Research on Scholarly Practices and Communication at Cornell Information Science

In the first half of this talk we will explore the high level issues of disciplinary culture and interoperability. A key enabler of cyberinfrastructure development is the ability to discover and deploy functionality that leverages commonalities amongst the practices of scientists in diverse fields, thereby allowing data sharing and other collaborative activities amongst them. However, there is good evidence from the literature that individual disciplinary cultures are deeply culturally embedded and based on the number of factors including the nature of the research, the economic value of the research products, and in some cases dysfunctional, historically-based path dependencies. Our own work examining the research practices and collaborative patterns of chemists and physicists has shown strong evidence of this. We use an innovative integration of qualitative and quantitative methods to reveal the nature of scholarly communication cultures and their structural dependencies.

The second half of the talk will zoom in on an important component of scholarly cyberinfrastructure – annotation. YUMA is a Web-based annotation framework that allows users to annotate digital media objects on the Web (audio, video, images, etc.). We will explain and demonstrate how we use YUMA to enable scholarly annotation of historic high-resolution maps and focus on two types of annotations: (i) geo-referencing maps by setting control points and (ii) semantically augmented annotations that link full-text annotations with resources on the Linked Data Web (e.g., DBPedia, Geonames). The latter enable features such as multi-lingual map search. We will also present our ongoing work in the Open Annotation Collaboration (OAC), which has the goal to provide interoperability among annotation applications. Finally we give an outlook on the further development of the underlying technologies and initial ideas how we could apply them in the scholarly publication area.

Speaker Details

Carl Lagoze is an Associate Professor in the Information Science Department at Cornell University. His early research explored Digital Libraries and focused Web-based protocols and architectures for interoperability. His early work produced with colleagues a number of fundamental digital library and web-based technologies including the Fedora digital repository architecture, Dienst, the Open Archives Initiative Protocol for Metadata Harvesting, the ABC metadata ontology, and the Open Archives Initiative Object Reuse and Exchange Model. His current work focuses on the sociotechnical aspects of information exchange and sharing in scholarly communities, with a focus on data.

Bernhard Haslhofer finished his PhD at the University of Vienna in 2008 and is a postdoctoral associate at Cornell University Information Science since 2011. His research interests are in the area of Linked Data, with a special focus on data quality and semi-automatic linkage mechanisms, and in the area of scholarship practices, where he investigates how to enhance scholarly practices with data linkage techniques. He has contributed to the W3C Linking Open Data Initiative from the very beginning (2007) and served as chair, PC member, and reviewer for several conferences and journals in this field.

Carl Lagoze and Bernhard Haslhofer
Cornell University
    • Portrait of Jeff Running

      Jeff Running