I agree completely with UCOP's Roy Tennant that distributed
author/institution self-archiving rather than just central
discipline-based author self-archiving is the most promising
route to open-access for the refereed research literature.
However, I would still very strongly urge calling a spade a spade:
Whether it is literally done by the hand of the author himself,
or by proxy, on his behalf, by his institution, it is most definitely
still SELF-archiving that we are talking about! Nothing is to be
gained by calling it by another name; rather, old misunderstandings
that have been holding us back for a decade would simply be courted
anew by trying to invent yet another new category, and a less
The heuristic function of the "self" in self-archiving is not that the
author must by all means perform the act with his own hands, if he
hasn't the time (or the skill). On the contrary, although it is
undeniably extremely easy (if someone as technically incompetent and
maladroit as myself can do it), I have always strongly advocated
institutional "proxy" self-archiving for any author who needs it:
Indeed, a reliable institutional proxy self-archiving service is one
of the two most important prerequisites for successfully filling an
institutional Eprint Archive. (The second prerequisite is a strong,
explicit, official institutional self-archiving policy for all faculty
But to call this "institutional archiving" instead of
"author/institution self-archiving" is simply to invite researchers
(and institutions) to continue to ignore the real contigency that is
involved here, the one that makes it all legal, feasible, desirable
and urgent: The author is self-archiving his OWN give-away work!
By the way, let it be noted that it was institutional self-archiving
by authors, not central self-archiving, that was the gist of the
original Subversive Proposal: http://www.arl.org/sc/subversive/
Then, as now, the issue was never the trivial one of "who owns the fingers
that perform the keystrokes?", but "who is the 'self' who owns (and can
hence give away) the intellectual property in question?" I regret to
say that many universities are still a little confused about this, and
conflating it with other intellectual property issues (such as patents,
pedagogic materials, etc.).
The special literature that the Budapest Open Access Initiative is
targetting is the author give-away literature (pre-refeeing preprints
and peer-reviewed, published postprinta). The only one who is
empowered to archive that literature is the author himself -- the
self-same "self" that figures in author/institution self-archiving.
There are many things an institution can archive, including the
give-away intellectual property of its own researchers, if those
researchers elect to give it away, and allow someone at their
institution to key it in on their behalf, by proxy. But nothing is gained,
and everything is at risk of being lost, if this special subset
of the many things a university can archive is conflated with all the
rest, by calling it just another piece of university archiving.
It is not. And universities' Eprint Archives risk remaining empty,
as so many archives are, if this critically important distinction is
not maintained and given the prominence it requires. (Along with all
necessary clarification on the subject of the availability of university
fingers to do the proxy self-archiving on the author's behalf.)
Enough about this vexed piece of semantics. (It is remarkable how many
subtle points exist on which there is the potential for paralytic
misunderstanding in this domain which looks so trivial and straightfoward
on the face of it...).
My only other point is about the free eprints.org software, which
we are supplying to universities in order to hasten and facilitate
author/institution self-archiving: University of California's
California Digital Library (CDL) was one of the earliest
implementers of the earliest release of the eprints.org software
(See Ken Weiss's and Catherine Candee's messages below):
Since that time, CDL have moved on to create their own institutional
archiving software. Meanwhile, eprints.org has continued to be developed
and upgraded. Later releases have been adopted and reviewed by, for
example, California Institute of Technology
The eprints.org software continues to be upgraded to incorporate
each successive upgrade of the Open Archives Initiative (OAI)
protocol, and continues to add the features requested by its
growing number of users, but the objective is not to make sure
very university adopts this software! The objective is to make
sure every university implements self-archiving!
A few commentas follow below:
On Tue, 28 May 2002, Roy Tennant wrote:
> On Friday, May 24, 2002, at 01:40 PM, Peter Suber wrote:
> > Excerpts from the Free Online Scholarship (FOS) Newsletter
> > May 23, 2002
> > [text deleted]
> > There are two primary paths to FOS: open-access journals and
> > self-archiving...
> Knowing full well what I may be in for, I want to take issue with the
> term "self-archiving", which Stevan and others are apparently using as
> an umbrella term for activities that I think should now be split apart.
> When "self-archiving" is used, I tend to think of arxiv.org and other
> repositories where the author is indeed depositing their own paper.
You may tend to think of that, but I don't think that most of the world's
universities and researchers across disciplines, most of whom have not
even heard of arxiv.org, tend to think of it. What I hope they think
of when they hear of self-archiving is self-archiving.
What will be very helpful in encouraging them along this line of thinking
is to make it clear to them that a service is available at their
university to do their self-archiving for them, if they feel unable to
do it for themselves.
> But this term seems much less useful to describe repositories that are
> institutionally sponsored, and for which the depositing process may be
> out of the hands of the author (performed by a staff member, for
> example). I'm beginning to find this latter model much more compelling
> in many instances than true "self-archiving".
As I said, this is the model that I and others find the most compelling
too. But it will not be hastened and facilitated by subsuming it under
general university archiving.
[And if we are concerned about terminology that will encourage rather than
alienate authors in their inclination to self-archive their own work,
may I suggest that you describe the service UC is ready to
provide as "taking the chore of keying it in OFF their hands" (rather
than "taking the (prerogative to self-archive their own work) OUT
OF their hands"!]
