am speculating now: freeing the give-away research literature by self archiving is in no way conditional on this) although implementing quality-control and certification (QC/C) is likely to be the only ESSENTIAL service left for publishers to perform for give-away researchers in the online era, there may well be ADD-ONS that can continue to be sold for S/L/P for some time, perhaps forever. But those add-ons must be called and treated as what they really are, namely, OPTIONAL ADD-ONS, not OBLIGATE "ADDED-VALUE," as it is currently fashionable to call them, as a justification for continuing to hold this give-away literature hostage to access tolls. But yes, this may still entail considerable pain for publishers, as Andrew says. (Is it then justified to try to prevent self-archiving in order to prevent that pain?) > That will be > outweighed by the gains for scholars and the general public, but > those large gains will be spread much more thinly. As a result > there are few people with a large interest in pushing for a rapid > change. I disagree. Refereed journals are neither written for nor read by the general public. They are written by researchers, for researchers. And researchers are the ones who would gain from the freeing of this literature. (The public too would gain, but only indirectly, as its benefits from scholarship/science always are.) It is every scholar who has ever sought, and failed to gain, access to any piece of literature to which his institution does not provide him with access who would gain -- and gain in each individual instance (each potential "click") in which he seeks, or would seek access to any such piece of literature. I think the potential gains are monumental; it is only our imaginations that are failing us in conceiving them. > The one exception is university administrators. Starting > with "The tragic loss ..." I have been predicting that dramatic shifts > might come when these decision makers see that traditional journals > are nowhere as important as they used to be, and start imposing > serious cuts on libraries. Given recent developments (described in > "Competition and cooperation: Libraries and publishers in the > transition to electronic scholarly journals") I think the first > phase will involve elimination of paper versions of traditional > journals, to cut down on internal library costs. As far as I know, the economics of online institutional subscriptions, site-licenses, and pay-per-view (S/L/P) have not left the universities better off than before. To cut paper subscriptions is not to cut expenses. The serials purchasing budgets must shrink, and they are shrinking, but access is shrinking with them. S/L/P cuts might help lower S/L/P prices somewhat, but there is no hope of their reaching a range where all institutions can afford all of the journal literature. Why the greedy "all"? Because ALL of this research literature is a give-away from the author/researcher's standpoint; so there is no earthly reason it should continue to be a buy-back (to that researcher, or to any other researcher) if there is any means of freeing it for all. And there is such a means -- a subversive means. To wait for "market forces" to drive the costs down to the minimal essentials (i.e. QC/C) via library serials spending cuts is to wait till Doomsday, I believe. And there is no need for research and researchers to wait that long. > It is hard to > predict when this might occur, since this depends not only on > the natural evolution of scholarly communication, but also on > the general economic conditions. It is hard to impose drastic > cuts on anything at universities today, in the days of plenty, > but when the next recession hits, it will be a different story. > In "Tragic loss ...," written in 1993-4, I ventured a guess > that there would be little visible change in scholarly publishing > in the first 5 years, but that there would be drastic changes > within 15 years. The first part of that prediction has proved > pretty accurate, now we just have to wait to see about the > second. Contrast this with freeing it all overnight, tonight, by self-archiving... > 3. Although change in traditional publications is slow, I am > convinced it is inevitable. In particular, the paper "The rapid > evolution of scholarly communication," which was mentioned on > this list recently by Tom Walker, discusses the vigorous growth > of electronic scholarly communication aside from the traditional > journal system. Although the incentives for scholars to have > their work freely accessible are weak, they do exist, and, as > with all network effects, get stronger the more people participate. > Archiving of preprints does substantiate priority claims, as > Joseph mentions in his most recent message. That is a substantial > factor to quite a few people I know, although less for the reason > Joseph mentions (that a referee might steal the results), and more > to protect themselves against a slow and/or capricious refereeing > process (such as when a paper is submitted to an ultra-selective > journal A, gets rejected after a year [yes, such delays are > not uncommon in some areas], and gets sent to journal B, while > in the meantime somebody else publishes similar results in > journal C). But why set up any opposition or trade-off between the author's desire and right to seek refereeing and acceptance in the established journals for his give-away research, and his desire and right that everyone should have free access to it? Neither publishers nor peer review need to reform so researchers can have their cake and eat it too; all that is needed is subversive self-archiving of both the pre-refereeing and the post-refereeing research. > The main incentive for scholars to insist on free access > to their papers, though, is [that] this is the way to get them read. At last! > Easy availability is essential on the Internet. Some criticize > this trend, claiming it produces superficial impressions, an > overload of material, etc., but instant gratification (even > if it "saps the moral fiber of society," in the words of > critics of some earlier technological innovations) is the > way we are going. The criticism sounds silly to me: as if the S/L/P access barriers were what was maintaining quality, rather than the QC/C! (By the same token, even conventional libraries are bad for book readers: their reading needs to be restrained by bookstore prices, to guard their "moral fiber: against "superficial impressions"...) It is a head-shaker that we are reduced to even contemplating such nonsense. (I don't mean Andrew, who of course finds this sort of Zeno-rationale as ridiculous as I do.) > "The rapid evolution ..." is largely > about this phenomenon. Interestingly enough, the paper by > Karen Hunter from Elsevier at the PEAK conference (which is > where "The rapid evolution ..." was presented, both available > at <http://www.si.umich.edu/PEAK-2000/program.htm>) has a nice > quote on what Elsevier has learned from the PEAK experiment > (which involved various types of payment schemes for access > to scholarly journals): > > [Elsevier's] goal is to give people access to as much > information as possible on a flat fee, unlimited use > basis. [Elsevier's] experience has been that as soon > as the usage is metered on a per-article basis, there > is an inhibition on use or a concern about exceeding > some budget allocation. > > Now Elsevier is interested in perpetuating its model, so > Karen talks of flat fee pricing, but the general principle > is clear. Flat fees encourage use, and zero fees do that > even better. Indeed. The publishers' fantasy is a click-through oligopoly, funded by global licensing agreements. Let that option be fairly weighed as an add-on, when it competes against authors' free self-archived "vanilla" versions of all the very same papers. If the add-on still has a market, who can possibly raise an objection? The give-away literature will be available free, as home-brew, and those who prefer to pay instead for high-end versions will be able to do so. And Andrew and I can retire to our tents and get back to our research, reaping the benefits of the give-away corpus, free at last! -------------------------------------------------------------------- Stevan Harnad [email protected] Professor of Cognitive Science [email protected] Department of Electronics and phone: +44 23-80 592-582 Computer Science fax: +44 23-80 592-865 University of Southampton http://www.ecs.soton.ac.uk/~harnad/ Highfield, Southampton http://www.princeton.edu/~harnad/ SO17 1BJ UNITED KINGDOM NOTE: A complete archive of this ongoing discussion of providing free access to the refereed journal literature is available at the American Scientist September Forum (98 & 99 & 00): http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html You may join the list at the site above. Discussion can be posted to: [email protected]
