Hot new tool? ITSO CUL

You may have already seen the news flash posted to CU-LIB back in February.
On Friday, February 13th, 2004 the Integrated Tool for Selection and Ordering at Cornell University Library (ITSO CUL -- pronounced it's so cool) was implemented CUL-wide to support the purchase of new library materials.

So what is ITSO CUL? The brainchild of Scott Wicks, head of CTS Acquisitions, Bibliographic Control and Documents, and realized through the technical expertise of Adam Chandler, CTS Information Technology Librarian, and Peter Hoyt, DLIT Systems programmer, ITSO CUL is an aggregation of locally developed programs that provides brief bibliographic information for newly published titles to selectors through a web-based interface. But these programs extend beyond selection, allowing us to load most of the selected bibliographic records directly into Voyager. As selectors make their choices, the tool captures MARC cataloging and order data that takes advantage of the Voyager bulkimport software. That software in turn creates bibliographic, holdings, and purchase order records. With ITSO CUL, a large percentage of monographic firm orders can be placed with minimal intervention, freeing up staff to focus on those orders that do require the human touch.

To assist in the selection process, ITSO CUL presents bibliographic information sorted by subjects and keywords into individual selector “buckets”. Each selector is able to review records that match his/her selection profile and flag titles to order for the collection. Selectors can also pass titles on if they themselves aren’t interested but believe the item may be of value to a colleague. Fund codes and default prices are all accessible at the touch of a button.

ITSO beautiful: the Deferred Books tab of ITSO CUL

Cornell has arrangements with several major book dealers worldwide. Traditionally, the dealers sent printed forms describing newly published titles that match our selection profile criteria (e.g., in particular subjects, at certain levels of treatment, or covering a set of geographic areas). With ITSO CUL, printed forms are replaced by MARC records that can serve two purposes—selectors will be able to review most newly published titles through a single, familiar interface, while technical services will be able to reuse the records to support acquisitions and cataloging activities.

The benefits of ITSO CUL are numerous. The tool allows us to recycle bibliographic data used in the review process to support ordering and cataloging. It frees up the staff time spent sorting paper review slips to selectors while also minimizing duplications between the slips. Selectors and staff have greater control over encumbrances and fund codes, and the overall amount of paper shuffling between Acquisitions and Collection Development is greatly reduced. Best of all, it means quicker turnaround from selection to order, which in turn means less time from order to shelving.

It really is so cool.

Birth of the ITSO CUL: Scott Wicks remembers …

The origin of ITSO goes back to an initiative by Collection Development to move away from the printed 3x5 cards supplied through the LC Alerts Service. At Ross Atkinson’s (AUL for Collections) request, David Block (Head, World Area Programs, IRIS) organized a team that also included Peter Hoyt of Library Systems and Howard Brentlinger of Collection Development. Their goal: to create a mechanism that would sort the appropriate MARC records to selectors based on their existing subject and area profiles. For a price, the Library of Congress makes each week’s cataloging records available to subscribing customers (just as they had done with the printed 3x5 cards). Selectors would review records sorted specifically to them by machine and print out those titles they wanted to order, recording their funding information and any notes for the Acquisitions staff.

They were making progress, but when I heard about it, I wondered why we couldn’t reuse the MARC records. After all, they would already be there, and we could leverage our investment by using them to serve two purposes at once. The more I thought, the more I considered a much fuller service—one that would capture the selectors’ efforts, reuse the MARC records to support Voyager ordering and help reduce the costly firm order process. We could create our own embedded order records!

With something like ITSO CUL, we could cut the cost of firm orders. We had trained staff to place firm orders one title at a time, searching titles in Voyager, OCLC or RLIN, one title at a time, and adding titles to purchase orders, one title at a time. One title at a time is fine if there is a reason to examine the title, but for a large percentage of Cornell’s firm orders, a machine could tell what was wanted, where to get it, and that we didn’t already have it. There was no real need for staff to search one title at a time.

From clouds to concrete

Having a bright idea is a good thing. But how do you go from idea to product? Answer: Sit down with smart, practical, well-grounded people who can take your brilliant, virtual ideas and add a bit of reality. I had the benefit of Adam Chandler and Peter Hoyt. Without the dedication and unrelenting work of these two key players, ITSO CUL would have remained just another idea. Together, we took the concept I had outlined to determine an efficient means to build the full—fledged ITSO CUL.

Part of securing the commitment for staff resources (we already had fulltime jobs) is building alliances with the stakeholders. Collection Development wanted to replace the printed LC Alert Service. Ross Atkinson didn’t really care what technical services did with the selections, so long as they were ordered. Technical Services AUL Karen Calhoun understood the potential to recover staffing resources and what ITSO CUL might offer other institutions.

Coincidence?

Never underestimate the power of coincidence. At the time we were scrambling for the resources to build ITSO CUL, Cornell computer science professor Bill Arms was looking for projects for his Spring 2003 CS 501 class. Ross, Adam, and I submitted a proposal to use the computer science class to build the web interface for selectors. Adam was appointed the Library’s contact, overseeing the students’ work. Yoram Szekely (formerly of Collection Development) coordinated the feedback loop between students and selectors to develop a tool that was focused on the selector as user.

Another important coincidence was LC's announcement that they would no longer print the 3x5 cards that selectors had been using for decades. Suddenly, we had a drop-dead date on the horizon--December 31, 2003.

While the students were working with Yoram and Adam, I was working with Pete and Adam. The selector web-based interface was a critical part of ITSO CUL, but we still had to work out the before and after selection issues. Two critical pieces were missing: a method to take the MARC records (or create our own MARC records from vendor information) and sort them to the appropriate selector, and a method to take the selected records and import them into Voyager (assigning the fund code and vendor along the way, while also deduping against existing holdings.)

David Block’s earlier initiative had given us a head start in developing a strategy for pushing the appropriate titles to the appropriate selector(s). I was able to build on this early work by taking the Cornell Primary Collection Responsibilities (CPCR) document that assigned selection responsibility by LC classification and create a table matching LC classification to selector, something that Pete's program would understand.

Many months later, after much work from the students, Adam, Pete, selectors and myself, with many helpful suggestions along the way, the tool was launched. A final detail --a catchy name-- was supplied by Jean Pajerek of Law Technical Services.

ITSO CUL continues to grow as records from new vendors are added and we gain familiarity with all of our users' needs (both selectors and technical services staff). The cast of characters supporting the changes is lengthy, but includes Howard Brentlinger, Peter Hoyt, Adam Chandler, Scott Wicks, Jim Spear, and Pedro Arroyo.

©Cornell University, 2004