Human & Machine Actionable Data Citations


Earlier this year the Joint Declaration for Data Citation Principles was released.  The preamble of the declaration states "These principles recognize the dual necessity of creating citation practices that are both human understandable and machine-actionable."

A group was formed to discuss the issues regarding implementing these principles and currently has a paper in process, "Achieving human and machine accessibility of cited data in scholarly publications" with a proposal of how to achieve the goal in the preamble.

We will present an short overview of both the data citation principles and our proposal and solicit comments on the paper during its review period.


Creative Common License: 
Creative Commons Attribution 3.0 License

Name: Ruth Duerr
Organization(s): The Ronin Institute

Name: Robert Downs
Organization(s): CIESIN


As for the correction, the table decribes content-negotiation per the HTTP/1.1 spec, not the RFC that I cited. The experimental RFC has a provision for 'Accept-Features', which could be used for tagging by schema (although I haven't managed to find examples of it being used in the wild. Also, browsers can request the alternatives by sending 'Negotiate: vlist' . The next draft has a more detailed appendix. Also, some people are using XSLT to serve landing pages. One of the co-authors of the paper mentioned that web browsers were starting to remove XSLT support, so we specifically did not mention it. (I'm seeing if I can get that edit made before the paper is re-submitted)