Re: Free Access vs. Open Access
On Mon, 27 Oct 2003, Jan Velterop wrote:
> If online material is 'open' in the sense of 'free' that is of course a
> great step forward, but if it's only available in pdf...
> that is decidedly sub-optimal...
> Not being optimal... shouldn't be an excuse for not making freely available
> ...by (self)archiving ...but
> ...we shouldn't lose sight of the ultimate... goal, 'open' access
> (as defined in the Berlin Declaration, the Bethesda principles,
> by Wellcome, PLoS, BioMed Central, and others) as opposed to merely
> 'free' access.
> It doesn't help to be sub-ambitious
I can only repeat that the open/free distinction is a red herring, no
matter how often it is invoked formally and informally.
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2956.html
These matters have not been thought through rigorously; and things
decided in haste have been invoked ever more solemnly without having
been examined for their usefulness or even their coherence.
Please, let's not lose sight of the problem, which is still there,
as pressing as ever, but now being kept at a distance by yet
*another* groundless, confusion-generating, and -- most important --
*inaction-encouraging* reservation.
The problem -- it can never be repeated often enough, apparently -- was
and is this: There are 24,000 journals publishing 2.5 million articles
per year, most of them not accessible to most of their potential users
worldwide because of access-tolls. This was also the problem in the
paper era, but in that medium there was no solution because of the
true costs and limited power of paper. (One could not diffuse paper
over the airways, let alone data-mine it!)
Now we are in the online era, which offers many new possibilities,
including online data-mining. But the *relevant* possibility -- relative
to what we do have, and what we still lack, *exactly* as we lacked it in
the paper era -- is the possibility of toll-free access to the full-text
online. That is what was missing then, and that very same access is
missing now.
So what do we do? We start to talk about this *absent* access as *not
enough*, "sub-optimal," not the "ultimate goal"!
This is rather like declaring (while still sitting in the total darkness):
"Let there be light -- but let it not be just be the good old
sunlight we've been deprived of for centuries, but voice-activated,
computer-controlled, fluorescent/incandescent light, 100K Lux!"
So if someone proposes: "Why don't you just open the curtains and let in
the sunlight?" the reply is "It doesn't help to be sub-ambitious"!
Considering the actual circumstances -- the curtains being still
closed, and most of us still sitting in the dark -- it does seem
rather impractical to be referring to the call to open the curtains
as sub-optimal and sub-ambitious while that simple act is still not
being performed, even though it could be, immediately, because people
still don't understand that it can be, nor what advantages it will
bring. Instead, we get ahead of ourselves, and fixate on the advantages
it will *not* bring!
Yet even those alleged shortfalls are spurious! (See the prior postings
on this thread.) Free online access to the full text (even if only
PDF!) still means being able to do *everything* one could do with paper
(if one could afford the access-tolls!), including reading the printed-off
version! But there is also on-screen browsing, reading and navigation,
downloading, storing, forwarding, *and* the capacity to convert
automatically to html or ascii for text-mining. Free-access online texts
are also harvestable and harvested, invertible and indexable, hence
boolean-searchable and otherwise navigable, singly and collectively. Not
to mention collectible into global virtual archives like oaister, the
google of the refereed research literature.
Data-mining? First, let us not forget that the text of a journal article
usually does not contain the empirical data on which it is based (in
part because it would have been too expensive to publish all those data
on paper in the paper era!), only the summary tables and analyses. So
the empirical data were and still are a separate database -- one that
should likewise be made freely accessible, alongside the refereed article
literature, certainly, but that is a separate matter, not to be conflated
with the open-access movement's first, second and third goal, which is
to free access to the refereed article literature!
http://www.ecs.soton.ac.uk/~harnad/Temp/data-archiving.htm
What quantitative data do appear in the text of an article are just text,
like the rest of the article. Even in XML format, the problems of how
to make generic data interoperable remain to be worked out, so let us
not delay opening the curtains on that account either!
> 'free' will come in the wake of the open access movement, but
> I doubt if the reverse is true.
It's not the "wake" that's the problem, but the *wait* (and, yes, it is
indeed beginning to feel ever more funereal!).
Since "free access" is one of the necessary conditions for fulfilling
the definition of "open access", it is tautological that "free" will
follow (!) "open": It "follows" it logically, as surely as "p" follows
"not-not-p"! But the crucial question is *when* will we have "open"?
For we can already have "free" now, if we just open the curtains!
So in real-time, we can already have "free" now (by each of us immediately
self-archiving, one by one, of all 2.5 million of our annual research
articles), whereas to have "open" others first have to create or convert
23,500 more open-access journals, one by one.
I'd rather not wait for that, to get access to the sunlight.
The reverse is true? Getting free access now will reduce the probability
of later getting any online text-mining powers we may be missing? I would
like to see the logic of that sketched out explicitly, for I don't see it
at all! Surely a free, online, full-text corpus will inspire all manner
of further online optimizations sooner than sitting and waiting for them
in the dark!
I know Jan is not actually recommending that we wait in the dark! But
the fact is that we *are* waiting in the dark, and what we really need
is elucidation of the feasibility and benefits of the opening of the
curtains that is within our immediate reach. It does not help to
encourage our Zeno's Paralysis by suggesting that letting in the light
would merely be "sub-optimal"!
http://www.eprints.org/self-faq/#25.Mark-Up
http://www.eprints.org/self-faq/#15.Readability
http://www.eprints.org/self-faq/#16.Graphics
> ...to analyse, data-mine, and text-mine... it has to be in a
> machine-readable format.
