Re: Should Publishers Offer Free-Access Services?

From: Stevan Harnad <>
Date: Mon, 21 Jun 1999 19:55:52 +0100

On Fri, 18 Jun 1999, Thomas J. Walker wrote:

>sh> (1) Authors can provide free access to their work RIGHT NOW by
>sh> self-archiving it all on their institutional servers and/or LANL and/or
>sh> CogPrints and/or (eventually) E-biomed/Scholar's-Forum.
tw> True--if they want to go to the trouble of putting their work in a place
tw> where their colleagues aren't conditioned to look

100,000 Los Alamos authors have not found it any trouble to do so, and
about 50,000 users a day have managed to get "conditioned." The word
does spread, you know.

tw> and if they don't mind violating the copyright agreement they have signed.

(1) Many copyright agreements are not explicit or unequivocal about
online self-archiving rights

(2) Where they are, authors should not sign

(3) Where they have signed, they should use an alternative draft,
containing the substantive refereed final draft in their own version.

tw> But how many authors are posting to CogPrints?

Incomparably fewer. History will note that Physicists were faster
off the mark, and that the rest of the disciplines were led to the
water but took longer to drink. Have patience! There are a few more
inducements on the way, such as citation linking...

tw> The point of PDF reprints
tw> is to quicken the pace toward free Web access and to help
tw> authors/institutions and publishers make the transition from users-pay to
tw> authors/institutions pay.

The PDF proprietary version is not worth the price of the paper it isn't
printed on... Let author's self-archive their own versions for free.

tw> Incidentally, as I understand it, LANL authors generally provide the text
tw> of the refereed version but not the formatted, official version.

Correct. There is no need for it.

tw> As you
tw> and Paul G. say in your citation-linking proposal, "Authors might wish to
tw> have arrangements for official links with the published version in order to
tw> provide an authenticated draft, or one in which the paper page images can
tw> be viewed or cited by page and line."

Yes, APS will have an official overlay, and can provide authentication
if/when needed, and can charge for it if they like.

tw> Publishers should be paid for the services they provide, even if it is for
tw> posting the formatted, official version on the Web.

They should be paid for their useful services: quality control and
certification. (Peer review and acceptance in the journal.) Useless
services, like the PDF version and page images should fend for

>sh> (3) Hence there is no earthly reason why they should want to pay anyone
>sh> to do that for them.
tw> Here are three reasons: convenience, ethics, and better exposure.

Self-archiving wins hands-down on 1 and 3: What's the ethical issue?

>sh> (4) There is equally no reason why they should want to pay anyone for
>sh> reprint rights.
tw> Authors should pay for services they receive.

Don't purchase reprints and you don't have anything to pay for...

And of course reserve the self-archiving rights in your copyright
agreement, as you hand over the text to the publisher (for free)...

tw> PDF reprints have no more to do with "reprint rights" than do paper
tw> reprints. Authors who choose not to buy paper reprints can and do
tw> distribute photocopies.

Correct. And on the Web they can and should self-archive. End of story.

tw> Infinite PDF reprints are a better buy for an
tw> author than are 100 paper reprints when they are priced the same.

Of course; but they are not a better buy then infinite self-archived
non-PDF eprints.

tw> Selling
tw> PDF reprints at that price is also more profitable for the publisher,
tw> because PDF reprints are cheaper to produce. Publishers don't offer them
tw> because they hope to delay or prevent free Web access to the journal
tw> literature.

Fine. If publishers allow authors self-archiving rights for the final
refereed draft, and the only version at issue is the PDF one, let the
market decide what the author prefers to do. (I have no doubts about the

>sh> (5) There IS an UNearthly (needless, unconstructive, unjustifiable,
>sh> counterproductive) reason authors might feel they HAVE to pay for the
>sh> right to self-archive their own papers, given to their journal
>sh> publisher for free, and that is if they were foolish enough to sign a
>sh> copyright agreement that DENIED them that right (except if they buy it
>sh> back again).
>sh> But the solution to (5) is for authors to refuse to sign any copyright
>sh> agreement denying self-archiving rights, and to self-archive any paper
>sh> for which they have not explicitly signed away their right to do so.
tw> That would be the solution _if_ you could get authors to spend time
tw> confronting a system that, in their view, isn't broken. As they taste the
tw> advantages of free Web access to their articles through self-archiving and
tw> PDF reprints, they will realize the system, broken or not, needs changing!

