One outcome of this is that I image we could derive citation metrics for things other than scientific papers, such as specimens, sequences, etc. That is, we could quantify the "value" of these data. Or it's a quantification of their popularity - which does not always correlate with "value" (whatever that is). John Grehan