Institutional Repositories and Enhanced and Alternative Metrics of Publication Impact

Abstract

Open Access is one of the most popular terms in the library and information science community. Within the community there is a wide and controversial discussion on what would be the most successful ways to bring scientific authors to provide their publications as open access publications. One way would be to encourage the use and to enhance the visibility and usage of institutional repositories and Open Access journals, whereby usage is understood in the sense of reading as well as in the sense of citing a publication. The DINI workshop in Berlin focused on 3 topics: 1) alternative metrics of impact based on usage data, 2) interoperable and standardized usage statistics and 3) Open Access citation information. To accomplish these there is a need to promote and standardize Institutional repositories as the basis of digital libraries, especially concerning the issues of data exchange formats and long term availability. To calculate alternative metrics of impact based on usage data it is clear, that a common definition of usage data is necessary, meaning for example that web crawlers and robots must not be counted in the access statistics and that the logs need to be generated by link resolvers. The log data needs to be aggregated to achieve an overall picture. It was understood that IRs should at least be OpenURL enabled. In this way linking server logs can be serialized and exchanged as OpenURL ContextObjects and exposed by on OAI PMH data provider. The repository retains full control of what data how this data is exposed and can therefore take care e.g. of anonymization of user IDs. A trusted third party can harvest and aggregate these logs from different repositories and derive page ranks or journal ranks from them. For this reason the usage statistics need to be interoperable and standardized.