PID setup for personal data #14

Open
opened 2025-04-21 15:04:37 +00:00 by mih · 1 comment
Owner

We should not be using the same PIDs for person(al) data for internal and public records. Doing so would make it easier to deduce information from internal records, even if the actual person records are inaccessible (because a public record with the same PID would reveal other public information, such as the author name from a publication record).

We are considering to use a truncated hash of a person's name and their starting date as the (internal) PID. Those could be prefix like inm7:person/<hash>. For any to-be-exported record, we could generate a public identifier from this information by adding a "salt" and rehashing it.

This could also be just a fallback, and we could, by default, use an ORCID, if that is on record for a person.

We should not be using the same PIDs for person(al) data for internal and public records. Doing so would make it easier to deduce information from internal records, even if the actual person records are inaccessible (because a public record with the same PID would reveal other public information, such as the author name from a publication record). We are considering to use a truncated hash of a person's name and their starting date as the (internal) PID. Those could be prefix like `inm7:person/<hash>`. For any to-be-exported record, we could generate a public identifier from this information by adding a "salt" and rehashing it. This could also be just a fallback, and we could, by default, use an ORCID, if that is on record for a person.
Author
Owner

https://concepts.inm7.de/about/ now shows that filter and transformation operations are possible for metadata being exported to other scopes/models. This would be the place where arbitrary ID mappings could take place. The sibling operation for metadata intake is also included in the design, and would allows for the inverse mapping to take place.

https://concepts.inm7.de/about/ now shows that filter and transformation operations are possible for metadata being exported to other scopes/models. This would be the place where arbitrary ID mappings could take place. The sibling operation for metadata intake is also included in the design, and would allows for the inverse mapping to take place.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
inm7/inm7-concepts#14
No description provided.