[cross-project-issues-dev] Anonymisation of public data

[Boris Baldassari <boris@xxxxxxxxxxxxxx>]

Hello good people,

In the context of the Crossminer research project [1], we plan to 
publish a number of datasets to the public and for the research 
community. This includes public data from the Eclipse forge (i.e. data 
is fetched from public data sources and APIs only), and we want to setup 
an anonymisation process that would:

* Efficiently and safely remove all personally identifiable data -- we 
don't want to help spammers or malicious harvesters, and
* Still provide valuable information and datasets for the research 
community -- e.g. ability to identify identical IDs across sources 
without specifically knowing them.

The basic idea is to simply replace all identifiers with asymmetrically 
encrypted strings, so all IDs have the same ciphered result. RSA is used 
for the encryption, and the private key is thrown away once the encoding 
is done, making it impossible (according to common encryption standards) 
to retrieve the original string.

A prototype has already been published [2, 3] and we would like to ask 
people to review it so as to make sure that our privacy-preserving 
mechanism is safe.

Any feedback, concern or contribution is warmly welcome.

[1] https://www.crossminer.org/
[2] https://github.com/borisbaldassari/data-anonymiser
[3] https://borisbaldassari.github.io/data-anonymiser/

Thanks in advance, have a wonderful week!

--
boris