Re: The 1994 "Subversive Proposal" at 15

From: Stevan Harnad <amsciforum_at_GMAIL.COM>
Date: Wed, 2 Dec 2009 14:34:36 -0500

On Wed, 2 Dec 2009, Stevan Harnad wrote:

> On June 27, 1994, Stevan Harnad wrote:

> SH2: What on earth do you mean by "esoteric"? Are we supposed to have
> different criteria for a publication depending on how big a readership
> it is likely to have? In that case we need a sliding scale whose value
> we cannot possibly know in advance for every candidate piece of writing.

SH3:
Just dull-wittedness. It should have been obvious already then that the
primary target was refereed journal articles and that "esoteric" was a
red herring.

> SH2: Paper publishing? Is this, then, merely about getting published
> articles online? That's not likely to be a very radical proposal, since (today, in 1994) it is surely a foregone conclusion that publishers will all have online editions within a few years.
> [So is this about] online, online only, or free online?

SH3:
More somnambulism: It should have been clearly stated as "free online
access to refereed journal articles" (i.e., OA).

> SH2: Give-away writing might be a natural kind, but what distinguishes
> give-away writing from non-give-away writing? How does one recognize it
> in advance? And surely the distinction is not just based on probable
> market but on some other aspect of academic motivation. After all,
> textbooks are as "academic" as one can get, yet textbook authors are
> certainly motivated to sell their words, otherwise many would not do the
> work of writing them.

SH3:
Addle-brainedness, yet again:

Refereed research articles are written purely for research usage and
impact not for sales revenue. That's how you distinguish them. And you
recognize them by the journal-names.

> SH2: Who are "peers"? And what is the reason for this obsession with reaching their "eyes and minds"? The fact that they are all in some sort of "esoteric" club surely is not the explanation.

SH3:
Peers are the fellow-researchers worldwide for whose usage peer-reviewed
research is conducted and published.

"Eyes and minds" should have been research uptake, usage and impact
(e.g., as measured by downloads and citations).
http://opcit.eprints.org/oacitation-biblio.html

> SH2:
> And this "building on one another's contributions" sounds cosy enough,
> but what is really going on here. It's certainly not about verbal Lego
> Blocks!

SH3:
Research uptake, usage, applications, citations.

> SH2: Fine. These authors are saints, or monks. But why? For what?

SH3:
Their research progress, their funding and their careers are based on
the uptake and usage of their research findings, not on income from the
sales of their writings. (User access-barriers are also author
impact-barriers.)

> SH2: The criterion sounds like it's esotericity itself, but why? Besides, that's circular: Is give-away writing esoteric because
> its target readership is tiny? Or is its target readership tiny because
> the writing's esoteric?

SH3:
Fuzzy thinking again: Esotericity, though roughly correlated, is a red
herring. Give-away writing is give-away writing, and wants to be freely
accessible online because access-barriers are usage- and
impact-barriers. (Yes, the potential users of most refereed research are
few, but that's not the point, nor the criterion: the need to maximize
usage and impact is the criterion.)

> SH2:
> And FTP archiving sounds fine, but isn't it already obsolete? This is
> June 27 1994, but Tim Berners-Lee created the Web 5 years ago!

SH3:
Ignorance, sir, pure ignorance.

> SH2:
> And there you go again with "electronic publication"? Is this just about
> moving to electronic publication? But that's surely going to happen
> anyway.

SH3:
Fuzziness, pure fuzziness. It is and was about free online access, not
about online publication.

> SH2:
> And is "esoteric" publication, then, merely "vanity press" publication?
> If so, then it's no wonder its likely readership is so tiny...

SH3:
It's about refereed publication, hence not vanity-press. (But there was
definitely muddle and ambiguity regarding unrefereed vs. refereed
drafts. The focus should have been directly on refereed drafts, with
unrefereed drafts being only a potential entry point in some cases.)

> SH2: But physicists (who are doing it on the Web, by the way, not via
> FTP) have already been doing much the same thing (sharing their
> pre-refereeing drafts) on paper for years now, even before the web,
> or FTP, email, or the online medium itself. Is *that* all you mean by
> "esoteric"? And if so, the online medium's there now: those who want to
> share drafts are free to share them that way. That isn't even
> "publication," it's just public sharing of work-in-progress.

SH3:
You're, and yet another gap in my original logic. Nothing is or was
stopping those who might wish to make their unrefereed drafts publicly
accessible online from doing so; but that is not the point, nor the
problem, nor the objective.

The problem is access to refereed, published research. All potential
users need access to that; and all authors want their refereed research
to be accessible to all its potential users (not just those whose
institutions can afford to subscribe to the journal in which it happens
to be published); whereas not all (or even most) authors want their
unrefereed drafts to be accessible to all.

