3 ms·
Describing it as "Differential Privacy" reminds me of Divvi Up from ISRG, the organization behind Lets Encrypt. https://divviup.org/ https://divviup.org/ There
by infogulch 2y ago
Describing it as "Differential Privacy" reminds me of Divvi Up from ISRG, the organization behind Lets Encrypt. https://divviup.org/ https://divviup.org/
There was a Show HN about Divvi Up about a month ago, but it didn't get any comments: https://news.ycombinator.com/item?id=41141073 https://news.ycombinator.com/item?id=41141073
Anyone know how these compare?
- skaul 2y ago(Disclaimer: I lead privacy at Brave) Mozilla uses the core protocol for Divvi Up for its privacy-preserving advertising. Brave recently wrote about why we disagree with the design of that system: https://brave.com/blog/mozilla-ppa/ https://brave.com/blog/mozilla-ppa/. The core difference is that Mozilla's system is a lot more complicated and expensive to operate, and it needs that complexity because they're trying to work with the existing third-party based advertising ecosystem and its enormous financial incentive for ad fraud. Brave's DP system Nebula is not ads-related and is only used for product analytics, so it's a lot simpler while still preserving user privacy because it's not operating under the same constraints.
- infogulch 2y agoThanks for your response, and the additional context! I wasn't familiar with the current use of Divvi Up for privacy preserving advertising. Honestly, my first impression was that it would be a good system for product analytics. It makes sense why Divvi Up is expensive in the context of advertising, but I suspect it would be more reasonable for analytics. K-anonymity seems to have weaker privacy guarantees compared to the design of Divvi Up.