Enhancing scholarly repositories: 1) augmenting interoparability, and 2)
automatic detection of duplicates and plagiarism.

Simeon Warner
Cornell University
Computing and Information Science
simeon@cs.cornell.edu

1) Open-access repositories could be the foundation of a global
and natively-digital scholarly communication system, yet the current
infrastructure falls far short of this goal. One missing element is the
interoperability necessary to support the many workflows and value-chains
involved in scholarly communication. What is the minimum set of features
necessary to allow heterogeneous repositories to support such a global
communication system? Can we find a common core content model to support
interoperability?

I will describe ideas and experience from the NSF Pathways project
between Cornell University and the Los Alamos National Laboratory. In
particular, I will use as an example our prototype implementation of
an overlay journal over a set heterogeneous repository architectures
(aDORe, arXiv, DSpace, Fedora).

2) The arXiv.org repository contains over 390,000 articles and will
grow by more than 10% in the next year. We would like to be able to
automatically detect both accidental duplicate submissions and plagiarism.

I will describe the algorithm and implementation of a service to
compare the full-text of each new submission against all 390k existing
submissions in real-time, based on a modification of the "winnowing"
document fingerprinting algorithm.



Simeon Warner

Simeon Warner is a Research Associate in Computing and Information Science
at Cornell University. He is one of the developers of the arXiv e-print
archive (http://arXiv.org/) and his research interests include web
information systems, interoperability, and open-access scholarly
publishing. He has been actively involved with the Open Archives
Initiative (OAI) since its inception and was one of the authors of the OAI
Protocol for Metadata Harvesting. He worked at Los Alamos National
Laboratory before moving with arXiv to Cornell in 2001. Prior to working
on arXiv, he worked in the Physics Department at Syracuse University in
computational physics, a discipline in which arXiv has eclipsed
conventional journals as the preferred means of scholarly communication.