Human language description for datasets, landing page

[katjabercic]

In addition to the data, schema, etc., we need to have a longer description of the data aimed at human readers. This description should be easily accessible from the frontend and would be part of a "dataset landing page". This description could be the same as the exposee from step one of the submission process. The description should also include a definition of the mathematical objects from the dataset aimed at humans.

The landing page should also contain links to

• a formalised appendix (pdf),
• a paper dealing with the objects (informal provenance) - in particular it should support claims of completeness in case the dataset is a complete census,
• link to the code repository.

See arXMLiv 08.2019 - An HTML5 dataset for arXiv.org for something a little similar.

[kohlhase]

And of course copyright and licensing information about the dataset. We should think about a schema how that can be formalized, so that becomes FAIR.