List members will doubtless correct me if I am wrong, but it seems to me
that the nub of this issue is that Peter Murray-Rust believes that when
a research library pays a subscription for a scholarly journal (or a
collection of journals) the subscription should give researchers at that
institution the right both to read the content with their eyeballs, and
to mine it with their machines -- and that this should be viewed as an
automatic right.
Publishers appear to take the view that text mining is not an automatic
right for subscribers. Consequently, some publishers believe that
machine access needs to be separately negotiated, and on a case-by-case
basis (preferably with the institutional library rather than individual
researchers), while others simply will not contemplate allowing it (or
will only allow it on payment of an additional fee -- again, on a
case-by-case basis).
Publishers' position, I must assume, rests partly on copyright claims,
but mainly on what is written into the contract when the institutional
library negotiates the subscription with the publisher.
There is also the question of what researchers can do with the the data
once they have extracted it, which seems to me to be a more complex issue.
I wonder if any other publishers might like to comment on this?
On whether this is an OA issue, and so relevant to GOAL, I would suggest
that it is partly an OA issue, and partly not an OA issue. One might
argue that it is not an OA issue because it concerns access to
subscription content, rather than OA content. On the other hand, one
could argue that it is an OA issue, since both Stevan Harnad and Peter
Murray-Rust are asking for access to subscription content. Where Stevan
is asking for eyeball access to this material, Peter is asking for
machine access.
I realise, of course, that Stevan is happy to have access to the version
of a paper as it exists prior to publication (the pre-print), but,
nevertheless, he is asking for access to papers that have been published
in subscription journals.
Either way, since this issue is inextricably bound up with discussion of
gratis vs. libre OA, I think it is relevant and appropriate to be
discussed on GOAL. If others disagrees, please say so.
Richard Poynder
GOAL Moderator
On 12/05/2012 09:23, Peter Murray-Rust wrote:
[Note that as this is a moderated list my replies may not appear
immediately - this may give the impression that I am ignoring mails
when I am not].
On Sat, May 12, 2012 at 8:29 AM, Wise, Alicia (ELS-OXF)
<[email protected] <mailto:[email protected]>> wrote:
Hi Peter,
Thanks for this. I've communicated that we are happy in principle
for you to mine our content,
Good - that is agreed.
and there are only some practical issues to resolve. We have
successfully concluded the technical discussion, and I believe
you, your colleagues, and my technical colleagues are all happy
with the proposed technical mechanism.
I don' think this is true at all. Rather than allowing us to mine the
actual journal articles in siti you have pointed us at some zip files.
We have no idea how comprehensive these are or how up to date. We
understand publication of articles, we do not understand your zip files.
Getting content from the published journal articles is trivial and is
the same mechanism for all publishers. Fine the DOI, download it.
Getting articles from 100 publishers through 100 different arcane
mechanisms is a nightmare. If you want to make life easy for us, let
us use the articles.
Next, I'ld like to double-check that I have correctly understood
what you and your colleagues will do and who will have access to
which content/extracts.
I will extract factual data and I shall publish them Openly to the
whole world. For your interest my first research is likely to be in
the phylogenetic trees of arthropods - whcih impacts on trillions of
dollars (sic) of human welfare. Polination allone is 150 billion
dollars. I shall not publish discursize text unless it is required. I
shall not publish the "publishers PDF". But I shall publish anything
that represents facts, in botha critical and uncritical manner.
And in the area below - just one project, I expect to use 10,000
articles per year. So , in principle, I intend to mine millions per year.
I shall take silence as assent that this is agreed.
Finally, we have an existing agreement with the U of Cambridge
library and we need to ensure there is some language in that
agreement -- or a side letter - to enable content-mining. We
aren't far off at all -- and I suspect we could resolve this in 1,
possibly 2, quick conversations. If you prefer not to interact
with the Cambridge librarians, I can do this separately.
The librarian has indicated that it is important that I be allowed to
text-mine and believes that publishers should set out terms. I will
ask her permission whether I can reproduce her letter publicly. She
does not believe I should be wasting my research time negotiation with
indivdiual publishers and she sets out one publisher (not Elsevier) as
having made their terms clear. So the first thing is for you to say
something in plain language which says what I can do.
I do not regard this as a negotiation. I believe I have rights and
will - with otherrs such as RichardP be putting out those rights
shortly. If libraries wish to agree on a more limited use I shall be
unhappy with them.
Perhaps it would be helpful for me to clarify the important role
that I believe the Cambridge library has to play. This role is
not to vet your research to see if you can carry it out, but to
ensure that the language necessary to enable this to happen is
included in their various agreements with publishers.