Why on earth, if faced with a system that is "not broken," will they
arrive at the realization that it needs changing more readily by
PAYING to self-archive the PDF version rather than self-archiving the
non-PDF version for free? (Please don't reply that it's because it's
easier for the journal to do it for you: First try the form interface
of Los Alamos or CogPrints, from a word-processor version of your
text,and THEN tell me it's worth the price of the PDFprints to escape
having to do all that!)

>sh> And if/when they HAVE inadevrtently signed it away, they should DEMAND it
>sh> back, not BUY it back; the latter would be adding insult to injury, and a
>sh> horrible precedent for the future. There is no reason (scientific,
>sh> moral, logical or practical) why what is permitted to (e.g.) APS
>sh> authors should not be permitted to all authors.
tw> They don't inadvertently sign their rights away. They do it because they
tw> don't yet see the bad consequences, and they want to get on with their
tw> research rather than risk a hassle.

Then it is time for us to open their eyes, as the eyes of the Physics
community have already been opened.

>sh> And if they cannot get back the right to self-archive the final draft,
>sh> they should simply self-archive a penultimate draft, rather than the
>sh> final one, incorporating into it whatever there is of scientific
>sh> substance that needs be incorporated to make it as archivally useful as
>sh> the copyrighted final draft. Self-archiving of the prior unrefereed
>sh> preprints is also a good way of asserting this incontestable right.
>sh> There is a slippery slope there in favour of the author, in those
>sh> anomalous cases where author and publisher are in conflict rather than
>sh> in harmony -- as they ought to be, about copyright "protection."
>sh> (Copyright law was not drafted to protect authors from themselves!)
>sh> For the record: My advocacy of author-instution-end publication charges
>sh> is most definitely NOT so that authors can buy back their
>sh> self-archiving rights, having given away both their papers and their
>sh> rights for free. It is only to pay for peer review and certification --
>sh> and that only after S/L/P toll-gate receipts no longer cover that cost,
>sh> and all other costs and barriers have been eliminated. And the source
>sh> of those funds will be the S/L/P savings themselves; before that, yet
>sh> another unnecessary and counterproductive access toll, this time on the
>sh> author rather than the reader, would be outrageous.
tw> You and I agree that access should be free and that authors/institutions
tw> should pay. I see PDF reprints as one means of speeding the transition
tw> from the present system.

As long as there is no attempt to prevent free self-archiving via
restrictive copyright agreements, I have no problem whatsoever with
letting the market decide whether authors prefer to self-archive
or to pay to have it done for them...

>sh> That is true, but the full benefits of
>sh> free access are already available from
>sh> free self-archiving. Hence nothing more is to be learnt from adding
>sh> needless expenditures to them! We must UNlearn the habit of supporting
>sh> needless, access-denying cost-barriers.
tw> PDF reprints are not an access-denying cost barrier. They make an author's
tw> work totally and permanently free in the official form for the cost of 100
tw> paper reprints--something that more than 90% of authors pay (in the
tw> journals for which I have data).

But they are being offered in a context where the alternative to them
APPEARS to be access-denying S/L/P
(Subscription/Site-License/Pay-Per-View). I agree completely that if
the third alternative of self-archiving is also clearly available, we
are merely speculating here about what authors' preferences will be.

tw> Free Web access to journal articles is not the status quo. PDF reprints
tw> achieve it for a fee that most authors already pay.

But needn't.
Received on Wed Feb 10 1999 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:33 GMT