Re: Institutional Repositories

From: Arthur Sale <>
Date: Sat, 24 Dec 2005 09:30:08 +1100



I have distinctly different views from this.


(1)    I view an institutional repository as a place which encompasses
(a) and (b), while adding digital theses and eResearch (research
datasets) to that research output. This is the sense in which the phrase
has been used in the past, and the &#8216;institutional&#8217; was added
to distinguish it from &#8216;central&#8217; repositories or
&#8216;subject&#8217; repositories. The word repository derives from the
deposit of papers by authors.

(2)   Roles (c), (d) and others are recent ring-ins from what is more
properly called a digital library. This phrase was indeed used for them,
and conflating them with the research-oriented (a) and (b) is very
dangerous. (a) and (b) are simple, well-known, standardized, easy to
achieve, low-cost targets. (c), (d) and the many others which I
won&#8217;t even list here, are uncertain developmental areas, with no
standardization and plenty of experimentation: the exact antithesis of
(a) and (b).

(3)   The mental trap that this misuse of the term &#8216;institutional
repository&#8217; leads one into, is thinking that one piece of software
(one database if you like) can serve all functions. The term
&#8216;Digital Library&#8217; didn&#8217;t do that. This is as stupid as
thinking that a university&#8217;s central administration can use one
piece of software and one database for its financial records, its
personnel records, its student database, its timetabling, etc. No
administrator would be that foolish, but some people do not seem to
recognize the same is true of broad digital library functions.
&#8220;Horses for courses.&#8221;

(4)   It is time we reverted to using the term &#8216;institutional
repository&#8217; for open access functions to do with deposits of
research output.

(5)   All teaching functions, e-publishing, digitization of image
collections, presentation of cultural artifact collections, etc should
revert to being described as being digital library functions, and those
interested in them should preferentially stop trying to stuff them into
the same database as institutional repositories, to the detriment of
both. R&D activities simply do not blend well with well-understood





From: American Scientist Open Access Forum
Behalf Of Richard Poynder
Sent: Friday, 23 December 2005 21:13


Hi All,


I am aware that the concept of "the institutional repository" (IR) tends
to mean different things to different people, so I would be interested in
hearing people's views on the topic.


Amongst the different roles I have heard people argue that IRs can/should
play are:


 a) as a repository for a university's research output, with the aim of
increasing access to that research, and so enhancing its impact


 b) as a tool for preserving and curating a university's research output


c) as a tool to assist a university in its digital publishing ambitions,


d) as a tool to enable universities offer digital courseware and online
learning services.


Would others agree that IRs are viewed as potentially assisting in all
these tasks?


Are there other tasks people see being assigned to IRs?


Can an IR play a number of different roles simultaneously, or should it
be restricted to just one (possibly two) roles? What are the pros and
cons of using an IR to play more than one role?


Also, what solutions are people choosing when setting up IRs? I am aware
that there are now a number of software and service providers (both for
profit and non-profit), including Eprints, DSpace, Fedora, ProQuest and
BMC. But what are people's views on the roles these different providers
can/should play? Are they playing them well?


Finally, I am also curious as to the current state of play in the major
universities (Stanford, Yale, University of California, MIT, CalTech,
Harvard, Oxford, Cambridge etc. etc.). I assume they all now have IRs up
and running, but what sort of IRs are they creating, what solutions are
they choosing when creating them, what roles are being assigned to these
IRs, and how are these universities going about populating them with the
institution's research output?



Richard Poynder

Freelance Journalist


Received on Sat Dec 24 2005 - 00:15:21 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:48:10 GMT