Free Access vs. Open Access

From: Stevan Harnad <>
Date: Mon, 11 Aug 2003 03:39:33 +0100

BioMedCentral's "Open Access Now" is a useful newsletter, but its first
editorial contains some inadvertently misleading information that needs
to be corrected. What
actually said was this:

> "Free Access is not Open Access"
> "There seems to be a general misunderstanding that the aim of the
> Open Access movement is to make the scientific research literature
> free online. But there is a difference between "free access"
> and "open access"...
> "The benefits and promise of Open Access will only be realized
> when this distinction is clear in the minds of authors and
> publishers. Only then can the literature move from being `free'
> to being truly `open'."

I will quote/comment the full (short) editorial in a moment to show
why I think what it *should* instead have said is this:

    "Open Access Calls for Both Free Access and Open Usage"

    "There seems to be a general misunderstanding that the aim of the Open
    Access movement is *only* to make the scientific research literature
    free online... That is the first aim, but it also aims to make it
    fully usable."

The difference between the two messages is substantial. We are very far
from having free access to the refereed research literature, even though
it is within reach; vast amounts of potential research impact are for
this reason being needlessly lost; and it is free access that is urgently
needed to put an end to this loss. What free access we do have today,
however, is not constrained by any usage constraints. Hence the
difference between "free access" and "open access" is merely
hypothetical right now: What is needed is more free access, not an
extension of free access to open access. To imply otherwise is simply to
saddle the research community with yet another red herring, instead of
what it really needs.

Here is the current situation, in rough practical and statistical

    (a) What the BOAI seeks is unrestricted toll-free
    full-text online access to the entire refereed research corpus
    (20,000 journals, 2,000,000 articles per year).

    (b) The way to achieve this is for researchers to (1) publish their
    papers in open-access journals whenever suitable ones exist
    (under 5% currently) and, for the rest of their papers (95%), to
    (2) self-archive them in their own institutional archives. [(1)
    is BOAI Strategy 2, and (2) is BOAI Strategy 1.]

    (c) Any form of restricted, gerrymandered online access (such as
    "ebrary"-based access that prevents down-loading, saving or
    printing-off) would not be open access (but there is none in sight
    so far to speak of).

That is all there is to it! Now, for those who are interested, a more
detailed quote/comment of the full (short) BMC editorial:

> Free Access is not Open Access

Not necessarily, in theory; but in reality and in practise, *all* of the
growing body of research today that is free-access is also open-access:

It can all be downloaded, saved, grepped, printed out, quote/commented,
and the URL can be sent to anyone who wishes to do likewise. All data
therein can also be used, *exactly* as they could be if read and copied
from the on-paper version. (It is simply an error, in other words, to
think of refereed, published articles as analogous to the genome database
or to software. It consists instead of texts, which are written to
be printed off, read, used, applied, built-upon, quoted/commented,
and cited. There is no question -- or need -- of republishing them or
altering them. They are already freely accessible to anyone with access
to the Web, and the only ones to update them are the authors; everyone
else must settle for quote/commenting, applying and citing.)

> There seems to be a general misunderstanding that the aim of the Open
> Access movement is to make the scientific research literature free
> online. But there is a difference between "free access" and "open access".

The aim of the Open Access movement *is* to make the scientific
(and scholarly) refereed-journal research literature -- full-text --
accessible toll-free online. Though there may be hypothetical ways
toll-free online access could be constrained so as to prevent
downloading, grepping, or printing, no such thing is happening. All the
free-access literature is also open-access.

> This distinction was part of what motivated the Bethesda definition of
> Open Access Principles that we published in the first issue of Open Access
> Now (July 14, 2003). That definition clearly states that access to the
> information should be free, but in addition the work should be open to
> re-use and redistribution

"Re-use and redistribution" has to be thought out more fully and clearly
than it is in the Bethesda definition -- insofar as refereed journal
articles are concerned. We are not talking about shared empirical
databases here but about the articles that appear in the 20,000 existing
(toll-access) peer-reviewed journals. The use one makes of those full
texts is to read them, print them off, quote/comment them, cite them,
and use their *contents* in further research, building on them. What is
"re-use"? And what is "redistribution" (when everyone on the planet with
access to the web has access to the full-text of every such article)?

Fifty-five percent of toll-access journals already formally support
author/institution self-archiving (BOAI-1) in their copyright/license
agreements, and many of the rest will agree if asked. No Bethesda-style
definition of "open access" needed to be adopted: The author merely
needed to assert, and the publisher to support, self-archiving.

> and that it should be deposited immediately
> upon publication in a public online repository (such as PubMed Central).

For the 95% solution, BOAI-1, depositing those toll-access articles in
the author's own institutional repository is the *means* by which they are
made free-access, by definition. For the remaining 5% (BOAI-2),
the fact that they are published by an open-access journal *entails*
(again, by definition) that they must be made freely accessible online
*somehow*. Likewise depositing them in a public online repository
(whether in a central one, like PubMed Central, or -- why not? -- in
the author's own institutional repository, this time too) seems like a
congenial solution to providing this essential feature of what it is
that makes an open-access journal open-access!

> Publishers who offer free online access on their own websites still have
> a long way to go before their research articles can be considered Open
> Access.

I know of no publisher-provided toll-free online full-text access with
"ebrary"-style constraints on downloading, grepping, printing, etc. But
if there *are* any such cases (and they can successfully prevent downloading,
grepping, printing, etc.) then that sort of gerrymandered access should
not count as open access, and that publisher certainly doesn't count as
an open-access publisher.

But what is the point? BOAI-1 is institutional self-archiving, not
publisher self-archiving, and it involves no ebrary-style
gerrymandering; and BOAI-2 *does* guarantee unconstrained
access. The fact that toll-access publishers do *not* provide toll-free
access is the whole point of the BOAI movement! If they did, we could
all go home now (and access it all)!

> The benefits and promise of Open Access will only be realized when
> this distinction is clear in the minds of authors and publishers.

I think authors know perfectly well when they can and cannot access the
full text of an article (including download, storage, grepping
and printout) toll-free. Toll-access publishers know the difference
too. The difference between unconstrained free access and gerrymandered
ebrary-style access will also be fully felt and appreciated -- if and
when it ever comes to pass. So far, it's nowhere in sight! Hence, at
the moment, *all* the benefits of Open Access reside in free, full-text,
online access of the sort that a growing number of articles already have
(most of them through BOAI-1) but that most of the 2,000,000 articles
published annually still lack. It will not help them get it if we seek the
benefits and promise from promoting the free/open distinction, rather
than from promoting free access!

> Only then can the literature move from being `free' to being truly `open'.

The "move" we should all be dedicating 100% of our energy and attention
to is the move from toll-access to free-access. That's the move that
awaits us impatiently, to at last stem our daily needless
impact-loss. There is no free-access literature straining to move from
free-access to open-access anywhere in sight at the moment.

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online is available at
the American Scientist September Forum (98 & 99 & 00 & 01 & 02 & 03):

Discussion can be posted to:
Received on Mon Aug 11 2003 - 03:39:33 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:02 GMT