OAI4: To CERN and Back Again
I was fortunate enough to be able to attend the OAI4 conference at CERN in Geneva, Switzerland between October 20 and 22, 2005. This text outlines what I heard and learned there.
Day 1, Thursday
I began by participating in a pre-conference workshop entitled "Advocacy and IPR" facilitated by Morag Greig (Glasgow University). Greig gave an overview of how they implemented their institutional repository (IR). In fact, they created two. One (eprints.gla.ac.uk) is an eprint service. The other (dspace.gla.ac.uk) contains working papers, technical reports, theses and pre-prints. The first is intended to be more scholarly. The second is intended to more like "gray literature" and not necessarily peer-reviewed. Through the workshop quite a number of tactics used to advocate the implementation of institutional repositories were articulated:
- Use what ever marketing mechanisms you can think of to promote the IR: mailing lists, presentations, newsletters, one-on-one targeted approaches, flyers, etc.
- Employ bottom-up as well as top-down communication strategies.
- Provide valued-added services against the IR to encourage author participation. Dealing with copyright issues can be one of the more persuasive services a library can provide. Others include citation analysis, bibliography generation, hit counts, and visibility in Google.
- Provide the means for the library to do the data-entry; don't give the authors additional work. "If you build it, then they don't necessarily come."
- Emphasize that open access publishing is not in opposition to traditional scholarly communication. Instead, open access publishing augments, supplements, and provides additional avenues for scholarly communication.
Herbert Van de Sompel (Los Alamos National Laboratory) opened up the main session with "What's new from the OAI?" In a nutshell, Van de Sompel reviewed the success of the OAI-PMH protocol, and he advocated the protocol be used to harvest not only meta-data, but data itself. The current state of the protocol, with its required use of Dublin Core elements, makes distributing the data itself difficult. Identifiers are not always pointers to the data but splash screens. Similarly, date elements are poorly defined. Instead, Van de Sompel advocated the use of richer XML schemes be used in the transmission of records. Such schemes could include MPEG-21, DIDL, or METS. Van de Sompel also advocated the greater implementation of mod_oai, an HTTP add-on for Apache Web servers which turns the entire server into an OAI-PMH data repository.
Stu Weibel (OCLC, Inc.) shared some of this idea regarding unique resource identifiers (URI's). Unlike URL's, URI's are/were intended to be unambiguous and persistent pointers to information resources. He outlined a number of issues surrounding URI's and they included: global uniqueness, authority, reliability, appropriate functionality, and persistence. He also illustrated a stack diagram depicting layers questions needing to be address when creating URI's:
In summary, he advocated URI's be not necessarily URL's but have characteristics of OpenURL's. He says, "Be ware of semantics, adopt accepted standards, HTTP has its limitations, and commitment is the long-term key."
Van de Sompel took the floor again and this time elaborated on the aDORe repository model he was advocating. The whole thing reminded me of Ockham where we use open protocols to harvest content centrally and the provide sets of services against the content. "We advocate the implementation of digital libraries using protocols as opposed to specific applications" -- Ockham-like.
Eric Lease Morgan (University of Notre Dame) described how us Ockham-ites used various open source tools and "light-weight" protocols to create MyLibrary@Ockham :
This article (presentation) describes the design and implementation of two digital library collections and services using a number of "light-weight" protocols and open source tools. These protocols and tools include OAI-PMH (Open Archives Initiative-Protocol for Metadata Harvesting), SRU (Search/Retrieve via URL), Perl, MyLibrary, Swish-e, Plucene, ASPELL, and WordNet. More specifically, we describe how these protocols and tools are employed in the Ockham Alerting service and MyLibrary@Ockham. The services are illustrative examples of how the library community can actively contribute to the scholarly communications process by systematically and programmatically collecting, organizing, archiving, and disseminating information freely available on the Internet. Using the same techniques described here, other libraries could expose their own particular content for their specific needs and audiences.
Weibel then stood in for Jeff Young (OCLC, Inc.) and outlined how Young had combined wiki technology with Open Worldcat to create a Review This Book service in the catalog.
