[GOAL] Re: Definition of OA and its Priorities and Obstacles

[Stevan Harnad]

On 2012-08-29, at 3:35 AM, Peter Murray-Rust wrote:

I have been asking several times for definitions of "Open Access". I get no
> answers but am flooded with political slogans…


Gratis OA: Free online access

Libre OA: Free online access + various re-use rights (there is no agreement
on which ones, but maybe up to and including CC-BY)


 It is one person's (PS) analysis of the situation, not an agreed communal
> view. It may be that many of the community agree it, but it is still one
> person's view.



Peter Suber has been the principal spokesman, and was the principal drafter
of the original Budapest Open Access Initiative (BOAI), and is now the
principal drafter of the 10-year revision.

Many of the community agree. Peter  Suber is extremely dedicated to
reaching a consensus. But a consensus in human affairs almost never means
complete unanimity.

It would be helpful, though, if dissenting voices were to address matters
of substance, rather than matters of definition.

I think what you (Peter Murray-Rust) are saying is that for the needs of
your field, Gratis OA is not enough: You need Libre OA, with particular
re-use rights (let's say CC-BY).

That's fine. It is not the definition of Gratis and Libre OA that is at
fault for the fact that we don't yet have Libre OA for your field: what is
at fault is the fact that we don't yet have Libre OA for your field.

It is not a "political slogan" to say that Gratis OA is much easier to
reach than Libre OA, and that Green OA mandates are a way to reach it. It
is simply a practical reality.


It is not "an update of the BBB definition" - this cannot be unilaterally
> decided by one-and-a-half signatories on a rolling basis. If there were a
> community process rather than individual pronouncements I would probably
> feel more comfortable.
>


To repeat, tilting at a definition (BBB is roughly equivalent to Libre OA)
is not going to bring us any more OA. More likely, it will give opponents
of OA -- as well as those for whom free online access is not enough -- the
chance to say that "'OA' mandates do not generate 'OA'".

A rather hollow gain from tilting at a definition instead of focusing on
viable, practical ways to generate either (1) free online access or (2)
free online access + various re-use rights.


 "Green" and "Gold" are not definitions of the state of Open Access, they
> are - at best - definitions of a process.
>


Correct (and I don't think anyone has said otherwise).


Where is "free online access" defined? It is not a simple concept. (a) is
> it permanent or temporary?



Permanent.

"free, immediate, permanent online access" to the full text of refereed
research articles for anyone, webwide: http://bit.ly/2380-refs


(b) how is it recognized?



I don't understand the question.

If you can access it for free on the web, it's freely accessible to you
now.

("Permanent" is a tall order: Please consult David Hume on "problem of
induction <http://plato.stanford.edu/entries/induction-problem/>")


(c) is it a property of (i) a document



Yes


or (ii) a location



Yes


or (iii) a process?



Well, you have to get it there somehow


Or some combination?



See above


(i) What are the "some" re-use rights?



Very far from being universally agreed.

For some (like yourself, I think) these include data-minability by machine,
and the right to re-publish ("derivative works").

For others it is CC-BY.

Libre OA potentially covers the entire CC spectrum.


(ii) Where are they listed and defined?



Please see the Creative Commons site: http://creativecommons.org

The OA movement is not the same thing as the Open License movement, though
Open License is one way to characterize Libre OA.

It is not that Libre OA lacks a definition, but that there are a huge
panoply of potential re-uses, hence of re-use rights.

The important practical thing is, again, not a matter of definition. It
depends on what users need and what creators want to provide -- and, most
important of all, how to ensure that it is indeed provided.


(iii) How are they recognized?



Not my speciality, but I believe the CC people are working to make clear
licenses machine- as well as human-readable.


[If] "permission to deposit in a University repository" could be claimed as
> a re-use right, in which case all Green University OA was by definition
> libre.



Some think there needs to be a license to access any item on the web. If
so, trillions of Web items that are freely accessed daily are in need of
licenses.

Yes, logically speaking, "free online access" could be defined as a "re-use
right". Do you think that would clarify matters? Do you think it would help
generate (1) more free online access, or (2) more free online access +
various other re-use rights?

("Re-use" rights could also be re-defined as various free online use
rights: Do you think that would help anything?)


3. Gratis OA is a necessary condition for Libre OA.
>
>

I would agree, although without operational definitions I cannot be sure



I'm grateful for that.

(On the limitations of "operational definitions," I again suggest
consulting Hume on the problem of induction.)


.6. Global Gratis Green OA is within reach of Green OA mandates (ID/OA +
>> "Almost-OA" Button)
>
> We  now have another concept "Almost OA" which again is not defined. And I
> assume that the "OA" in ID/OA is not "Open Access" but "Optional
> something-or-other".
>
>

"Almost OA" refers to individual access to papers that have been deposited
in an institutional repository and that are under publisher embargo. During
the embargo, only the metadata are visible to users and harvesters. But the
repository has an automated Button: Any individual user can paste his email
and click the Button which sends an automatic email-eprint-request to the
author, who can, with one click, authorize the automated emailing of one
copy.

