search

nlesc-sigs / data-sig

Linked data, data & modeling SIG
Other
5 stars 3 forks source link

Sensitive data handling #44

Closed arnikz closed 5 years ago

arnikz commented 5 years ago

Discuss NLeSC projects that require a.o. data anonymization.

eriktks commented 5 years ago

Background information: At this moment I am collecting information about the usage of privacy-sensitive data in eScience projects (current, past and future) to find out what our best practices are in this area. I will present a summary of these findings at our data sig and at a national task group on data anonymization I joined recently: https://www.edugroepen.nl/sites/RDM_platform/LCRDM/Wikipages/Taakgroep%20Anonimisering.aspx

I have sent the following questions to engineers involved in projects that might deal with privacy-sensitive data:

  1. Did your project involve privacy-sensitive data like person names?
  2. Answer only if the answer to question 1 was yes: Did our escience engineers have access to these privacy-sensitive data?
  3. Answer only if the answer to question 2 was yes: Were the data preprocessed to remove privacy-sensitive information, for example by data anonymisation?
  4. Answer only if the answer to question 3 was yes: How were the data processed to remove privacy-sensitive information?

If you think the answer to question one for any of your projects would be yes then I am interested in your experiences.

jiskattema commented 5 years ago

Hi Erik, good to see you working on this! Wrt. to the AVG (GDPR) we already did a similar inventory (looking only at point 1 and 2) some time ago; but projects probably have progressed (or new ones started) so good to repeat it. Can you keep me informed of the outcome? In any case, engineers working with (possibly) personal identifiable data should come talk to our contact person for privacy (me at the moment).

arnikz commented 5 years ago

@jiskattema you are welcome to join and to share your experience and the "inventory" as well.

c-martinez commented 5 years ago

There is a report from this issue -- thanks @eriktks !