4 ms·

Technically, how is it even possible to verify the authenticity and completeness of leaked emails unless by the owner or the hoster?

by stef25 8y ago

DoctorOetker 8y ago
this is a good question, see rester's comment: https://news.ycombinator.com/item?id=18466194 https://news.ycombinator.com/item?id=18466194 DKIM DomainKey Identified Mail, obviously diplomatic communications must prevent spoofing (!), so those mails are signed anyway, ensuring provenance of the emails
- stef25 8y ago
  Thanks for the link. Aside from provenance, can it be proven that the contents of the leaked emails is valid and untampered with? Like one would use hashes for file verification. I have to admit I have no clue as to what is the format of "leaked emails".
  - DoctorOetker 8y ago
    signatures are evaluated over the hash of the contents protecting integrity, the associated public key proves the provenance... I do remember some of the diplomatic mails being scrubbed of identifying information of individuals (say names). This would have a low entropy, so it would seem easy to download the archives from WL, take an email with a singlle censored name, then bruteforce over alphabetic characters, and verify the hash signature. Unless WL has taken the precaution to censor at least x unpredictable bits (i.e. a single character in a name or word does NOT equate to 8 bits) where x is an infeasible bruteforce depth, in theory it should be possible to recover redactions by WL. Edit: I may be conflating the scrubbing of names with the Snowden leaks, but I believe WL has also scrubbed some individual's nammes...
  - resters 8y ago
    Yes, DKIM/DomainKeys is a hash of the content and some of the message headers of an email, which is added as an additional header. A public key for the domain name owner is used to verify the hash was generated (and thus the email sent) by the owner of the domain. Email systems can thus quickly/easily check the validity of the hash when receiving inbound messages and flag or discard messages which lack the header or which fail the test.

Technically, how is it even possible to verify the authenticity and completeness of leaked emails unless by the owner or the hoster?

by stef25 8y ago

Technically, how is it even possible to verify the authenticity and completeness of leaked emails unless by the owner or the hoster?

DoctorOetker 8y ago

this is a good question, see rester's comment: https://news.ycombinator.com/item?id=18466194 https://news.ycombinator.com/item?id=18466194 DKIM DomainKey Identified Mail, obviously diplomatic communications must prevent spoofing (!), so those mails are signed anyway, ensuring provenance of the emails

stef25 8y ago
Thanks for the link. Aside from provenance, can it be proven that the contents of the leaked emails is valid and untampered with? Like one would use hashes for file verification. I have to admit I have no clue as to what is the format of "leaked emails".
- DoctorOetker 8y ago
  signatures are evaluated over the hash of the contents protecting integrity, the associated public key proves the provenance... I do remember some of the diplomatic mails being scrubbed of identifying information of individuals (say names). This would have a low entropy, so it would seem easy to download the archives from WL, take an email with a singlle censored name, then bruteforce over alphabetic characters, and verify the hash signature. Unless WL has taken the precaution to censor at least x unpredictable bits (i.e. a single character in a name or word does NOT equate to 8 bits) where x is an infeasible bruteforce depth, in theory it should be possible to recover redactions by WL. Edit: I may be conflating the scrubbing of names with the Snowden leaks, but I believe WL has also scrubbed some individual's nammes...
- resters 8y ago
  Yes, DKIM/DomainKeys is a hash of the content and some of the message headers of an email, which is added as an additional header. A public key for the domain name owner is used to verify the hash was generated (and thus the email sent) by the owner of the domain. Email systems can thus quickly/easily check the validity of the hash when receiving inbound messages and flag or discard messages which lack the header or which fail the test.