Please see the above on the distinction between data-mining and
text-mining, the difference between article texts and their empirical
data, and on the convertibility of PDF to html and ascii. (PDF is a red
herring anyway, for authors with data tables can also self-archive the RTF
and HTML, and eventually no doubt also a vanilla XML generated from
their own word-processor version.)
> Open Access really is more than just an economical goal (although it goes
> without saying that being able to access literature without having to be at
> an institution that can afford the access tolls helps enormously).
Please translate this into the terms of the "let there be light" analogy
introduced here an intuition pump for the toll-free access that was, is,
and will remain the essence of the open-access movement.
"Toll-free access to the refereed literature is more than just an
economic goal"
Of course! And also *other* than an economic goal. It is the toll-curtains
that keep us in the dark, but the light can be let in by opening the
curtains on our own work without changing the overall economics: what we
need and want is light, not economic-change. (It is open-access journals
that offer the light through economic change.)
"(although it goes without saying that being able to access
the refereed literature toll-free helps enormously)"
Is this not another tautology (once it has been freed of the
spurious open/free distinction that I have to tried to show to
be functionally empty above)?
> Perhaps the difference in approach between open access publishing and
> self-archiving, while both working in parallel to strengthen one another, is
> the sense of priority of a qualitative (in terms of usability) versus a
> quantitative one.
Not at all. Both approaches yield *exactly* the same thing, both
quantitatively and qualitatively: toll-free access to the digital
full-text of all refereed journal articles online. The difference is
that one approach requires the founding, funding and filling of 24,000
new open-access journals with the digital texts of the yearly 2,500,000
articles, whereas the other requires only the the founding, funding and
filling of institutional eprint archives with the digital texts of the
yearly 2,500,000 articles. You will find that the number, cost and
difficulty of the respective founding/funding/filling steps is far lower
for the one than the other, both quantitatively and qualitatively:
http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0025.gif
Stevan Harnad
NOTE: 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):
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/index.html
Posted discussion to: american-scientist-open-access-forum_at_amsci.org
Dual Open-Access Strategy:
BOAI-2: Publish your article in a suitable open-access journal
whenever one exists.
BOAI-1: Otherwise, publish your article in a suitable toll-access
journal and also self-archive it.
http://www.soros.org/openaccess/read.shtml
> > -----Original Message-----
> > From: Stevan Harnad [mailto:harnad_at_ecs.soton.ac.uk]
> > Sent: 24 October 2003 18:20
> > Subject: [Manifesto] Re: Open Access and Humanities Monographs
> >
> > On Fri, 24 Oct 2003, Stefan Gradmann wrote:
> >
> > > [Willard,] as you state, the online version of a
> > > book is not satisfying (and this already has caused the
> > death of the rather
> > > silly e-book paradigm), and thus self-archiving of book
> > material (even if it
> > > was available for the authors) may not be a solution at
> > all. Open access to
> > > electronic information only gets attractive in our context once this
> > > material is published in a way that is appropriate to the electronic
> > > environment and that makes use of its innovative potential in a way
> > > PDF-documents modeled on the printing analogy simply don't!
> >
> > I *completely* disagree! Consider the following (I think much more
> > realistic) logic:
> >
> > (1) It is a *good* thing that online access to full-text monographs is
> > not as attractive as having the book on paper. That removes one
> > prima-facie obstacle to self-archiving them and thereby providing open
> > access for those who cannot afford to buy the monograph yet
> > might still
> > make some use of the text!
> >
> > (2) Once open access -- reminder: that means toll-free
> > full-text online
> > access for anyone on the web -- becomes widespread for
> > monographs, there
> > will be much more motivation for designing ways to make online access
> > more convenient, useful, effective.
> >
> > It makes no sense whatsoever *not* to self-archive a monograph merely
> > because online access may not be optimal! It's certainly 100% better
> > than no access! (This reasoning is simply the flip-side of the equally
> > self-paralytic reasoning that they should not be self-archived because
> > they *would* be preferred over the paper version! At least the latter
> > would have a publisher, and possibly a royalty-seeking author
> > to endorse
> > the reasoning; but with the online-is-nonoptimal argument it is purely
> > a rationalization for inaction! No losers; no winners.)
> >
> > >wm> "Open" is a word like "free", whose meaning and import
> > >wm> greatly depends on the preposition that implicitly follows.
> > >
> > > You are perfectly right in pointing out some facets of the
> > connotation aura
> > > of a term like 'open' (and much more could be said here) -
> > I would only like
> > > to add that the same kind of reflexion could be made
> > regarding the term
> > > 'access' which may have very different connotative values
> > depending on
> > > whether you use it with a 'text culture' or with an
> > 'empiristic' background
> > > ...
> >
> > It is here that I feel that we non-hermeneuticists and
> > non-semioticians
> > may have a bit of an advantage, in not getting too wrapped up in
> > far-fetched connotations. Here is a black and white distinction:
> >
> > (1) 2,500,000 articles in 24,000 journals can only be read
> > online if the
> > user's institution can afford to pay the access tolls.
> >
> > (2) Open access means being able to do the same thing as those lucky
> > users, but without having to be at an institution that can afford the
> > access tolls.
> >
> > Open access is not about access to the printed edition. (But
> > the online
> > edition can always be printed off, if one wishes.)
> >
> > No philosophical problem. It is clear what we do not have
> > now, and what
> > we would have if there were open access to the journal article
> > literature. Ditto for the monograph literature. (And note that nothing
> > was said about the superiority or even parity of online
> > access compared
> > to on-paper access for monographs. It's only about about tolled
> > vs. toll-free online access.)
> >
> > Cheers, Stevan
> >
Received on Mon Oct 27 2003 - 14:09:34 GMT
