Integrating Information Resources: The tale of the Primo development partnership at Vanderbilt

Vanderbilt formed a partnership with Ex Libris to develop Primo, a new search and retrieval platform for library and university resources. The session will cover:
· Project management experience from the project
· Working with staff that are not confident with the product
· Shifting time schedules
· What happens when the deliverables change
· Staffing needs
· Technical development
· How to integrate two different ILS philosophies
· Creating a GetIT system
· Extracting information

Speakers: Dale Poulter and Jody Combs, Vanderbilt University

Saturday 10:50am; about 50 in audience


Explanation of link between strategic plan & partnership was used to develop a good support base

– Users were searching for articles in opac – i.e. looking in the wrong place – documented by log files
– Vanderbilt library also manages blackboard system for campus
– In terms of staff – you can work with the “constructively skeptical” but not the “terminally skeptical”

– Amazon data was used for enhancement metadata
– Had an integrated comment feature in their preview version
– Created a test pilot web site, put links on opac and other library web page
– Combining all the holdings records into a single MARC record for processing – one of the most difficult things to do.

Alpha Search Jean & Alexander Heard Library – Demo:
– Sign-in is also remote access authentication.
– Can save search sets, send to Endnote, tag, output to
– E-shelf (my book shelf feature)
– Spell check like google – suggestions “do you mean” – but only recommends if there is something that matches in their database
– “Facets” = new jargon for search limit
– Suggested new searches functionality (“more like this”)
– When entering comments have to click on an agreement to make comment public and to be used by Vanderbilt

Speaker comments:
* Goal: once user is at the item can get it as quickly as possible, or provide them with alternatives to request item
* “took a great deal of work” to get the authority checking/collapsing function into the product
* Can define what group can access what – like circ groups – called “scope”
* Primo has multiple access level account functionality
* “total cost of ownership is very low”
* Can change rank to control what the default search is.
* Per tab and per view default view settings can be modified
* Don’t have universal one-login works everywhere system yet.
* Version 4 is faster than 3.3 – federated searching will always be slower than single interface search

Questions from audience:

* What’s the role for native interfaces after Primo is rolled out?
No. They plan to go with Primo as *the* interface
Then later said, yes we would link out to the native interface and they would stay available to link out to.

* Will librarians be teaching native interfaces and well as Primo?

* Have you thought about pipes going directly to databases instead of Metalib?
Depends on the vendor, might be in future development
Vendor might create a Primo-compatible index at the database side to incorporate into the Primo federated search. Very complex option.
But the vendor would keep control of their data. Might be better to always have a separate tab for separate searching.

* Do you need to have an OPAC underneath it?
Yes. The ILS data or digital repository data needs to be maintained in the background.
To clarify: you wouldn’t NEED to run the local opac, but they will do that because some people still prefer it.

* Primo is built on Lucene – why use Primo instead of Lucene, which is “free”?
Partnering with very large institutions which have the resources to create a working product in a short period of time.
Wanted to ultimately achieve a very user-oriented product.

* Have you done much customization of Ex Libris products, Primo ramifications?
They haven’t made a lot of changes and have not yet broken anything.

* Scenarios where user has to bounce out to the native OPAC? Does it confuse users?
Demo of Primo search, use “get it” button – brings up their opac record – right now then click on ILL and have to link again – but they are working on making it all seamless, one login and all processes in Primo.
If a given function can be offered in all data sources, it would make sense to push that to the Primo interface.
They haven’t had any confused user feedback.

* Why did they choose default search as combo OPAC and TV News?
They wanted to demonstrate that you can search all local data sources at once.

It’s Up and Running, Now What? Strategies for Building Content in an Institutional Repository

Speaker: Catherine M. Jannik, Georgia Institute of Technology Library and Information Center
Program blurb: In August 2004, Georgia Tech Library launched SMARTech with approximately 2,500 legacy items. In the beginning, we focused on authors self-archiving pre-prints and postprints, research and technical reports, and electronic theses and dissertations. As interest in archiving other materials increased and we realized that our faculty was not properly motivated to submit their own work, we changed our approach to collecting materials for our institutional repository and added a dark archive for strictly archival material. We have launched an electronic publishing service, Epage @ Tech, to support the creation and capture of e-journals, conferences, and lecture series to facilitate scholarly communication. As we provide these tools to faculty to accomplish their goals and they in turn become more aware of the need for repositories, we are more likely to convince them to deposit their personal materials. We will discuss:
· The technical support and training we provide departments to digitize and submit their own materials
· How we partner with departments to capture materials using their current electronic workflows
· How we provide production services to support e-journals, conferences, and the capture of lecture series, symposia, and the like
· Planned services to introduce these services into individual faculty members’ workflows
Session had about 37 in audience

Technology easy, it’s the social conditions and psychology of the institution
By the 2nd semester of ETD grad school was sold and wanted to make ETD required
Were able to start with “legacy materials”
Pre and post prints very difficult to get from faculty
Self submission never happened – library went to departments to get their stuff
Efforts to get self submissions had the effect of marketing the service to the campus
Alumni association “mines” the student yearbook They got their alumni association to do the scanning and PDF production
Student newspaper gives them their online issues in e-format
Download stats come from the “bitstream view” count.
Need to educate users to use the item handle not the download link
“We never asked people to catalog the books themselves that they requested” – so needed to develop a service layer for IR

Aiding depts in capturing things that now disappear – lectures, presentation of thesis equivalent performances
Offering server space as a library service to the CAMPUS.
Taking grant files when the agency requires that the product must be kept forever (sustainability). NSF grant applicants at Georgia Tech asked for letter from library stating they can make their research available forever.
Publish articles using open journal system software OJS from Public Knowledge Systems

Inserting themselves into departmental communication flow – like announcement systems, etc.
Research funded reports that are marked “final” the Office of Sponsored Programs will transmit the report & metadata (in the “coming soon we hope” category)
Presentations may have already gotten permissions for public access, or you can try and get it after the fact, or make it restricted to campus only
Do innovative things like taping students as they install an exhibit to go with exhibit photos or other materials

How do users respond to the epageTech service offering?
They don’t know what the library is taking about. Faculty are concerned about getting tenure credit for depositing with IR or creating an open source journal.
They like it when they understand it after they’ve had enough of their time to explain it.

Staffing levels?
I person worked with Systems, Archives, etc. IR architect joined 6 monhths, then a web designer, every 6 months got a new person, as of July 1st – Digital Initiatives is 6 people but not organized that way. Library has been reorganized.

What about loading scientific data?
Not yet.

Statistics on self submission vs. library submisstion?
Low self-sub, library input is growing

Multimedia materials – does metatdata describe the whole thing or the parts?
Links can be created among different parts, but most described as a whole.

Is multimedia streaming – is it fully encapsulated or coming from another server?
Streaming video is located on IT services – launches an external application.

How do you handle removal of item?
“We don’t”. If anyone asks we say no. Moved to the dark archive. If it is illegal or unethical they will remove it.

When you get datasets will you still be using dSpace.
Probably not, might use Fedora, can’t predict