Integrated Access Council
May 16, 2005
SML Room 409
Present: Dale Askey, Katie Bauer, Matthew Beacom, Meg Bellinger, Ann Green, Katherine Haskins, Julie Linden, Fred Martz, Kim Parker, Bobbie Pilette, Kalee Sprague, David Stern, Rich Szary
Absent: Dan Chudnov, Carol DeNatale, Kenny Marone, Jack Meyers, Tom Saul, Joan Swanekamp, Frank Turner
Guests: David Gewirtz, Audrey Novak
Meg started the meeting by announcing that Fred will be retiring from his position as Head of Integrated Library Technology Services in August. He has worked as a manager in the Library for 35 years. Fred said it is time for him to pursue a different type of work and begin a new chapter in his career, perhaps in a different field but possibly still in the Library. He has been systems Department Head for 10 years, and it has been a rich experience. He mentioned that 40 years ago this July he first worked in this very Library as a student book mover. He has a lot of stories and history to tell. The Library once was a place where people took refuge from technology – a remarkable contrast to the present environment.
1. Report from the Digital Preservation Committee – Bobbie, Audrey, Ann
Bobbie began by saying that last Friday she sent the draft digital preservation policy to all IAC members. The policy is a high level review of preservation goals. The Digital Preservation Committee would like feedback from the IAC about whether or not they believe the policy is headed in the right direction at this point. Are there any sections that the committee missed? If so bring it their attention. Once the high level policy is finished they will be working on a summary of best practices. She said the committee worked with CoDGeR on policies and also is coordinating work with the Metadata Committee. They are scheduled for a SCOPA forum in September.
Kim asked if we need to say in this document what procedures need to be in place to support the policy or point to places where procedures are articulated. Ann said the committee talked about where policies stop and procedures and best practices begin and the committee intends to make references to the specifics. Meg asked about the road map and time frame for completion. Bobbie replied that they plan to have a final draft of the high level policy by the September SCOPA forum and seek feedback from across the Library at that time. This policy document is meant for the libraries, but anyone outside could also be informed by and follow the recommendations in the policy. Ann said the underlying principles can apply to multiple units on campus. The best practices and procedures documents will be outlined by that time, but noted that the committee has a 3-year appointment and the bulk of the work is yet to come.
Kim asked if the committee has considered the difference between feedback and buying into the document. There needs to be a time when no more feedback can be received and the document is presented as finished. Meg said we need to develop a communication plan between all of the committees.
Ann said that the website has the notes from all of the meetings and a more extensive bibliography. Bobbie asked IAC members to send suggestions by June 1st. Audrey asked if anyone needs to see this beyond IAC. Rich suggested it be presented to ITS. David said CDC needs to review also. Bobbie said that David Gewirtz from ITS is also part of the committee. Meg also suggested that the future committee on Information Architecture would also need to review the policy.
2. Overview of the Proposed VITAL/Fedora Trial – David Gewirtz
David G. presented a possible partnership between the Library and VTLS, the vendor of the product VITAL. VTLS would help us develop and test an infrastructure based on the open source repository software Fedora. Fedora gives us the ability to build information management systems such as digital collections, digital libraries, rescue repository, or digital archive (http://www.fedora.info/).
He pointed out that vendor-supplied systems are very expensive, the offerings are proprietary solutions, commercial systems are not designed for the academy, and customization of commercial products is expensive and may not yield the desired results. But there is a possible solution by using open source software such as Linux and Apache. Fedora is a good example of an open source product that meets academic repository needs. Other examples can be found at http://sourceforge.net/. By partnering in developing open source software with a commercial vendor that is committed to its use, the risks and resources are significantly reduced.
In another open source context, David noted that at the Fedora users conference there was discussion of using Fedora in the classroom with Sakai. Although this capability is not yet in place, there is significant interest. David also discussed the importance of standardized infrastructure which reduces costs over time. The transport, storage, and preservation layers need to be standardized to achieve optimal cost savings and integration.
