Paper 2024/982 Original: eprint.iacr.org/2024/982 ↗

SoK: Programmable Privacy in Distributed Systems

Daniel Benarroch
Bryan Gillespie
Ying Tong Lai
Andrew Miller
Abstract

This Systematization of Knowledge conducts a survey of contemporary distributed blockchain protocols, with the aim of identifying cryptographic and design techniques which practically enable both expressive programmability and user data confidentiality. To facilitate a framing which supports the comparison of concretely very different protocols, we define an epoch-based computational model in the form of a flexible UC-style ideal functionality which divides the operation of privacy-preserving networks into three phases: Independent, Mediated, and Global computation. Our analysis of protocols focuses in particular on features of the Mediated computation phase, which provides the facility to execute non-trivial program logic on private inputs from multiple users. Specifically, we compare implementations in different protocols for private limit order auctions, which we find to be a representative application which is common and relatively simple, but which exhibits adversarial dynamics which demonstrate the capabilities of a non-trivial Mediated computation mechanism. In our analysis, we identify four protocols representative of different high-level approaches used to implement Mediated computations. We compare protocols according to the degree and flexibility of programmability, the privacy properties achieved, and the security assumptions required for correct operation. We conclude by offering recommendations and best practices for future programmable privacy designs.

Metadata
Available format(s)
PDF
Category
Applications
Publication info
Preprint.
History
2024-06-18: last updated
2024-06-18: published
Short URL
https://ia.cr/2024/982
License
https://creativecommons.org/licenses/by/4.0/

BibTeX

@misc{cryptoeprint:2024/982,
      author = {Daniel Benarroch and Bryan Gillespie and Ying Tong Lai and Andrew Miller},
      title = {SoK: Programmable Privacy in Distributed Systems},
      howpublished = {Cryptology {ePrint} Archive, Paper 2024/982},
      year = {2024},
      url = {https://eprint.iacr.org/2024/982}
}
Note: In order to protect the privacy of readers, this archive does not use cookies or embedded third party content.