(And, yes, "esoteric" is once again a red herring. It ought to have been
"peer-reviewed research" all along, to short-circuit potential
ambiguities and misunderstandings.)

> SH2: Why didn't you say that in the first place? "peer-reviewed" rather than "esoteric."

SH3:
Mea culpa.

> SH2:
> But, again, nothing stands in the way of authors sharing unrefereed
> drafts online with their tiny intended public prior to submitting them
> for peer-review and then publication, does it? What's your point?

SH3:
The point is and should have been about peer-reviewed drafts. Earlier
unrefereed drafts were just one potential entry point. (Perhaps I was
just too timid or unimaginative to say "post your peer-reviewed drafts"
at that time.)

> SH2: But is [what's sought] really the "patina" of paper publishing, or the patina of peer-review, and a given publication's prior track record for peer-review quality standards?

SH3:
Just peer-review and track-record. The rest was again just
ill-thought-through muddle.

(There may have been some faint excuse for such muddle way back in
1994; but one can hardly invoke that today, 15 years later, when all
of these muddles have since been raised, rehearsed, and resolved, many
times over, in countless online discussion forums, FAQs, conferences,
and published articles, chapters and books. Hence the frayed patience of
weary archivangelists even if they themselves are not free of original
sin, insofar as not having thought all things through sufficiently
rigorously at the very outset is concerned. There is no excuse for the
same old muddles 15 years on...)

> SH2:
> And what, exactly, is the scope of "peer-reviewed publication"? Apart
> from journal articles (and refereed conference proceedings), aren't
> monographs, edited books and even textbooks "peer-reviewed"? And aren't
> some of them "non-esoteric," because revenue-seeking?

SH3:
There is genuine uncertainty about the cut-off point. All peer-reviewed
journal articles are, without exception, author give-aways, hence all
can and should be made freely accessible online to maximize their usage
and impact.

The same may be true for *some* monographs, edited books (and possibly
even some textbooks, if the authors are magnanimous). But none of these
other categories is exception-free (rather the contrary). Freeing
authors' writings online against their wills cannot be the objective of
the Open Access (OA) movement. Nor can providing free access
to writings to which the author does not want there to be free access
serve as the basis for OA mandates by institutions and funders.

That is why the exception-free give-away content -- written solely for
usage and impact -- is the primary target of the OA movement (and of OA
mandates).

By the way, another enormous oversight in the Subversive Proposal
(though I can hardly imagine how it could have been anticipated at that
time) was the failure to call for (what we would now call) Green OA
self-archiving mandates by institutions and funders. It only became
apparent after another half-decade had passed with researchers' fingers
still not stirred into motion by the Subversive Proposal that mandates
would be necessary...

> SH2:
> The (obvious) flaw with the hope of making all refereed publications
> free online by first making their unrefereed drafts free online is that,
> unlike physicists (and, before them, computer scientists, and
> economists), most authors in most disciplines do not wish to make their
> unrefereed drafts public (either because they consider it unscholarly,
> or because they fear professional embarrassment, or because they don't
> want to immortalize their errors, or because they thing unrefereed
> results could be dangerous, e.g. to public health).
>
> Hence if the road to free online access is reserved for papers that
> their authors are willing to make publicly accessible as unrefereed
> drafts first, it will not cover much of refereed research in most
> disciplines.

SH3:
All true, and, again, mea culpa.

The road to the optimal solution -- the one that covers all refereed
research, immediately upon acceptance for publication, has been somewhat
circuitous:

First, the Subversive Proposal recommended self-archiving all unrefereed
preprints (but that would not work for the many researchers and
disciplines that do not wish to make unrefereed drafts public).

A variant on that strategy was the "preprints plus corrigenda" strategy,
which recommended self-archiving unrefereed preprints and later also
self-archiving a file containing all corrections arising from the
refereeing. Likewise inadequate, partly because, again, many authors
don't want to make unrefereed drafts public, and also because it would
be awkward and inconvenient for authors to have to archive -- and for
users to have to consult -- separate preprint and corrigenda files.

It has to be added that the P&C strategy was never really intended as
an actual overt practice: it was just intended to assuage the worries
of those who thought there was some sort of insurmountable obstacle in
principle to self-archiving the refereed version in cases where the
publisher objected.

In reality, some publishers have objected even to self-archiving
the unrefereed preprint [this is called the "Ingelfinger Rule"],
but most have since dropped this objection. And the sensible strategy
for the refereed postprint is to self-archive it and reconsider only
if and when a publisher requests a take-down. Sixty-three percent of
journals already endorse immediate OA self-archiving of the refereed
postprint. And in the past two decades, there have been virtually no
publisher take-down requests for the many million refereed postprints
that have been self-archived. It's absurd to let a one in a million
exception drive practice, especially when all it would entail would
be a take-down!