> It may appear that I'm splitting hairs, but I think not. By depicting
> only two primary paths to free online scholarship you run the very real
> risk of turning away those who have no interest in spending a lot of
> time and effort to do what is required to "self-archive". And despite
> Peter's enthusiasm (see below) this process can still, in some cases, be
> both time consuming and painful.
I agree that everything possible should be done to draw researchers
toward self-archiving their giveaway work, rather than turning them
away from it. A university proxy service to key it in for them will
be a great help in facilitating the first wave and in reaching critical
mass. But don't risk alienating your authors either, by giving them the
imppression that you are going over their heads with their intellectual
property (even though it is giveaway!). It is a service you are offering,
to help them self-archive their own work. You are not doing an autonomous
piece of university archiving, co-opting their work!
Having said that: I usually try to remember to call it "author/institution
self-archiving" as often as I can. As long-winded as these portmanteaux
terms sound, they may well be necessary in order to short-circuit the
predictable misconstruals that have already held us back for so long.
I will do a systematic swap in a lot of our documents to make sure the
long-winded version appears, to make it clear that the self-archiving
we are promoting is a collaborative effort between authors and their
> > If you want to deepen the discussion, focus on why self-archiving isn't
> > spreading more rapidly than it is. Creating an archive is now painless
> > with free software, maintaining an archive takes minimal effort, hosting
> > one takes server space that any university could donate without
> > noticing,
> > and the benefits are immediate and cumulative.
> As someone who has created several, I can tell you that creating an
> archive is far from painless. There is free software to be had,
> certainly, but the out-of-the-box interface requires a good deal of work
> to both brand it and make it sufficiently understandable as to be
> moderately usable. Once it is usable, the garden variety faculty member
> (mostly the people NOT on this list) will nonetheless find it difficult
> to understand and time-consuming to use. If you don't believe me,
> perhaps you will believe the experiences of the authors of this article:
> http://www.ariadne.ac.uk/issue31/eprint-archives/ when they say "The
> eprints.org software has a self-archiving facility but our experience of
> this is that it is rather long winded and requires a certain amount of
> IT literacy. Some users may well be put off." That has been our
> experience as well.
As I said, this free, generic software is constantly being upgraded
to make it as easy as possible to install and use. (Perhaps if you
make the UCOP software freely available, we can get some comparative
reviews for ease of installation, maintenance and use...?)
The objective at the moment is not to wait for the optimal software,
but to get OAI-compliant software that is good enough to speed
self-archiving on its way to reaching critical mass, all over the
world. Once we manage that, optimizing the software will be a much
But certainly a university proxy self-archiving service will be
a very welcome ally in this cause!
(And in fairníess, the Pinfield, Gardner & McColl article you cite above
also had some rather good things to say about the eprints.org
software. It wasn't all grumbling...)
> Even should the interface be dead simple, a number of faculty will
> nonetheless find other reasons not to do it. Therefore, our model is to
> use existing organizational structures within the university to do the
> depositing. That is, we target staff at university "organized research
> units" (institutes or centers) and academic departments for training in
> depositing the papers of their associated faculty. Our premise is that
> most faculty shouldn't ever have to know how to do it, just as many do
> not need to know or care about how what it takes to put their papers up
> on their institute's web site. So far it appears that this model will
> allow us to scale up this service fairly rapidly and minimize our
> support overhead.
That sounds terrific, and I hope it will be wonderfully successful. If
it is, we will certainly recommend it as a model for others to adopt.
> Therefore I think it does the effort to free online scholarship a
> disservice to conflate staff-supported institution-based repositories
> with "self-archiving". Were I a faculty member with interests other than
> freeing online scholarship (of which I assure you there are many) I
> would find the term "self-archiving" off-putting. I would wonder why on
> earth I should take over a task that had never been mine to begin with.
The logical and practical core of the motivation to self-archive is NOT
to free the online scholarship of others, but to free the access to
the researcher's OWN research output, thereby maximizing its visibility,
access, uptake, hence impact. Free access to the research of others is
merely a bonus, a side-effect, of universal self-archiving and the Golden
Rule, a stable, self-interested piece of reciprocal altruism.
Neither the library serials budget crisis nor general crusades to free
access to online scholarship is a significant factor in this motivation
(though both of course stand to benefit greatly from its success).
So I can only repeat: There is every reason to emphasize the "self"
in the self-archiving initiative, both to make it clear that it is
the authors' own (giveaway) work that they (sic) are giving away,
and that it is their own research impact that stands to benefit from
> All of this is not to take away from the useful work being done by
> Stevan, Peter, and many others. We are, after all, advocating many of
> the same things. So please take this message in the spirit in which it
> is intended -- to try to tease out differences and nuances in the model
> that has so far been put forward and bring them to light.
Roy's message (and all other such thoughtful reflections on the puzzling
and often frustrating new paths we are trying to navigate here) are most
welcome here. Anyone who thinks he is a prophet and knows the sure way
has not been at this long enough...
> These are my personal comments, and are not intended to necessarily
> represent the views of my employer, the eScholarship initiative of the
> California Digital Library.
> Roy Tennant
NOTE: A complete archive of the ongoing discussion of providing free
access to the refereed journal literature online is available at the
American Scientist September Forum (98 & 99 & 00 & 01):
Discussion can be posted to:
See also the Budapest Open Access Initiative:
and the Free Online Scholarship Movement:
Received on Wed May 29 2002 - 18:13:16 BST