EIESC Telecon

Meeting Notes
Wednesday, February 25, 2004, 1.00 pm-2.00 pm.

Notes taken by Casey Jones

1. Collections Assessment Task Force - Update (Judy Ridgway)

Way to catalogue resources in NSDL MR. Wanted to do this based on metadata, but discovered inconsistency and wide variation in provided metadata for audience, subject, resource type, and format. Task force initially developed terms for each category to be used to describe NSDL collection. Thought that maybe some crosswalking could be done.

Later, other means were investigated. Lexical Analysis and Support Vector Machines were the two methods we looked at. A grad student (Steve Bethard) helped us and suggested we were having limited success with both of these approaches. Two cataloguers were hired to catalogue 125 resources each and to check each other's work. This supervised training technique was explored.

Preliminary results were reported for holdings of collections and the items available. These were reported in the first NSDL Progress Report.

Liz Liddy has now agreed to take a high-level approach to looking at the metadata repository using clustering and extraction methods. Next steps will be having Liz explore different approaches we might take. Liz commented that this is a big job, more than what might be possible on volunteer time. Liz will respond with some recommendations. Steve's line of inquiry provided us with a great start. Tammy has suggested a possible hybrid of using metadata that exists and machine indexing. We currently have a small amount of training data. Crosswalk approach was abandoned because many of the fields were empty. One possible way forward would be to write a proposal. The proposal would need to come from someone with a computer science background. A clustering experiment would be useful input for a possible proposal.

Tammy will explore options, and Anita mentioned a possible contact at Nanyang University who might be able to provide input.

2. Annual Report - Update (Tammy Sumner, Carol Terrizzi, Casey Jones)

Laura Bartolo is the EIESC representative on the Annual Report committee.

What types of data do the committee want to provide?

Annual Report meeting tomorrow will discuss possible ways of combining Annual Meeting (Theme: Desigining Strategies for Success) with information gathered in Annual Report, process, and timeline.

Last year we provided 4 bits of information - analysis of collections (including analysis of newly funded projects), data on usage metrics, PI survey data, and NSDL representation in papers and conference participation.

Potential to highlight projects that have strategies in a particular area and projects with a component in education. We could highlight information textually as well as find something we could support with data. The conferences information is good to highlight, but the data is hard to come by. Worth trying to see if we can tap into information NSF collects. Connie Della-Piana and Gabe Della-Piana have been helpful in supporting impact of NSDL - they might be able to give us data or access of data.

Mimi volunteered to represent EIESC on the first cross-committee task force for addressing impact. This might be something we can help with.

3. 2004 PI Survey - Update (Sarah Giersch)

How are the distributed library building and community governance processes working?

First survey was during the 2002 pilot study to anyone working on the projects. Results presented at 2002 Annual Meeting, and EIESC decided to undertake this on an NSDL-wide scale. The 2003 PI Survey (open to PIs only as representative of their project and project personnel) was developed and validity tested although rushed in order to include data in the first NSDL Progress Report.

One issue last year was synchronizing PI Survey and Annual Report timelines. Another issue is that we don't want to ask everyone the same questions year after year.

potential software: Survey Tracker, Survey Monkey

Can this group help to roll surveys up into one survey or at least incorporate other SC issues into our survey process? If other SCs want to be involved, we should ask for a representative from the SC.

Survey Tasks - survey development, testing, administering the survey, and analysis. Judy and Tammy have volunteered to help. Have SC reps identified by March PC meeting. Running under assumption that we want to ask the same questions. June was not a good time to administer the survey because many people were travelling at that time. Last year's foci were communication, collaboration, and coordination - do we keep those 3 focal areas? One area we may have been remiss is ensuring that the powers-that-be have been taking findings and implementing them. We might want to develop a briefing document for the SCs and PC. Goal is to have next year's at April.

4. Committee Meeting at JCDL

The afternoon of Monday June 07 will probably be our JCDL meeting day, but we should be careful not to compete with a relevant tutorial.

5. Webmetrics Task Force - Discussion (Tammy Sumner)

One of the questions the EIESC has been considering is How are people using the library? This is based primarily on the 2002 pilot study. It was difficult to provide a consistent high-level of data monthly. The Technology SC has become interested in the technology issues in answering this question. A joint task force between the EIESC and Technology SC would have EIESC providing expertise on types of information to be gathered and the Technology SC addressing the technology issues.

Robby Robson's Re-usability Workshop is also addressing the technology issues. Dr. Paul Stroop is a colleague of Tammy's working at SUN who has expertise and interest in this area.

Interested People - Bob Downs, Tammy Sumner, Mimi Recker, Jim Dorward
Telecon will happen sometime during the week of March 8.

6. Other Business

PC meeting in Chicago, end of March - Anita will be attending as Chair of EIESC.

On the Call