John Bollen (Los Alamos National Laboratory) was studying with Van de Sompel at LANL, and he was spending his time assessing the impact of scholarly publications. He compared and contrasted various types of scholarly impact based on rankings from ISI and statistical analysis of various log files of open access content. If I understand Bollen's conclusions correctly, the ISI impact factors for particular journals match rather closely with the results of his statistical analysis. I believe Bollen's presentation has affected my thinking regarding the implementation of the University Libraries of Notre Dame institutional repository. If I can demonstrate to authors that their impact factor increases through the use of open access publishing techniques, then I think I will have an easier time convincing authors to contribute.
Tim Brody (University of Southampton) advocated the use of OpenURL's to the output of OAI responses to improve retrieval of described records as well as to facilitate the implementation of additional services against the content. He alluded to "user context" a number of times. This phrase was mentioned at least a couple of times during the conference. Shades of MyLibrary.
Day 2, Friday
Wilma Mossink (SURF Foundation) presented on copyright issues. There was not a lot of new information here. The most interesting description was "copyright as a package of balances between the author and the user." Mossink went on to outline the Copyright Toolkit.
Alma Swan (Key Perspectives, Ltd.) was a consultant who provided a very good overview of the open access movement. Again, there was not a lot of new news here for me, but the presentation was easy to follow and would be very apropos for the people who did not know the history of open access. This was the first time I had heard about OpenDOAR in public.
Andrea Powell (CABI Publishing) described the perceived role of secondary publishers (a.k.a. indexers) in an open access environment. She described the challenges of indexing serial content that is not as regular as traditional print publishing. Content moves. It comes out irregularly. When is it complete and "good enough" to index? Etc.
I then attended another workshop. This one was called "Our authors are central" and it outlined steps in the creation of an institutional repository. True to the title, it advocated user-centered design in the creation of repositories. Save their time. Create publication lists for them. Make their content more visible. Do not describe the repository as a solution to the librarian's serials pricing crisis problem because that is a non-issue for authors. It was in this workshop where I first articulated for myself, "institutional repositories are not replacements but a supplement to scholarly communication and ArXiv is a good example."
Jennifer De Beer (Stellenbosch University) described the open access publishing efforts taking place in South Africa.
Bill Hubbard (SHERPA) described OpenDOAR as a directory of open access institutional repositories. It is analogous to DOAJ, the Directory of Open Access Journals. He described OpenDOAR as a tool for many types of users: administrators, funders, IR managers, service providers, open access advocates and stakeholders. He also described it as a directory, a registry, a bridge, and a resource.
Day 3, Saturday
On Saturday Martin Halbert and I took a day trip to the Chateau Chillon , a castle at the other end of Lake Geneva. Unlike other castles I've visited, Chillon seems to have started out more like a toll gate as opposed to a military outpost or town center. Situated between the lake and a mountain, people traveling on pilgrimages were forced to pay a fee or travel around the mountain. Later the castle was used as a military base while soldiers patrolled the surrounding areas. Again, unlike other castles I've visited, Chillon has more or less (mostly more) been restored. It is not an empty shell of a building. Getting there and back was easy because of the rail system. Martin and I lucked out because the weather for our trip was the most favorable of all the days we were in the area.
CERN is an interesting place. Lot's of physicists walking around. If you walk fifty yards east you are in France. The conference was stimulating. It confirmed much of what I had already known. It provided an opportunity to reinforce existing knowledge and articulate current thinking. I strengthened a few relationships and established others. Most importantly, I learned about some of the challenges of creating and maintaining institutional repositories. The issues are not necessarily technical but rather social, legal, and political. I sincerely believe open access publishing through things like institutional repositories can supplement and enhance the scholarly communications process. The goal is not to remove traditional print publishing, but to increase the sphere of knowledge in the most effective means possible.
Geneva was beautiful.
Creator: Eric Lease Morgan <email@example.com>
Source: This text was never formally published.
Date created: 2005-12-01
Date updated: 2005-12-01
Subject(s): CERN; Geneva, Switzerland; OAI (Open Archives Initiative); travel log;