davebrowning
commented
1 year ago (Apologies for the delayed response).
Thanks for raising this issue. This has sparked a bit of a discussion between the editors and then within the WG as a whole. The outcome of that is that we think its a potentially interesting area but we haven't got any use case that needs to communicate the existence of privacy-relevant data within a dataset , nor does any of the active WG membership have any implementation experience with such requirements. On the other hand, we can imagine how it might be useful to know that a dataset had some such information though what a user of such a catalog record would do on the basis of this is more difficult to speculate about. That might be quite domain specific and sit well within a profile, but that's very much getting ahead of our experience/requirements at this time.
What the WG has agreed is that we mark this for future work and defer discussion on privacy to see what the Data Privacy Vocabularies and Controls Community Group develops, and other efforts by other W3C groups. (The WG resolution is documented in the minutes at https://www.w3.org/2022/10/25-dxwg-minutes#r02 )
I hope this is satisfactory.
pchampin
npdoty
Communicating privacy-related restrictions regarding a dataset: beyond copyright, could there be terminology about the privacy-relevant restrictions on a dataset? E.g. this dataset has particular policies regarding who can access it, how the data can be used, that individual records can't be re-published, etc.
Indicating whether a dataset contains information about people, or particularly sensitive types of information about people. Could those indicators be used to delete data in dangerous situations? Consider the historical, but still very relevant, example of the destruction of civil registry records in Amsterdam in 1943. Could those indicators help facilitate audits of datasets or use of datasets?