But for those authors (and for those mandates) that insist on refraining
from making the refereed postprint OA for the remaining 37% of articles
until their publishers endorse it (most endorse it after an embargo
period), the best author practice is to deposit the refereed final draft
in their own institutional repositories (IRs) anyway, immediately upon
acceptance for publication, but to set access to it as "Closed Access"
instead of Open Access during any embargo. That way the repository's
semi-automatic "email eprint request" Button can provide almost-immediate,
almost-OA to any would-be user during the embargo.
http://openaccess.eprints.org/index.php?/archives/274-guid.html

At the time of the Subversive Proposal, however, neither the OAI
interoperability protocol, nor OAI-compliant institutional repository
software, nor the notion of self-archiving mandates yet existed. So
today's Best Practice solution was not yet in sight, namely: deposit, and
mandate deposit, of all refereed final drafts immediately upon acceptance;
set access to the 63% of deposits that are published in Green journals
to OA immediately; and, if you wish, set access to Closed Access for
the remaining 37%, and rely on the Almost-OA Button during the embargo.
http://romeo.eprints.org/stats.php

Once such IDOA -- Immediate Deposit, Optional Access -- mandates are
adopted globally by institutions and funders, the days of embargoes are
numbered anyway, under the overwhelming pressure of the benefits of OA.
http://openaccess.eprints.org/index.php?/archives/494-guid.html

And another thing that was not yet in sight in 1994 was the fact that
the benefits of OA (likewise not yet named then!) could and would be
demonstrated to authors and their institutions and funders
quantitatively, in the form of the scientometric evidence of the "OA
Advantage": significantly increased download and citation impact for OA
articles, compared to non-OA ones. This too would eventually go on to
encourage mandates as well as the increased the use of OA content to
generate rich new metrics for measuring and rewarding research impact.
http://opcit.eprints.org/oacitation-biblio.html

None of this was quite obvious yet in 1994.

> SH2: And what about all the published reprints that authors would prefer
> not to have shared with the world when they were just unrefereed drafts?

SH3:
Self-archive the refereed version immediately upon publication (and rely
on the Button if you wish to observe the access-embargo).

> SH2: How and why did this "subversive proposal" (to the author community)
> turn into speculations about publishing and publishers?

SH3:
This is the plaint plagues and shames me the most! For the needless
and counterproductive speculation about the future of publication --
along with all the essential features of what would eventually be called
"Gold OA publishing" -- were all introduced in that proposal, with the
result that premature "gold fever" contributed to distracting from and
delaying the ("Green OA") self-archiving that was the essence of the
Subversive Proposal.

But I do think it was unavoidable -- in responding to the (now at least)
38 prima facie worries that immediately began to be raised time and time
again about self-archiving -- particularly worries #8, #9, #14, #17,
#19, #28, #30, & #31 http://www.eprints.org/openaccess/self-faq/ -- by
sketching the obvious way in which publication cost-recovery could
evolve into the Gold OA model if and when universal Green OA
self-archiving should ever make it necessary.

But I never imagined that the prospect of gold would become such an
attraction -- mostly to those, like librarians, not in a position to
provide Green OA themselves, but groaning under the burden of the
serials crisis, but also to publishing reform theorists more interested
in publishing economics and iniquities than in researchers' immediate
access needs -- that gold fever would propagate and distract from
providing and mandating Green OA, rather than reassuring and reinforcing
it. (For some reason that neither Peter Suber nor I can quite fathom,
people take to Gold much more readily than to Green, even to the extent
of imagining that OA is synonymous with Gold OA publishing.)

Well, one reaps what one sows, and I accept a large part of
the blame for having already began to sprinkle gold dust way
back in 1994, and continuing to stir it for some years to come --
http://www.nature.com/nature/debates/e-access/Articles/harnad.html#B1
--  until I at last learned from sorry experience to stop speculating
about tomorrow's hypothetical transitions and focus only on the tried,
tested and sure practical means of reaching 100% OA today: universal
Green OA deposit mandates by institutions and funders.

I still think, however, that the proof-of-principle for Gold OA
publishing by BMC and PLoS was, on balance, useful, even though
premature, because it did serve to allay worries that universal Green OA
self-archiving would destroy peer-reviewed publication altogether,
by making subscriptions unsustainable, and hence making publication
costs unrecoverable. No, it would merely induce a transition to Gold-OA
publishing to recover the costs of publication. (Moreover, the costs of
publishing then, after having achieved universal Green OA,  would be
far lower -- just the costs of peer review alone -- and paid for out
of a fraction of the self-same annual institutional windfall savings
on which the premise of subscription collapse underlying this set of
worries is predicated.)
http://www.publications.parliament.uk/pa/cm200304/cmselect/cmsctech/399/399we152.htm

But there I go, succumbing to gold fever again...