This is not OA, just Almost OA. But again it is a practical compromise, in
order to make it possible for all institutions and funders to mandate
immediate-deposit, whether or not there is a publisher embargo on OA.

The rationale (not ideological, but practical) is that many institutions
and funders think they cannot adopt a Green OA mandate because publishers
have varying embargo periods. So they adopt policies to the effect that
"authors must deposit only if and when their publishers say they may do
so."

The ID/OA mandate (Immediate Deposit, Optional Access) is the way to
upgrade such policies (which are not mandates at all, but merely echo
publisher policies) by mandating immediate deposit in every case, even if
they do not mandate making the deposit OA immediately.

What has to be kept in mind is that the only thing standing between the
status quo and universal Green Gratis OA (free online access) is author
keystrokes.

Sixty percent of journals (including almost all the top journals in almost
all fields -- not, notably, chemistry!) already recognize their author's
right to make their deposits OA immediately. Forty percent don't.

This means that once ID/OA is adopted universally today  -- as it can be,
because it moots all publisher embargoes -- the result is 60% immediate
(Gratis) OA plus 40% Almost-OA.

The likely outcome (based again on Humean induction, and perhaps on Human
Nature in general) is that with all articles being at most only one
keystroke from being OA, 60% of them being OA already, and 40% of them
being made accessible individually at a cost of two keystrokes, the
palpable benefits of OA to both users and authors will soon make all
embargoes die their inevitable and well-deserved deaths!

The big keystroke barrier is the deposit. Deposit mandates remove that
barrier. The last keystroke will take care of itself.

To repeat: None of this is a definitional or ideological  matter. It is a
practical matter: How to maximize access to refereed research, as soon as
possible, and as much as possible.

7. 8, 9, 10, 11, 12, 13, 14. Irrelevant (a mixture of political assertions
> and opinions-stated-as-facts).



No. They are practical strategic contingencies, designed to generate OA as
soon as possible. Some of them may be opinions, but most are opinions that
have been tested and proved to work.

My suggestion to you Peter, is that you forget about definitions and think
about practical ways to accelerate the transition to the particular form of
Libre OA that you seek for your field.

All I ask is that you respect the fact that the primary and most urgent
need of all fields is Gratis OA, that there is already a practical way to
generate that (Green OA mandates) and that the specific needs of your own
field should not be taken as grounds for disparaging the practical
fulfillment of the immediate needs of other fields.

I get shouted down because asking factual questions is uncomfortable.

Factual questions are more than welcome (once!).

*What is important, though, is not to keep asking the same questions over
and over, but to listen to the answers, and then add some constructive and
realistic practical suggestions of your own. *
*
*
Many of us would be much more willing to answer your questions, Peter, if
you showed some sign of having taken the replies on board, rather than just
continuing to repeat the questions.

 the OA movement is not the Open License movement
>
>
> I have never argued that it was. I am arguing that unless Open Access is
> clearly defined then huge amounts of energy and money are wasted and great
> confusion ensues. The confusion is not of my making - I am trying to
> resolve it by asking questions.
>
>
The problem to solve is how to maximize access to refereed research, as
soon as possible.

The problem is not with the definition of Gratis OA or Libre OA.

This is science-wide at least. I am also arguing on behalf of people
> outside academia. I am arguing that many beneficial developments are
> stalled unless we have BOAI-compliant OA. For example the next generation
> of search engines depend on unrestricted access to full-text. (I will not
> use the term LibreOA until I get a clear answer as to what it is in
> practice.
>

1. OA is wider than science.

2. Many beneficial developments are stalled without Gratis OA, in all
fields.

3. In some fields, further beneficial developments are stalled without
Libre OA (with whatever specific license you have in mind).

Now can we focus on realistic, practical remedies for 2? And remedies for 3
that are not at odds with remedies for 2?


Here are some more questions. They should have a simple YES/NO/factual
> answer:
> - many repositories consist largely of metadata-only. Is metadata-only
> counted as Open Access?



No

OA (Gratis and Green) pertains to the full text of peer-reviewed research
papers.

- some repositories have "ID/OA". (a) what does the O stand for?


Optional: Immediate-Deposit/Optional-Access

(b) where is ID/OA defined?


http://openaccess.eprints.org/index.php?/archives/71-guid.html
http://openaccess.eprints.org/index.php?/archives/494-guid.html

Peter Suber's term for this is DDR: Dual Deposit Release (DDR):
http://www.earlham.edu/~peters/fos/newsletter/08-02-06.htm

(c) is ID/OA counted as Open Access?


ID/OA is a mandate. OA is a property of a paper.