And to take out the restrictions that you have added.
This is the way that libraries have been able to create the
existing information environment on campus where you, and your
colleagues, can access e-journals from home or your office or out
in the field. All the agreements/arrangements/technology that the
library has put in place to create this environment, and to ensure
that it is easy to access and use, are generally invisible to
researchers - even those who use this information environment on a
daily basis.
And we pay enormous amounts for it. My grant income is paying
Elsevier, including you. IMO it is one fo the most inefficient uses of
public money.
This is the sort of environment/experience needed for researchers
who wish to text mine as well. As an early adopter -- indeed a
pioneer in text mining -- you are forging a trail. Librarians
will help to maintain that trail so that many, many others can
follow easily in your footsteps.
Let me be clear. This discussion is wasting my time - I could be
doing research. If I have to do this with every publisher it will
destroy this research.
If I leave it to my librarians, then they may well agree to the awful
and restrictive deal that Elsevier forced on Heather Piwowar at UBC
where one researcher was given permission for one project. She is not
allowed to publish the full research openly so it's of no use to me.
So I reiterate:
* a public statement that I can mine Elsevier journals in any amount
for whatever purpose and in whatever form.
* that I can publish the factual information extracted without
restriction.
I have no details of the negotiations you are transacting with my
library - I am a scientist not a contract negotiator. I have made my
wishes very clear to the library.
P.
With kind wishes,
Alicia
Dr Alicia Wise
Director of Universal Access
Elsevier I The Boulevard I Langford Lane I Kidlington I Oxford I
OX5 1GB
P: +44 (0)1865 843317 <tel:%2B44%20%280%291865%20843317>I M: +44
(0) 7823 536 826 <tel:%2B44%20%280%29%207823%20536%20826> I E:
[email protected] <mailto:[email protected]> I
*Twitter: @wisealic*
*From:*[email protected] <mailto:[email protected]>
[mailto:[email protected]
<mailto:[email protected]>] *On Behalf Of *Peter Murray-Rust
*Sent:* 11 May 2012 23:47
*To:* Global Open Access List (Successor of AmSci)
*Subject:* [GOAL] Re: OA and scholarly publishers
On Fri, May 11, 2012 at 1:16 PM, Richard Poynder
<[email protected] <mailto:[email protected]>>
wrote:
Many thanks to Alicia Wise for starting a new conversation thread.
Let's recall that Alicia's question was, "what positive things are
established scholarly publishers doing to facilitate the various
visions for open access and future scholarly communications that
should be encouraged, celebrated, recognized?"
Alicia Wise already knows my reply - she has had enough email from
me. The publishers show withdraw contractual restrictions on
content-mining. That's all they need to do.
My university has paid Elsevier for subscription to the content in
Elsevier journals. I believe I have the right to mine the content.
Elsevier has written a contract which forbids me to use this in
any way other than reading with human eyeballs - I cannot crawl
it, index it, extract content for whatever purpose. I have spent
THREE years trying to deal with Elsevier and get a straight answer.
See
http://blogs.ch.cam.ac.uk/pmr/2011/11/27/textmining-my-years-negotiating-with-elsevier/
The most recent "discussions" ended with Alicia Wise suggesting
that she and Cambridge librarians discuss my proposed research and
see if they could agree to my carrying it out. I let the list
decide whether this is a constructive offer or a delaying tactic.
It certainly does not scale if all researchers have to get the
permission of their librarians and every publisher before they can
mine the content in the literature. And why should a publisher
decide what research I may or may not do?
All of this is blogged on http://blogs.cam.ac.uk/pmr
Yes - I asked 6 toll-access publishers for permission to mine
their content before I submitted my opinion to the Hargreaves
enquiry. Of the 6 publishers (which we in the process of
summarising - this is hard because of the wooliness of the
language) the approximate answers were:
1 possibly
4 mumble (e.g. "let's discuss it with your librarians")
1 no (good old ACS pulls no punches - I'd rather have a straight
"no" than "mumble")
In no other market would vendors be allowed to get away with such
awful customer service. A straight question deserves a straight
answer, but not in scholarly publishing.
Just in case anyone doesn't understand content mining, the
technology is straightforward. The only reason it's not done is
because Universities are afraid of publishers. I estimate that
tens of billions of dollars worth of value is lost through being
forbidden to mine the scholarly literature.
If Alicia Wise can say "yes" to me unreservedly, I'll be happy.
P.
--
_______________________________________________
GOAL mailing list
[email protected]
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal
