Skip to main content

Privacy Preserving Measurement

Document Type Approved BOF request
Title Privacy Preserving Measurement
Last updated 2021-10-07
State Approved
Editor Eric Rescorla
Responsible leadership Roman Danyliw
Send notices to (None)

Name: Privacy-Preserving Measurement


There are many situations in which it is desirable to take
measurements of data which people consider sensitive. For instance,
one might want to measure web sites that do not render properly
or propagation of some disease. In these cases, the entity taking the
measurement is not interested in people's individual responses but
rather in aggregated data (e.g., how many users had errors on site X).
Conventional methods require collecting individual measurements and then
aggregating them, thus representing a threat to user privacy and
rendering many such measurements difficult and impractical.

New cryptographic techniques such as Prio and, more recently, a
protocol for collecting data on the most-common values in a privacy-preserving
manner (the "heavy hitters" problem), address this gap by
splitting up measurements between multiple servers which can jointly
compute the aggregate value without any server learning the value
of individual responses. The Privacy Preserving Measurement (PPM) work
will standardize a protocol for deployment of these techniques on the
Internet. This protocol will include mechanisms for:

  • Client submission of individual reports, including proofs of validity.
  • Server verification of validity proofs
  • Server computation of aggregate values and collection of results to
  • Reporting of aggregate results to the entity taking the measurement

This is a WG-forming BOF for a WG to standardize a PPM protocol. The
PPM WG will not itself define cryptographic algorithms for PPM but
will instead use algorithms defined by the CFRG.

Required Details

  • Status: WG Forming
  • Responsible AD: Roman Danyliw
  • BOF proponents: Eric Rescorla <>, Chris Wood <>
  • BOF chairs: Adam Roach <> + Alissa Cooper <>
  • Number of people expected to attend: 100
  • Length of session (1 or 2 hours): 2 hours
  • Conflicts (whole Areas and/or WGs)
  • Chair Conflicts: TBD
  • Technology Overlap: Security Area, PEARG, CFRG, SFRAME, QUIC, MASQUE, TSVWG, MADINAS, OARH
  • Key Participant Conflict: Chris Wood, Eric Rescorla, Christopher Patton,
    Martin Thomson, Richard Barnes


  • Presentation of use cases [15]
  • Overview of solution draft [30]
  • Charter discussion [30]
  • BoF questions [Remainder]