[OAI-eprints] Re: Guide juridique CNRS

Tue Nov 29 08:57:53 EST 2005

A knowledgeable colleague from France has pointed out to me that I did
not read the CNRS legal guide sufficiently closely, as there are other
corrections that need to be made in it:

    http://publicnrs.inist.fr/IMG/pdf/guide_bis_juridique_publiCNRS.pdf

(1) The passage requiring prior permission to deposit theses makes no
sense and should be removed. The decision is entirely in the hands of
the author, particularly after the thesis has been accepted. (Till then,
it is simply an unrefereed text, and can be deposited as such too, if
the author wishes, and if the institutional repository has a deposit
category for unrefereed texts. None of this is a legal matter at all.)

(2) There should be no restrictions among document-types and mark-ups
eligible for deposit. The usual types (HTML, XML, DOC, TEX, PDF, PS
etc.) should all be fine. (Again, document-type or mark-up are not legal
matters. The guide confuses institutional deposit policy rules with
legal questions.)

(3) In general, although a lot of the legal information is correct and
helpful, there is rather too much accent on legalism in the CNRS document,
which somewhat loses sight of the most simple and fundamental overarching
fact: that self-archiving is done by researchers, for researchers,
and it involves their very own writings.

(4) This simple, direct, overarching essence needs to be conveyed
clearly up-front, to keep all the legalism in context and
proportion. 

(5) There is still far too much of the permissions-librarian and
IP-jurist mentality behind the CNRS's and many other such "legal
guides." Permissions-librarians deal mostly with the 3rd-party right
to access writings by authors from *other* institutions, not the right
to make the institution's own writings accessible; and IP jurists
deal mostly with writings for which the author wishes to be paid for
access. Open Access is hence not a very natural concept for either of
these specialists. It is a researcher-to-researcher matter, and has no
real legal impediments:

    (5a) The authors of research articles and theses can and should
    self-archive all of their research writings, optionally before,
    but imperatively after they have been accepted for publication (if a
    journal article) or accepted for a degree (if a thesis).

    (5b) Both the metadata (authors, date, title, journal, institution,
    abstract, etc.) and the full-text of all writings of which the author
    is indeed the author can, without exception, be immediately deposited
    in the author's own institutional repository. If there are any legal
    questions at all, they concern the access-setting, not the deposit,
    which is an institution-internal record-keeping matter.

    (5c) Access to the metadata can and should always be set to
    Open Access (OA). If there is any reason the author does not
    wish to set access to the full-text as OA, they can set it as IA
    (Institution-jnternal access only).

    (5d) For unpublished writings or unrefereed papers not yet submitted
    for refereeing, or for theses, whether accepted or not yet accepted,
    access can always be set to OA.

    (5e) For accepted journal articles, the publishers of about 70% of
    them endorse self-archiving of the accepted final draft, and about
    23% more endorse self-archiving of the presubmission draft (to which
    corrections can of course be appended). (These endorsements are not
    really needed, but they show that for 93% of research output there
    is not even a notional legal issue.) http://romeo.eprints.org/stats.php

    (5f) For the remaining 7% of articles, the access can provisionally
    be set at IA if the author wishes, and the author can email eprints
    to all the would-be users who access the open metadata and need the
    full-text because their institution does not subscribe to the journal.

    (5g) A better option is to set access at OA for 100% of articles and
    consider whether or not to re-set it as IA only if and when there
    is ever a request from the publisher to do so.

The legal details in the guide are not wrong; with a clearer statement of the
practical priorities, they will be useful for reference.

Stevan Harnad

On Mon, 28 Nov 2005, Stevan Harnad wrote:

> The CNRS Guide is an excellent, well-documented, well-written,
> well-thought-out guide to the legal aspects of self-archiving. I strongly
> urge that it be translated also into English and adapted for use by
> other institutions.
> 
>     http://publicnrs.inist.fr/IMG/pdf/guide_bis_juridique_publiCNRS.pdf
> 
> The only point I would suggest revising is point 4. (The optimal default
> policy is to self-archive in all cases, and to consider publisher
> removal-requests, if/when received, on a case-by-case basis. This is
> simpler than explicitly negotiating self-archiving rights in advance.)
> 
> Especially useful in the CNRS guide is the following passage (highly
> pertinent to the support of the already over-cautious RCUK proposed
> policy):
> 
>     "Nous vous rappelons que vous détenez toujours le droit de déposer une
>     notice bibliographique dans HAL, complétée du numéro DOI du document,
>     et de déposer le plein texte d'un document en mode caché."
> 
> (You can *always* deposit the metadata plus text, with the metadata set
> to Open Access [OA] and the full-text set to institution-internal [IA]
> access. Then of course those would-be users webwide who try to access
> the full text on the basis of the metadata [author, title, joutnal, date,
> abstract, etc.] and hit an "IA" barrier can simply email the author for
> a copy of the eprint.)
> 
>     http://openaccess.eprints.org/index.php?/archives/20-guid.html
>     http://www.ecs.soton.ac.uk/~harnad/Temp/berlin3-harnad.ppt
> 
> It would also be more useful if, in addition to linking the Romeo/SHERPA
> directory of publishers' self-archiving policies (which indexes the
> policies in terms of the 126 publishers registered to date), the CNRS
> guide also linked the Romeo/Eprints version of this directory (which
> indexes the policies in terms of the 8630 individual journals registered
> to date): http://romeo.eprints.org/
> 
> Stevan Harnad
> 
> On Mon, 28 Nov 2005, Minh Ha Duong wrote:
> 
> > Hello/Bonjour,
> > 
> >   Le CNRS a préparé un Guide pratique des aspects juridiques liés au dépôt des 
> > publications. Les internautes sont invités à faire part de leurs observations 
> > sur le projet de texte que vous trouverez à: http://publicnrs.inist.fr/
> > 
> > Minh
> > -- 
> > Minh HA DUONG                                    Chargé de recherche CNRS
> > www.centre-cired.fr/perso/haduong/       CIRED, Campus du Jardin Tropical
> > tel: +33 1 43 94 73 81                   45bis ave. de la Belle Gabrielle
> > fax: +33 1 43 94 73 70                      F94736 Nogent-sur-Marne CEDEX
> > haduong at centre-cired.fr                                            FRANCE
> > 
> 
> 