While reading an NSF "Dear Colleague" letter on data citation in the geosciences, I came across this quote from the American Geophysical Union,
..the scientific community should recognize the professional value of data activities by endorsing the concept of publication of data, to be credited and cited like the products of any other scientific activity, and encouraging peer-review of such publications.I agree completely! Now what would it take for AGU, or any other physical science society, to say:
"the scientific community should recognize the professional value of scientific programming activities by endorsing the concept of publication of code, to be credited and cited like the products of any other scientific activity, and encouraging peer-review of such publications."
The data citation movement is really a great development for geoscience. Like code, data products often have many hands involved in them, more then the number of authors on a typical climate paper. They undergo revisions and can be used and reused for years. If the community can figure out things like what a "first author" means for a data product, what the "impact factor" is for a data product and get citations of data accepted in tenure cases co-equal with other publications then its a short step to doing the same for code.
1 comment:
"Data" is of no use without metadata. That's everything from units to how collected, how was data processed, how was it quality controlled, what's its provenance, etc. This has come to be called "information". This is well documented in the Open Archival Information System (OAIS).
I'm not aware of a similar reference model for software. Until such exists there's not much use for citations. Languages have different versions, change over time, or even disappear. Operating systems, hardware, build instructions, dependencies, etc, must all be identified in a consistent manner for the citation to be of any scholastic use.
Post a Comment