18 · SPEC Kit 292
deposit documents for authorized users (78% and
63%). Most respondents are using both methods for
deposit rather than one or the other, and many re-
spondents indicated that their deposit procedures
are still under construction. Almost all respondents
indicated that faculty may deposit their materials
in the IR and both implementers and planners al-
low the work of professional staff, students, and
support staff to be deposited, as well. Several re-
spondents also mentioned faculty sponsorship as a
means of bringing in work for students or outside
Fifty-nine percent of implementers and 73% of
planners have some method for reviewing docu-
ments for copyright compliance or other reasons.
In most cases, the review is by individuals outside
the IR unit. Only nine respondents report that IR
staff review and approve documents. These proce-
dures are not always systematic, but vary by col-
lection and type of document. Additionally, all but
six respondents require depositors to sign a deposit
Sixty-nine percent of implementers and 62% of
planners accept multiple versions of the same doc-
ument. In most cases, versioning is achieved by ap-
pending additional files to the original item; most
institutions do not permit the depositor to overwrite
an earlier version of the document. Eighty-two per-
cent of respondents indicate that IR staff or authors
may withdraw documents from the IR. Comments
suggest that most institutions permit withdrawal
only in cases of copyright infringement or other le-
gal issues. Institutions that permit the withdrawal
of documents generally leave a “tombstone”—a
reference to the withdrawn document—in the sys-
tem as a record. Many respondents felt that policy
in this area would solidify as they learned more
about the legal landscape.
Roughly half of the respondents import metadata
into their IRs from outside sources, typically by a
process of automated mapping from a variety of
schemas into Dublin Core. Many are converting
data from local schemas and a surprising number
mentioned converting records from the MARC for-
mat. Ninety-four percent of implementers and 78%
of planners allow depositors to enter simple meta-
data; many of these same respondents also enter
metadata on behalf of depositors (implementers,
60%; planners, 56%) or enhance depositor supplied
metadata after the fact depending on the material
and source (57% and 72%).
Survey results indicate that many institutions are
taking significant steps to ensure that their IRs are
interoperable with other systems. Ninety-four per-
cent of implementers and 88% of planners indicate
that their IR supports the Open Archives Initiative
Protocol for Metadata Harvesting (OAI-PMH) and
a similarly large percentage (89% and 82%) report
using persistent identifiers for materials in the IR.
This result is consistent with the widespread use
of DSpace and other platforms to which these ser-
vices are integral. Over half of implementers (54%)
have ensured that their systems are OpenURL com-
pliant and 41% of planners anticipate having this
functionality. Sixty-three percent of implementers
have incorporated their IRs into federated search
mechanisms. Since only one planner reports doing
so, it may be that institutions still in the planning
stages have not yet considered how to incorporate
the IR with other services, or these institutions may
lack federated search tools altogether.
Although most IR platforms in widespread use
are OAI-PMH compliant, only one respondent spe-
cifically mentioned being crawled by search en-
gines. The issue of optimizing exposure to search
engines may become more significant as IRs be-
come more prevalent and stable.
Content Recruitment
The difficulties faced by institutions when recruit-
ing content for their IRs is clearly borne-out by
respondents. Only one implementer found recruit-
ment “very easy” and only seven “somewhat easy.”
Fourteen (40%) found it “somewhat difficult,” and
Previous Page Next Page