Re: The Ultimate Danger of SkyReading/Writing

From: Albert Henderson <>
Date: Thu, 29 Nov 2001 11:03:34 +0000

on 27 Nov 2001 Stevan Harnad <> wrote:

> AH> The confusion comes in the link between the 'freeing' of the
> AH> scientific record and the [1] debasement of libraries, the
> AH> [2] demolition of journals, and the [3] shifting of costs
> AH> from institutions to individuals.

> sh> I regret I have nothing to say about debasement and demolition, but I
> sh> can definitely say that I have never said anything about shifting
> sh> costs from institutions to individuals! I have suggested redirecting a
> sh> small portion (10-30%) of INSTITUTIONAL annual windfall savings from
> sh> subscription/license cancellations for incoming research to
> sh> INSTITUTIONAL annual peer-review costs for outgoing research:

Here is a dream that will never come true! History teaches us that any
windfall in higher education -- real or imagined -- always goes to
profits, never to peer review or the pursuit of knowledge.

The appearance of the Xerox 914 in the 1960s justified the draconian
slashing of library spending as noted by a number of studies. We now
see that average profitability doubled while libraries have less than
half their 1960s allocation. The downward spiral of library spending
will soon end. Okerson and Stubbs projected in 1992 that major
university libraries would no longer buy either books or journals by

Moreover, the poor researcher has lost out -- waiting weeks for
interlibrary photocopies, according to studies by OCLC, ARL, and
others. So much for windfalls providing any benefit for education or

Long before the Internet, King, McDonald and Roderer pointed out that
the major costs of journals and monographs have always been in
preparation. This is particularly true for the most specialized
journals. They estimated average prerun costs at $1050 per article in
1977; runoff costs for subscription copies came in at $0.09.
[SCIENTIFIC JOURNALS. 1981 p. 218- 219] Even if, loaded with expensive
computers, one could save 9 cents per article, financial support for
the other $1050 must come from somewhere. If it does not come from
libraries, then where?

> AH> The confusion comes when 'liberation' as a utopian goal results in
> AH> 'anarchy' with all evils permitted in lawlessness.

> sh> I regret I have nothing to say about anarchy and lawlessness...

Yet you advocate the 'every man for hisself' policy of self-publication
as an alternative to libraries. Your solution would be more 'free for
all' than 'freeing the literature.' Libraries _are_ the main source of
free dissemination of the scientific record. Why not support them?

> AH> Peer review, if any, will operate differently in Harnad's 'sky'
> AH> than it does currently.

> sh> That would be odd, as it is the peer-reviewed articles to which
> sh> online access is freed by self-archiving them. (Does the act of
> sh> self-archiving, one wonders, produce some remarkable transformation
> sh> in those texts?)

Like dumping swamp water into the punch, the mixture of unrefereed
drafts with journal articles sullies the entirety. The scientific
record needs to be protected, as do its readers.

> AH> The freedom of self-archiving informal communications together with
> AH> formal publications will admit any paper from any author without
> AH> consultation of editors and referees.

> sh> Admit to what? The sky? But there's plenty of stuff up in the
> sh> sky already. There's plenty of room, and users will be just as
> sh> capable of reading labels on-line (such as "Journal of ...") as
> sh> they were on-paper. (Or does free on-line access produce some
> sh> remarkable degradation of these powers of discrimination?)

Well, you must admit that there is plenty of junk up there. You seem to
be proud of it. Who will regulate the use of "journal of" labels for
advertising that no science library would have? JAMA reported that the
FDA believed that even physicians lack the ability to recognize
promotions printed by manufacturers. FDA demands that manufacturers
label reprints "advertising," and restrict such reprints to
peer-reviewed journals. And yes, free online access degrades the powers
of discrimination, as many teachers have repeatedly pointed out.

Many students seem to believe that the Internet solves everything....
not to mention the Johns Hopkins scientist who killed a subject with
the ignorance of limited Internet resources not long ago.

> AH> Any refereeing of informal communications is done publicly or,
> AH> more usually, not at all.

> sh> And your point is...?

The point is that you have invented a world that
doesn't exist. For instance:

> sh> (Meanwhile, most pre-refereeing preprints are submitted to
> sh> journals and undergoing peer review, until the final, refereed
> sh> draft is ready to be self-archived with the label "Journal of
> sh> ...")