Green OA mandates all have certain compromises because of publisher
policies.

For this reason there are (2) no Immediate-OA + No-Embargo mandates and (1)
*no Libre Green mandates* (except with opt-outs, which means no mandate,
and hence little "compliance").

The viable Green OA mandates are two:

1. ID/OA

2. ID/OA plus Copyright Reservation, with opt-out permitted for copyright
reservation but not for deposit.

In contrast, "Deposit if and when your publisher allows" is not a mandate
at all.

Finch/RCUK have lately proposed what looks like a Libre Gold OA mandate,
but on closer inspection, it has profound flaws, apparently unanticipated
by the drafters of the policy -- flaws that make it extremely unlikely that
it is viable unless it they are promptly corrected:

Finch/RCUK mandate that the RCUK author is no longer allowed (sic) to
publish in a journal that offers neither (i) Libre (CC-BY) hybrid Gold OA
nor (ii) Green OA with a maximum embargo of 6 months (12 for EHRC and ESRC).

Finch/RCUK mandate that if the author chooses to publish in a subscription
journal that offers both Libre (CC-BY) hybrid Gold OA and 6/12 Green OA,
the author must pay for the Libre Gold rather than just provide Green.
(Finch/RCUK offer "block grants" re-directed from research funds to pay for
Libre Gold OA.)

RCUK's two profound, unanticipated flaws are that

(1) RCUK's promise of extra payment for Libre (CC-BY) Gold OA, over and
above world-wide subscription revenues, is likely to induce most or all
subscription journals to offer a paid Gold OA option.

(2) RCUK's  forbidding authors to publish in a journal that does not offer
either Libre (CC-BY) Gold OA or 6/12 Green is likely to those subscription
journals that hesitate to offer Libre (CC-BY) Gold OA to offer Gratis Gold
OA and to extend their Green embargoes to RCUK-impermissible limits, so
that RCUK authors must pay for Gold (with or without the help of the RCUK
"block grants") in order to be able to fulfill the RCUK  immediate deposit
requirement -- otherwise they cannot publish in their chosen journal at all.


This policy is not only likely to generate an RCUK author revolt, with
confusion and non-compliance because of the constraints on journal choice
and the obligation to pay publishers extra for Gold out of scarce research
funds rather than providing cost-free Green -- but if adopted it will also
hurt Green OA mandates in the rest of the world, because of the incentive
to subscription publishers worldwide to adopt long Green OA embargoes in
order to increase hybrid Gold revenues from the UK.

The UK only produces 6% of the world's published research. Even if it can
afford to pay extra for Libre (CC-BY) hybrid Gold OA out of its research
funds, over and above what UK institutions are paying for subscriptions, it
is not clear that the rest of the world, producing the remaining 94% of
published research, can afford to do that for Libre (CC-BY) hybrid Gold OA
-- nor that it would want to, even if it could, instead of just mandating
cost-free Gratis Green OA, while subscriptions are paying in full for the
costs of publication.

(d) is this concept defined by a community process?
>

I am not sure what you mean: Are you asking whether OA policy on the part
of institutions and funders is "defined" by a community process?

I'd say sometimes, like many policies (good and bad), it is a top-down
decision, with varying degrees of community consultation.

There seems to have been little consultation of the UK research community
in the case of the latest RCUK policy.

In contrast, there was a great deal of community consultation in the case
of the 2004 UK Parliamentary Select Committee recommendation to mandate
Green OA, as well as in the EC policy, in response to an EC petition, and
even moreso in the US, where both the FRPAA and President Obama's Office of
Science and Technology have been doing extensive community consultations.
Harvard's copyright reservation policy was the result of a faculty
consensus, and a few other universities have followed the same model.

But top/down policies can be well-received by researchers too, if they are
beneficial ones. Such was the case at Southampton ECS, the first Green OA
mandate of all, highly successful and well received, and since adopted by
many others. So too with the Liege model mandate, an improvement on
Southampton's (because of the link to performance assessment and the
integration with funder mandates), likewise highly successful and well
received, and since being adopted by others.

- Under what circumstances can a "green-created" document be copied from a
> repository? Or is it "Open Access" only when the reader has access to the
> Internet?
>

It can be copied by any user, including machines, because it is Gratis --
free online -- but it cannot be re-published: That would be Libre OA.

Gratis OA means free online access, including reading, linking,
downloading, printing, storing, and data-mining locally (as well as
harvesting, inverting and indexing of navigation and research by Google
Scholar and countless other search engines, none of whichseem to feel that
they need a Libre OA license to do so!)

- If only part of the community has access to a document in a repository
> (as in Ghent and many other repositories) is this document counted as Open
> Access?
>

If it is not online free for all, it is not Gratis OA.

Stevan Harnad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ecs.soton.ac.uk/pipermail/goal/attachments/20120829/f1736df2/attachment-0001.html