20 · SPEC Kit 292
stricted. For example, some repositories contain
materials that are limited only to campus users,
while others have materials that are limited to a
specific department or groups of people (such as
a specific group of research faculty). Copyright is
only one reason that access to materials is limited.
Cultural concerns with primary source materials
and pending patents were also cited as reasons for
restricting access.
Although some institutions restrict access to
materials within their repository, few implement-
ers (3 or 9%) supply IR documents to external users
for a fee. Primarily, fees appear to be for re-use of
images or electronic thesis and dissertations. This
seems somewhat incongruous when one consid-
ers that 44% of the implementers limit access to
materials within their repository. However, there
are several possible explanations for this. First, in-
stitutions could be storing institutionally licensed
materials, such as images, in their IR. Second, the
process of collecting per-use fees is missing from
several popular open source software packages for
IRs. This makes it difficult for institutions to col-
lect fees on a per-use basis without extending the
software.
Benefits
Respondents’ comments indicate that the top two
benefits of IRs are enhanced visibility and increased
dissemination of the institution’s scholarship (34
responses or 68%) and free, open, timely access to
scholarship (23 or 46%). Preservation and stew-
ardship of digital content and preservation of and
long-term access to the institution’s scholarship are
close seconds (18 responses each or 36%), followed
by collecting and organizing assets in a central lo-
cation (12 or 24%). Four respondents (8%) report
that another benefit of an IR is the opportunity to
educate faculty about copyright, open access, and
scholarly communication.
Challenges
Among the top three challenges that respondents
face in implementing, promoting, and running an
IR are content recruitment/building a critical mass
of content (16 responses or 32%), staffing (15 or
30%), and faculty awareness/buy-in/interest/en-
gagement (14 or 28%). Copyright issues and com-
municating the benefits of the IR to faculty are close
behind. Adequate funding and other resources and
integrating the staff and workflow of IRs into exist-
ing structures were also recognized as challenges.
Conclusion
Based on the survey, what were the major character-
istics of operational ARL institutional repositories
at the start of 2006? Most IRs had been established
in the last two years (or had just been established).
By far, the library was likely to have been the most
active institutional advocate of the IR. It was also
likely to have been the primary unit leading and
supporting the IR effort, sometimes in partnership
with the institutional information technology unit.
The main reasons for establishing an IR were to in-
crease the global visibility of, preserve, provide free
access to, and collect and organize the institution’s
scholarship. In most cases, a project team had been
used to plan and implement the IR and a pilot proj-
ect had been used to determine IR-related issues.
If it was not still ongoing, the IR implementation
process had most frequently taken six months to
a year, with one to six months being the next most
common duration.
By a large majority, the most frequently used local
IR software was DSpace, with DigitalCommons
(or the bepress software it is based on) being the
system of choice for vendor-hosted systems. Local
systems usually either ran under variants of Linux
or Windows on an Intel-based server or under
Solaris on a Sun server. A typical IR held about
3,800 digital objects, with ETDs, article preprints
and postprints, conference presentations, technical
reports, working papers, conference proceedings,
and multimedia materials being the most common
types of documents. IRs normally support OAI-
PMH and, a little over half the time, OpenURL.
Most IRs had written policies and procedures
and the majority of them had been submitted to an
stricted. For example, some repositories contain
materials that are limited only to campus users,
while others have materials that are limited to a
specific department or groups of people (such as
a specific group of research faculty). Copyright is
only one reason that access to materials is limited.
Cultural concerns with primary source materials
and pending patents were also cited as reasons for
restricting access.
Although some institutions restrict access to
materials within their repository, few implement-
ers (3 or 9%) supply IR documents to external users
for a fee. Primarily, fees appear to be for re-use of
images or electronic thesis and dissertations. This
seems somewhat incongruous when one consid-
ers that 44% of the implementers limit access to
materials within their repository. However, there
are several possible explanations for this. First, in-
stitutions could be storing institutionally licensed
materials, such as images, in their IR. Second, the
process of collecting per-use fees is missing from
several popular open source software packages for
IRs. This makes it difficult for institutions to col-
lect fees on a per-use basis without extending the
software.
Benefits
Respondents’ comments indicate that the top two
benefits of IRs are enhanced visibility and increased
dissemination of the institution’s scholarship (34
responses or 68%) and free, open, timely access to
scholarship (23 or 46%). Preservation and stew-
ardship of digital content and preservation of and
long-term access to the institution’s scholarship are
close seconds (18 responses each or 36%), followed
by collecting and organizing assets in a central lo-
cation (12 or 24%). Four respondents (8%) report
that another benefit of an IR is the opportunity to
educate faculty about copyright, open access, and
scholarly communication.
Challenges
Among the top three challenges that respondents
face in implementing, promoting, and running an
IR are content recruitment/building a critical mass
of content (16 responses or 32%), staffing (15 or
30%), and faculty awareness/buy-in/interest/en-
gagement (14 or 28%). Copyright issues and com-
municating the benefits of the IR to faculty are close
behind. Adequate funding and other resources and
integrating the staff and workflow of IRs into exist-
ing structures were also recognized as challenges.
Conclusion
Based on the survey, what were the major character-
istics of operational ARL institutional repositories
at the start of 2006? Most IRs had been established
in the last two years (or had just been established).
By far, the library was likely to have been the most
active institutional advocate of the IR. It was also
likely to have been the primary unit leading and
supporting the IR effort, sometimes in partnership
with the institutional information technology unit.
The main reasons for establishing an IR were to in-
crease the global visibility of, preserve, provide free
access to, and collect and organize the institution’s
scholarship. In most cases, a project team had been
used to plan and implement the IR and a pilot proj-
ect had been used to determine IR-related issues.
If it was not still ongoing, the IR implementation
process had most frequently taken six months to
a year, with one to six months being the next most
common duration.
By a large majority, the most frequently used local
IR software was DSpace, with DigitalCommons
(or the bepress software it is based on) being the
system of choice for vendor-hosted systems. Local
systems usually either ran under variants of Linux
or Windows on an Intel-based server or under
Solaris on a Sun server. A typical IR held about
3,800 digital objects, with ETDs, article preprints
and postprints, conference presentations, technical
reports, working papers, conference proceedings,
and multimedia materials being the most common
types of documents. IRs normally support OAI-
PMH and, a little over half the time, OpenURL.
Most IRs had written policies and procedures
and the majority of them had been submitted to an
