The following is a response to comments by G.F. Humphrey, University of Sydney
which appeared in The Australian
http://www.theaustralian.news.com.au/
(paper edition) on 19 March 2003, Page 038. The comments are on my
article, which appeared there 12 March:
on
http://www.ecs.soton.ac.uk/~harnad/Temp/unto-others.html
> Part of [Harnad's] solution is the relegation of publishers to the task of
> peer review, their funding being from the universities whose members
> provide the papers. The universities would save two-thirds of the cost
> of journal subscriptions.
My solution (to the problem of research impact being needlessly lost
because of toll-gated access) is the self-archiving of all peer-reviewed
research in each researcher's institutional Eprint Archives, to maximize
its access to would-be users, and thereby maximize its usage and impact.
Whether and when journal publishers must downsize to becoming peer-review
service-providers depends on whether and when the market for their
other services and add-ons (paper version, publisher's PDF, mark-up,
citation-linking) shrinks to the point where it can no longer sustain
the essential cost of peer review. No one
knows either whether or when that will happen
http://www.nature.com/nature/debates/e-access/Articles/harnad.html#B1
but meanwhile research access and impact will already have been maximized
by self-archiving.
"Distinguishing the Essentials from the Optional Add-Ons"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1437.html
> Unfortunately, the universities would also have to fund copy editing
That depends on whether copy-editing turns out to be part of the
essentials. If so, it can be wrapped into the peer-review service cost.
But it is not clear that the rather low level of copy-editing being
practised by many journals today (mostly just "which-hunting") is an
added value at all. Reference-checking (using the full, interoperable,
open-access database) will become increasingly automated with the entire
refereed literature openly accessible online, as will format-checking. And
some XML markup will no doubt soon be part of authoring tools, as html
already is today.
> No specific indication was given as to how this knowledge (hopefully peer
> reviewed) is to come about.
We are talking about peer-reviewed journal articles. Whether they are
on-paper or on-line, toll-access or open-access, has nothing to do with
whether they are peer-reviewed. This is either a red herring or an
inadvertent conflation of pre-peer-review preprints with peer-reviewed
postprints. Self-archiving is recommended for both, but which is which
is clearly tagged on-line, just as it was on-paper. The main purpose
of the self-archiving is open access to all peer-reviewed postprints.
http://www.eprints.org/self-faq/#What-self-archive
> "Researchers are paid to do research but not to report it."
> Incorrect. Research is not complete until it is published. Salary
> and expenses continue during the writing period.
G.F. Humphrey has misunderstood this point. Of course researchers are
paid (by their institutions) to publish or perish. That is what research
impact (and the motivation for maximizing it through open access) is all
about. The point was that they do not get paid royalties or fees by
their *publishers* in exchange for the sale of their work, as most other
writers do.
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#1.2
> Further, grant agencies base allocations on publishing history, partly
> since, as Harnad says, research input is measured by counting how many
> researchers use and cite the work.
But once again, the point is being missed: That impact-income (shall we
call it) is not coming from publisher toll-income. On the contrary, the
toll-gating of the author's give-away work is *diminishing* impact
income!
> Agencies significantly measure the worth of papers according to
> journal reputation. This criterion would not be available if an author's
> university were employing the group doing the peer reviews. It would
> be better if the universities employed autonomous scientific bodies to
> arrange peer reviews -- for example, the academies.
Again, the article has been misread. There is no proposal to cease
publishing in exactly the same peer-reviewed journals that researchers
are publishing in today. The proposal to self-archive peer-reviewed
research is not a vanity-press or in-house peer-reviewing proposal!
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#1.4
The proposal is merely to self-archive the peer-reviewed paper itself
too, to make it open-access (for all the would-be users worldwide whose
institutions cannot afford the access-tolls).
http://www.soros.org/openaccess/read.shtml
In addition, it is merely pointed out in passing that if and when the
cost of paying for the implementation of the peer-review alone can no
longer be covered from the access-toll-revenues (paid currently by
institutions for access to the *incoming* peer-reviewed research from
*other* institutions), it can easily be covered out of only a portion
of the annual institutional windfall toll-savings, in the form of a
service charge (paid to the [autonomous] publisher, as now) for
implementing the peer-review of each institution's own *outgoing*
research.
> "Research papers are similar to advertisements -- they bring rewards."
> Yes, often, but they usually bring justified adverse criticism, sometimes
> exposure of authors as frauds, and nearly always attract review changes.
All true, but not relevant, as the proposal is not to alter
peer-review but to maximize access, and hence visibility, usage, and
impact. (Open-access also maximizes the self-corrective feedback cycles,
supplementing peer review; Dr. Humphrey, who has published on the
detection of research fraud, should welcome this!)
> Further, the advertiser pays to get publication; never so the researcher,
> although occasionally the research grant pays page charges.
The statement was that peer-reviewed publications are *similar* to
advertisements, not that they are *identical* to them. Advertisements
are not peer-reviewed either; nor are researchers advertising a
product or service for sale. The point was that just as it would be
counter-productive to toll-gate *access* to advertisements, written to
maximize sales impact, it is counter-productive to toll-gate access to
peer-reviewed research, written to maximize research impact.
> Fortunately, we are later told that there has to be peer review. So
> just to [self-archive all research] would have to involve an academy,
> copy editing and a university publications committee.
This is the same misreading as before. The proposal is to self-archive
all peer-reviewed research; the peer-review continues to be implemented
by the autonomous journals, as it always was.
> With smaller incomes, publishers (that is, profit-making companies) might
> need higher profit ratios on the diminished incomes, thus increasing
> costs to perhaps one-half.
This is of course all hypothetical. What is actual (and tried and true)
is that open-access can be attained right now, through self-archiving.
Whether and when this will diminish publishers' toll-incomes, and what
can then be done to cut costs and cover the essentials (peer-review
service-provision) is a matter of speculation. But even if it were to turn
out that peer-review costs half of the current toll-revenue per article,
rather than under one-third (as I and many others have estimated), that's
still cheaper, still only half of the windfall toll-savings, hence
still affordable by institutions, and the reward is still open access and
maximized impact. Hence this is no argument against self-archiving, nor
for access/impact-blocking tolls.
> "Every journal has a paper edition and an online edition." Only some
> do and there is a user charge for online.
I should have said "just about every journal." (Certainly all the
biggest and most important ones do). And it is the user (access-toll)
charge that this is all about, whether on-paper or online.
> Electronic publishing... will replace hard copy in a
> decade... Nevertheless, there are many questions to be resolved. It is
> certainly a boon for researchers. No more grubbing around in the library!
Electronic publishing is a foregone conclusion. Hard-copies will be
user-generated until/unless on-screen reading becomes much improved
(but searching and browsing are infinitely better on-screen). That
is all old news, however. The new news is that access-toll-barriers
(for author-give-away writing, of which peer-reviewed research is the
most important example), and the potential usage and access they block,
are no longer necessary in the online era, and their negative effects
can already be eliminated through self-archiving (today, in fact
yesterday).
Stevan Harnad
Received on Wed Apr 02 2003 - 15:54:11 BST