David then talked about a 60-day trial period for VTLS/VITAL to assess the level of commitment by the vendor, the quality of support, and their technical extensions to the Fedora product. Then a decision can be made about developing a longer term relationship with VTLS. Kim asked what content would be used and what work would be done during this 60-day period. In regard to the work plan, David said they would load content into VITAL and test-drive the client manager and ingest process to determine what added value the vendor brings over Fedora’s basic client manager. Kim asked if we would test different metadata sets, and David replied we could but right now it will be limited to Dublin Core since the test is focused on the partnership with the vendor not with specific metadata implementation. David Stern asked if instead of putting efforts into a trial, could we learn from others who have already done this. David G replied that we need first-hand experience with the vendor to know the full story. Fred commented that this also is an opportunity for an evaluation of the software from a user interface perspective. Also this is an opportunity for PIC and other stakeholders to look at this solution and learn what is offered.
Ann asked how we are going to separate the value of Fedora and VITAL. David G said his knowledge of Fedora is extensive and that it will be possible to compare features provided by VTLS with those available in the native open source product. Fred asked if the value added from VTLS is enough to justify working with them. David G said we might want to engage a vendor in order to reduce the burden on Yale staff resources. Kim asked if others know Fedora well enough. Aren’t we assessing Fedora at the same time as VITAL? David said this is a way for us to think about common infrastructure.
David Stern commented that it seems as if VITAL doesn’t have a huge following. However, David G feels that the number of people interested in the development of this product is large enough and emphasized that we will be assessing the product and vendor as well. Julie asked what components would be evaluated and if we would be looking at this as a digital archive as well as digital infrastructure. David said we are trying to do both; tools for a common infrastructure that encompasses preservation need to be explored.
Fred said the capability of building a preservation archive is not yet done within Fedora. We need a service to bring the preservation archive into Fedora. This is years off though. Meg asked if there were any strong reservations to doing the 60-day trial. David S asked what efforts it would take. Kim asked if we were really prepared to take it for a trial run. Meg said that many feel this is potentially the best solution and the trial is a minimal investment in exploring that potential. The Finding Aids and the Yale Daily News will be content used in the pilot phase. It would also be highly desirable to include marked up text (TEI) and a sample image collection.
3. Discussion of activities reported by the Public Interfaces Committee’s forum on Measuring User Behavior in the Online World – Katie Bauer
Katie distributed a document from a PIC forum on evidence-based decision making about web interfaces and remote Library users. She said that we do have a number of tools in place that let us know about user behavior and preferences. The first report lists the top 30 web pages as collected by the Web Trends reports. These are collected from the Web server log files. They won’t capture any log history from external websites, only what lives on Library web servers. (There is nothing from Law or Medical). Kim asked if we surveyed areas with clusters to see what their homepage is and how that might affect the average time viewed and number of visits. Katie said they assumed that most would use the Library’s homepage. Katie did point out that the high average time viewed could be due to idle time, and this could be filtered out of the reports. She pointed out that Divinity crept into the top 3 recently.
There was also a report that listed the top 4 databases that were accessed (Academic Search, WorldCat, Web of Science, and Lexis Nexis). It covers users that go through the DBOW page (not those who have them bookmarked). Katie noted that all of these reports only show a part of the picture but have been useful in rethinking the front door and other web services. The 3rd page is another study of links from the Library front door. It shows that the quick links are not getting much use. Also included are Orbis stats that Kalee has been running for the past year. Z39.50 is at the top because of external harvesters. Fred said we are listed on many other library sites as a reference. Katie said that they looked at searches returning 0 hits and many were from people who didn’t know that they had to use the “and” operator, so changes were made in the search interface to make ‘and’ implied.
The last report summarized the Google Logs from the library web site. Katie said that this showed that people don’t really understand what part of the web site they are searching, and that subject searches were about 46% of the logs. So PIC developed a new "Search" page that allows users to search not only the library Web site, but also Orbis, databases & online journal titles, the staff directory, and "Google the Web" from a single page. At this point, only one of those options can be searched at a time. The ultimate goal is for a single, federated search of all those resources; Karen Reardon and Kalee Sprague are working on the technical issues. This covers some of the things that they have been looking at in the last year. In the future they will look at more subjective measures regarding what people like and don’t like about web services. Meg noted that funds are available for web development and assessment work.
Ann mentioned that the June meeting will be extended to 1-1/2 hours. The Meeting Maker proposal has been changed. Any comments on the April meeting minutes should be forwarded to her.