Wednesday, March 2nd, 2011

List of programs:


Assessment of Electronic Resources

Hana Levay: Information Resources Librarian, University of Washington,
Wednesday, March 2, 2011 — 8:30 – 9:30 am Salon DE
reported by Jane Smith

Hana Levay described the assessment of licensed journal packages and databases at the University of Washington, presenting a method that goes beyond usage statistics and cost data. First, she outlined the process “gathering the data, assembling the data, and analyzing the data” and then she explained the steps involved in each part. The data used in the assessment included usage reports, bibliometrics, title lists, cost, and package and license restrictions.

To collect usage reports, Ms. Levay used two methods: manual collection from vendor sites as well as ScholarlyStats, a collection service. She noted that the choice of method essentially came down to time vs. money; the collection service consolidated the usage data for her and included additional services while the manual collection allowed for greater control of data but was time-consuming. Bibliometrics data included Journal Citation Reports from ISI Thomson, Eigenfactor, developed at the University of Washington, custom metrics developed by the California Digital Library and local weighting of journals by UW subject selectors. Ms. Levay also included title lists (both subscribed and unsubscribed) in a licensed package, cost data, and whether or not the package provided archival or perpetual access.

To assemble the data, Ms. Levay created a Serials Decision database using Microsoft Access, though Microsoft Excel could have been used as well. She then analyzed the serials data by fund and by package. Analyzing the serials by fund allowed the selectors to see usage, cost and perpetual access for all titles in their funds. Analyzing the serials by package allowed for the comparison of journal packages. She was also able to assess the value of subscribing to a package as opposed to subscribing to individual titles.

To assess databases, she used two spreadsheets; one for usage data over time, and one with costs and notes.

She noted a number of limitations to assessment: most licensed title packages limit or prohibit individual title cancellations, usage of a title varies between disciplines, and the bibliometrics tools used cover the sciences and social sciences, but do not offer much coverage of other disciplines.

Ms. Levay was able to draw several conclusions from her study: most of the unsubscribed titles in a licensed journal package are of little value, assessment of electronic resources is time and labor intensive, and librarians need products that assess more than just usage. As a result of the study, the University of Washington library cancelled the Springer licensed title package and subscribed to individual Springer titles instead. They also cancelled individual titles within packages where possible.


Buzz Session: Security issues with online products

Wednesday, March 2, 9:40-10:25, Salon AB
reported by Gina Bastone

When many librarians think of unauthorized downloading, they picture an authorized user, often an undergraduate student, sharing his or her password with friends outside of the university. However, the biggest infractions come from hackers far away from the school. IP addresses often reveal hackers in other countries like China, India, or Brazil. Even then, hackers may use a server in one location while the end use may occur in a completely different location. Hackers will often set up automatic downloading of large numbers of files. So what are university libraries doing? Here is a discussion from various librarians.

Don’t always jump to the conclusion that a huge number of downloads from one user is the work of a hacker. Computer science professors and students sometimes set up robotic downloads. Vendors do not like it, but it is often legitimate research. Check licensing agreements for language on robotic downloading.

International students go back to their home countries and share their usernames and passwords with colleagues and their colleagues then use systematic downloading. To prevent this, require all traffic to go through a campus proxy server, and then monitor proxy logs. Good education and library instruction also helps. And remember, it has less to do with location and more to do with the quantity of activity. IP ranges from other countries can be blocked, but this only works well in schools that do not have a large international student population.

There are many blogs and sites that share usernames and include proxy links. The IT department should monitor these sites regularly. Outreach to the IT department can help prevent problems.

Sometimes vendors can be difficult to work with when these situations arise. They may shut off all access or put in stringent requirements that limit access. Therefore, be aware of campus needs before negotiating with vendors and check the language of agreements before signing. Be clear that the institution is not legally responsible for what individual users do. Fight ridiculous demands. For example, some vendors want users to destroy all data once they leave an institution, and there is simply no way to police this. Ask for at least 30 days to clear a breach, and that the vendor will go after specific IP addresses rather than shutting off all access. If the library participates in consortial agreements, make sure that violations at other schools do not affect access at all institutions. Keep track of this sort of language and incorporate it into agreements.


Demand-Driven Acquisitions: New Tools and Strategies for Long term Management

Bryan Keane,ebrary; Robin Champieux, EBL, Michael Levine Clark, University of Denver; Matt Nauman, YBP Library Services

Wednesday, March 2, 8:30-9:30, Salon AB

[View this presentation on Vimeo]