I hope when I have been asked about my assessment of electronic info
sources in pre-award I have been very clear about why IRIS has worked
well *for our institution,* and haven't come across as biased. To be
sure that is the case, let me reiterate:

Colorado State University is a land-grant Carnegie Class I Research
Institution, about $118M research expenditures per year, ranked between
75th and 60th in the country by NSF, depending on what they're
counting.  The other 2 schools in our system, for which we are the
flagship, are in the far western and southern corners of the state.
Enrollment at our campus is about 19,000.  Academic faculty on our
campus number nearly 1,000; another 1,000 Administrative-Professionals
can apply for projects.  Emphases in engineering, veterinary,
biomedical, environmental, and agricultural.  We distinguish between
contractual and donative, with separate offices;  we assign essentially
one professional position and one staff position to "research
initiation"--brainstorming with faculty, maintaining about 3,000
programs in-house, providing applications, and basic grantsmanship
training.  Budget guidelines for research initiation are "do it for
less."  Sponsored Programs enjoys an excellent reputation on campus as
being an office that takes time, has a personal touch, is responsive,
has current information, and can deliver it timely.  We want to keep it
that way.

In 1987 when I began looking at databases, IRIS and SPIN were the only
2 logical choices.  DIALOG's was just an electronic version of Oryx
Press' printed "Grants," not updated, with an expensive access, and
sorely outdated.  SPIN was very unstable administratively.  Cost-wise,
IRIS was the best deal for us--4 "outlets" on 3 campuses for $1700 (and
the price has gone down over the last several years) for an unlimited
number of searches, plus a very low CPU-based cost for each search
(rarely much more than $0.60).  We quickly migrated to Internet access,
which cut our connect charges to zero and gave us the option of
eliminating printing/mailing expenses as well.  We now run the entire
system (faculty and graduate students, 3 campuses and now 5 "outlets")
for annual expenses of something like $2500-$2800.

IRIS entries match up well with our faculty's interests.  The only
fault I've really had is that I would prefer more specificity for
NIH--maybe the PAs from the NIH Guide rather than just the program name
and each funding mechanism within it. IRIS used to include very few
foundations, which suited us fine because those were considered
donative and not ours to deal with.  Far more foundations have been
included in recent years, but we too have matured in how we interact
with our development office and now share IRIS results with them,
thereby helping them.  The accuracy of the listings has been
impeccable, and I'm a stickler for that.  The timeliness of the
updating has been very good. The University of Illinois system is
designed by, administered by, and operated for a university, not a
private company that may undergo mercurial changes.  Their interests in
maintaining a stable system parallel ours [and probably likewise, their
budgetary constraints prevent them from quickly implementing changes
I'd like to see, like including Boolean logic, or a mechanism for
screening programs on-line.]  The thesaurus structure is understandable
to faculty.  We have on-line access to the CBD and the Federal Register
through IRIS at only the cost of the CPU time.

We prefer the mediated approach to obtaining searches:  I get to know
faculty better, their research interests, can advise on campus policies
and schedules, link them better with other faculty, distinguish between
what we handle and the development office handles, and generally give
much more personal attention than if a system were available
unmediated.  We think we get much more value from our searches that
way, and apparently most faculty agree.

Electronic access to info is kind of a quick moving field, and I've
seen some "used car dealer" techniques used.  I think we'll see more of
that, as federal documents become available electronically, agencies
become more interested in getting their info out, publishers see the
advantage of electronic "publications." When I'm offered a cut rate, I
always wonder if someone else got a better deal, how long the rate will
last, why it is necessary to cut a deal if the product speaks for
itself. [And no, I don't own a Saturn.] I have talked with InfoEd about
SPIN, but have felt uncomfortable with their marketing tactics. Several
of us have talked about doing one search on both IRIS and SPIN, and
although I have done the IRIS parts several times, I've never received
the comparable SPIN results.

IRIS over the years has served us very well, is extremely stable and
reliable, targets sources well in our key areas, continues to make
enhancements, suits *our* way of helping faculty, and has kept our
search costs very low. That's not to say I don't consider other systems
coming along--I'd really like to see one which could be ported to a
campus weekly to avoid CPU charges--but in 5-6 years I've seen no
reason to switch from IRIS for this university system.

................................................................
Celia S. Walker
Assistant to VP Research      INTERNET:xxxxxx@vines.colostate.edu
Colorado State University     TEL:303/491-6355   DESK:303/491-7784
Ft. Collins, CO  80523        FAX:303/491-6147