The research into informal communications over the last 30 years
demonstrates that much if not most research described informally is
never submitted and never published. For example:

+ In 1959, the International Conference on Scientific Information found
that 48.5% of the research covered by the 383 papers presented at the
1948 and 1949 annual meetings of the Optical Society of America and the
1949 national convention of the Institute of Radio Engineers, and the
1950 meeting of the American Physical Society were published and
abstracted 5 years later.

+ In 1969, the American Psychological Association reported that 25% of
the authors of printed conference papers delayed or decided against
submission to a journal.

In 1998, JAMA reported that only 20% of the unpublished studies
originally submitted to a medical specialty meeting were later
submitted as a full manuscript to a journal. Moreover, investigators
were easily dissuaded, submitting a manuscript, to fewer than 2
journals before giving up.

> AH> The absence of criticism may leave the impression that errors,
> AH> duplications, omissions, and rhetoric represent the norms of
> AH> research.

> sh> Absence of criticism of what? Unrefereed preprints are unrefereed
> sh> preprints. Refereed postprints are refereed postprints. Both are
> sh> up there, the latter prominently tagged with the journal name.
> sh> Both will no doubt elicit some commentary. But what is the
> sh> point?

The point is that the individual researcher is already overtaxed by the
massive output of others. Mixing swill with the refined ingredient
does more harm than help.

By undermining the journal and the library out of the picture, we lose
authenticity and the protection of multiple gatekeepers.

> AH> A great deal of research is poorly prepared, as most editors will
> AH> affirm. Self-publishing will sully the record.

> sh> Albert, no matter how many times he is reminded that we are
> sh> talking about the self-ARCHIVING of refereed research, not about
> sh> self-PUBLISHING, will keep on making this dreary non-point...

You also advocate the self-'archiving' of preprints - unreviewed drafts
that may never be published.

> AH> Informal interaction will have little value. My impression is
> AH> that most researchers would rather present better supported
> AH> claims than attack someone else's work in print.

> sh> Those who wish to restrict their reading and writing to the
> sh> refereed literature can and will, even when it is on-line and
> sh> free (why can't they?).

How can anyone restrict themselves to the refereed literature when it
is mixed up with preprints and other unreviewed clutter. An
investigator screens thousands of titles a year to find papers that are
relevant _and_ refereed. There are too many refereed papers now,
without adding preprints from unknown sources.

> sh> And as to whether others will choose to interact more (about both
> sh> unrefereed and refereed research), why speculate? Why not just
> sh> look at the data (e.g.,

> sh> "Comments in Journals"
> sh>

> sh> "Open source tool for online peer review commentary"
> sh>

Thank you for these excellent examples of unrefereed self-serving
promotional materials.

This is just the sort of self-publication I have been warning against.

> AH> Clearly, under Harnad's proposals, the burdens of judgment,
> AH> together with library costs, are shifted from the community to
> AH> the individual.

> sh> Vide supra. (Peer review is unchanged, and institutional costs
> sh> are largely saved, partly redirected. Nothing is shifted to the
> sh> individual except free access to the refereed research.)

More like Fiat Aurum! [Let there be gold]

Institutional managers started to shift costs to the individual in the
1970s. They would have eliminated library spending entirely by now were
it not for the obstacle of copyright.

> AH> This would be the unfortunate reversal of policy going back to
> AH> the beginning of history -- the policy that has supported
> AH> libraries as the disseminators of free information.

> sh> No comment (but it does seem a bit tortured to cast this thinly
> sh> disguised defense of journal publishers' revenue streams as
> sh> support for libraries and free information...).

Free information has been the role of the
libraries, not the publishers.

> AH> Who has the capacity to plow through all the unrefined
> AH> self-published claims and comments? No one.

> sh> And no one needs to. That's what peer review and journal names
> sh> are for...

But your proposals eliminate journals as well as libraries....
'Freeing' libraries from the 'tolls, etc.' leaves journals with no
income and libraries with no need to exist.

> AH> This, the challenge of dissemination and not the exhaustion of
> AH> ideas (predicted by Holton, Horgan and others), will be the end
> AH> of science.

> sh> Albert is sounding more and more apocalyptic. Could it be that he
> sh> has finally realized that his blinkered defense of the
> sh> indefensible is utterly unavailing?

The challenge of dissemination has been very well documented. The
answer to the challenge is more effective journals and libraries, not
undermining them. It is more library spending, not a glib windfall,
that we need.

Thanks for your comments.

Albert Henderson
Received on Thu Nov 29 2001 - 11:03:48 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:19 GMT