On Tue, 12 Jun 2007, Millington Peter wrote:
> The policies in the OpenDOAR Policies Tool were derived mainly from an
> extensive survey of around 300 repositories that was undertaken at the
> beginning of 2006. The options therefore generally reflect the actual
> permissions, restrictions and practices of real repository
> administrators.
We definitely need an inventory of repositories and their contents,
but we definitely don't need an inventory of their arbitrary
restrictions. Restrictions are bad enough, but helping to advertise and
propagate them is not something a service that is meant to encourage
OA ought to be doing.
In a quip about SHERPA-Romeo I asked whether a publisher policy that only allowed
deposits on Tuesdays by authors with maternal uncles with chestnut-coloured irises
deserved to be immortalized in a "Chestnut" category of its own in SHERPA-Romeo. I
could ask much the same question about OpenDOAR: Do you think it helps inform and
encourage the opening of doors to aid and abet their shuttering by faithfully
cataloguing all of its benighted variants?
Neither Romeo nor OpenDOAR should emulate library catalogues, describing
holdings, along with IP restrictions on them. That is not what OA or
the services meant to be informing about and promoting OA should be about.
All Romeo need to do is say which journals/publishers are GREEN
on postprints, or, if not, then PALE-GREEN on preprints, or neither
(GREY). Even the length of the postprint embargo hardly merits a mention;
a link to the journal's policy page would do. That's all. Mechanically
cataloguing more of the details of arbitrary restrictions simply promotes
arbitrary restrictions while serving no useful purpose whatsoever,
insofar as promoting or even informing about OA is concerned: none.
The very same is true of the arcana of repository restrictions: Users
will not be consulting OpenDOAR in order to find out what they may *not*
do with their content. Let them find that out from the IRs themselves,
when they actually search and use. OpenDOAR will be consulted to see how
many IRs there are, what kinds of contents they house, how many items,
what proportion is full-text, how fast they are growing. It would be
very useful to track all of that, by both publication date and deposit
date, for full-text. It even makes sense to catalogue what proportion
of the full-text contents is in Open Access and what proportion in
Closed Access, so those growth rates too can be tracked and compared
across IRs, fields, countries, and time. But any further foregrounding
of the minutiae of arbitrary IRs' non-OA or anti-OA practices is simply
compulsive cataloguing, and counterproductive to boot.
> Some are sites VERY open with their material (e.g.
> Lincoln's "You may do as you wish with any metadata harvested from this
> site") to highly proprietorial (e.g. ULP Strasbourg's "Text, data and
> images should not be downloaded, nor redistributed nor published
> anywhere else").
Users will not be consulting OpenDOAR for information on what they may
not do with Strasbourg content. They will be consulting Strasbourg in
global searches, and doing what they can with what they hit.
(How, by the way, does one read what is in the Strasbourg IR without
downloading it? And does the eye-colour of one's next of kin have anything
to do with it? If Strasbourg said "You may click, but you may not peek"
would OpenDOAR faithfully catalogue that too?)
> It is up to users of the Policies Tool as to how open or restrictive
> they are. Administrators can, and perhaps should bypass the restrictive
> options. Les's example is a case in point. (A few repositories had shot
> themselves in the foot by prohibiting all harvesting, thus disallowing
> Google and other search engines from indexing their content. We
> therefore added the riders - adapted from Caltech's policies - to
> improve matters.)
What is the purpose of OpenDOAR? To canonize every arbitrary practice
being improvised today (why? for historical purposes?), to help IRs
enforce restrictions? or to give a rational picture of the state of play
in IRs (numbers, size, growth), with a view to encouraging more IRs,
more growth, and more openness?
> The Tool also provides links that automatically select "minimal" or
> "optimal" policies. The "minimal" policies generally do the least
> possible to comply with the spirit of Open Access, whereas conversely
> the "Optimal" policies go as far as possible to promote re-use (and
> encourage best practices).
But who is the intended user, and to what use are they expected to
put this information? Clearly someone searching the OA literature on
a certain topic will not first consult OpenDOAR to see which IRs to
include or exclude from his search. Is it for IR managers to trade
notes on their respective arbitrary restrictions? Or to help etch
those arbitrary restrictions in stone?
> We've tried to do a good job with the policy
> options and recommendations, but are not written in stone. There is
> always room for improvement, so if Les or anyone else has any
> constructive suggestions, we would be very happy to receive them.
If you mean that, then I suggest that all policy details that do not serve
a useful, constructive purpose should be omitted from cataloguing. (This
suggestion applies to both SHERPA-Romeo and OpenDOAR.) Index only the
properties that are relevant to monitoring and promoting OA's growth,
not the ones that will only help to inhibit it.
Stevan Harnad
> ________________________________
>
> From: Leslie Carr [mailto:lac_at_ecs.soton.ac.uk]
> Sent: 12 June 2007 16:26
> To: American Scientist Open Access Forum
> Cc: Millington Peter
> Subject: Re: Get the Institutional Repository Managers Out of
> the Decision Loop
>
>
>
> Peter Murray-Rust [PM-R] replied
> <http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=364> :
>
> "Stevan Harnad... has been consistent in arguing
> the logic [of what comes with the OA territory
> <http://openaccess.eprints.org/index.php?/archives/259-guid.html> ]...
> and I agree with the logic... [but]... several repository managers at
> the JISC meeting
> <http://www.jisc.ac.uk/events/2007/06/repositories_conference.aspx>
> [said] I could not have permission to do [such things] with their
> current content. I asked 'can my robots download and mine the content in
> your current open access repository of theses?' - No. 'Can you let me
> have some chemistry theses from your open access collection so I can
> data-mine them?' - No - you will have to ask the permission of each
> author individually.
>
>
> The OpenDOAR repository policies tool tends to act towards
> over-cautiousness in the policies that they suggest for data and
> document reuse.
> The current policies that they produce have options to
> explicitly allow services that do full text indexing and citation
> analysis, BUT THAT IS ALL.
> By enumerating the potential allowable services they are
> effectively stifling innovation and research, and that is a BAD thing.
> The last thing that OA advocates ought to do is build up ANOTHER
> rights-withholding infrastructure.
>
> I do hope that this a a short-sighted transition phenomenon, but
> it should certainly be addressed now (and strongly).
> --
> Les Carr
>
>
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses, which could damage your computer system:
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
Received on Tue Jun 12 2007 - 22:05:11 BST