Paper describing a FAIR implementation

[markwilkinson]

https://peerj.com/articles/cs-110/ "Interoperability and FAIRness through a novel combination of Web technologies"

[peter-wittenburg]

This is a great article and I am sure that the ideas will find their way into practices.
This raises the issue that we need to include good examples somehow somewhere so that people can look things up. I just had to read quite a number of infrastructure appers and it is great to see that some already refer to FAIR and that some started being FAIR already years ago. I guess that the oldest infrastructure that implemented FAIR principles goes back to a start in 2000 or so which demonstrates that it is possible and efficient to be FAIR and that one can indeed build persistent and stable data infrastructures based on these principles. Perhaps the astronomers started even earlier being widely FAIR - don't know - need to ask the IVOA folks.
The biggest problem is indeed as Mark points out"diversity/heterogeneity" at all layers and in general semantic diversity has not been solved yet so that scientists are satisfied. I remember my "old" community where it was possible to agree on metadata concept definitions across sub-communities, register them in an open registry and getting them into use, but it was hardly possible to extend this to the domain of concepts that describe the scientific phenomena. People of course insist on their fine semantic nuances since this will make their science possible and they hesitate to spend time to register/define them according to a formal scheme. How to tackle this semantic field in an optimal way?

Just one side comment to Mark: you state that your science area is extremely diverse. Whatever meeting I participated in in various scientific domains, I always hear this sentence. I guess that it is time to just accept that modern science in all disciplines is DIVERSE.

Thanks for the reference.