> SH2: In this speculation about publishing media and costs, what have
> "pages" to do with it? And what, exactly, does the 25% figure pay for (and what is the 75% that is no longer needed)?

SH3:
Pages have nothing to do with it. That was just a regrettable momentary
lapse into the papyrocentric thinking of the Gutenberg era.

The right reckoning is total publication costs per article. And once
authors are all systematically depositing their refereed drafts in their
institutional repositories, and users are using those OA drafts instead
of the publisher's proprietary version, the global network of IRs becomes
the access-provider and archive and the only remaining function (and
expense) remaining for journals is the implementation of peer review,
certified by their name and track-record. (The peers, of course,
continue to referee for free, as they always did.)

> SH2: You seem to be pretty generous with other people's money. ["advance
> subsidies (from authors' page charges, learned society dues,
> university publication budgets and/or governmental publication
> subsidies)" And you seem to have forgotten the money already being paid for subscriptions.

SH3:
More of the perils of premature speculation. Of course no extra funds are
needed if the transition to Gold OA only comes after universal Green OA
has been reached, and only if and when that universal Green OA in turn
makes subscriptions unsustainable. For then, by the very same token,
the subscription cancellation  releases the funds to pay for Gold OA --
whereas paying pre-emptively for Gold OA now, while it is unnecessary,
because most of the essential journals are still subscription-based,
requires extra money (and at an inflated -- because again premature -- cost).

But you see how easy it is to keep getting taken up with Gold OA
speculation instead of attending to Green OA practice, within reach
since 1994, yet still not grasped?

> SH2:
> But what, exactly, is this money supposed to be paying for? (Again,
> there seems to be conflation of online-only publication, and its costs,
> with free online access-provision: surely they are not the same thing.)

SH3:
Today: nothing. After universal Green OA -- if and when that makes
subscriptions collapse -- it will pay for peer-review alone.

> SH2:Â
> This still sounds quite muddled and vague:
> We've heard about "esoteric," give-away writings, but it has not yet
> been made clear what they are, and why they are give-ways.

SH3:
Refereed journal articles, written only for research impact.

> SH2:
> We have heard about online publication, and online-only publication.

SH3:
The Subversive Proposal was only meant to be about making refereed
research freely accessible online.

> SH2:
> We have heard about (some) authors making their unrefereed drafts free online. But how (and why) do we get from that to free online refereed publication?

SH3:
Forget about the unrefereed drafts; they're just extras. The way to make
refereed research free online is to deposit your refereed final draft,
free for all, in your Institution's OA Repository, immediately upon
acceptance for publication.

> SH2:
> And [how (and why) do we get] from there to paying to publish instead of paying to subscribe? (What needs to be paid for, how and why? And how do we get there from here, given that most authors do not wish to make their unrefereed drafts public?)

SH3:
Right now, nothing needs extra to be paid for. Subscriptions are paying
for it all, handsomely. All that's needed is author keystrokes, to deposit
all final refereed drafts, immediately upon acceptance for publication.

That's all that's been needed since 1994, but now we know the keystrokes
need to be mandated, to set the fingers in motion, so what's needed is
institutional and funder Green OA self-archiving mandates.

All of that is for sure, and will generate 100% OA with certainty.

The rest is speculation: If universal Green OA makes subscriptions no
longer sustainable, publishers will cut costs, downsize to the essentials
-- providing peer review alone -- paid for, on the Gold OA model, out
of the institutional subscription cancellation savings.

> SH2
> Sounds like a rather inchoate proposal to me... (And you reputedly
> expect this to happen overnight? Might we have some more details about
> what we might expect to happen on that fabled night?)

Inchoate it was, in 1994, though the practical means to do it overnight
(fingers) were already available in 1994. Since then, the OAI protocol
and the IR software have made it a lot simpler and easier. But the
keystrokes remain to be done. Thirty eight prima facie worries have kept
fingers in a state of Zeno's Paralysis, despite all being answered,
fully, many, many times over.
http://eprints.ecs.soton.ac.uk/12094/

Now it is time to mandate the keystrokes. That too could be done
overnight, by the stroke of a Department Head's, DVC's or VC's pen, as
Wendy Hall (Southampton), Tom Cochrane (QUT), and Bernard Rentier
(Liege) have since shown.
http://openaccess.eprints.org/index.php?/archives/136-guid.html

Will it be another 15 years before the remaining 10,000 universities and
research institutions (or at least the top 1000) wield the mighty pen to
unleash the even mightier keystrokes (as 68 Institutions and
Departments, and 42 Funders have already done)? Or will we keep
dithering about Gold OA, publishing reform, peer review reform, re-use
rights, author addenda, preservation and the other 38 factors causing
Zeno's Paralysis) for another decade and a half?

Stevan Harnad
Received on Wed Dec 02 2009 - 20:57:34 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:50:00 GMT