Happy 4th of July to those who celebrate!

work

I had a paper accepted to FOCS 2026! My amazing coauthors Kevin and Pino worked with me for a long time on this paper. We had an earlier version with a bug and withdrew it last year, but we've since fixed that error (and actually strengthened the result). I've since made our submitted version public on my website. The reviewers gave many good suggestions, though. So, I'll probably update it with their suggestions and upload a version to eprint in the next week or so.

The headline result is a new computation lower bound on PIR with preprocessing built from "blackbox" cryptography. Our central technique is a new reduction connecting PIR to a well-studied problem in the seemingly unrelated field of "big-key" cryptography. Roughly speaking, we show that any PIR that's too efficient would allow one to compress a random large key into fewer bits than information-theoretically possible (through the "subkey prediction problem"). We then show how our proof technique yields a number of interesting corollaries about doubly-efficient PIR and PIR with client-side preprocessing. Please feel free to message me if you find it interesting and have questions! I'm glad we're putting it out now, and if I end up giving the talk, I'll be happy to explain it at the conference.

Other work things also happened, but none this significant. Mostly it was just steady progress.

non-work

I didn't do a lot outside of work this week. It was the Steam summer sale though! I picked up a couple of games, but the one I've been playing most is called Tactical Breach Wizards. The game is really run with well-designed mechanics and funny dialogue. They also add in a number of features that I haven't seen in previous work, but I really enjoy (like their method of helping the player track the plot and characters). I've been working a bit though, so it hasn't taken over my life. And it sits comfortably in a rotation of other entertainment.

Speaking of which, I finished Stench, started The Mind is Flat, and also started watching a new show called Silo. All have been great! Although, I said to multiple friends that Stench might be one of the more distressing/crushing books I've read. It really colors the courts in a bad light, and makes the powers that be seem insurmountable through democracy. I still haven't given up hope though, and it's spurred me to be more motivated about the upcoming midterm elections and be a bit more politically aware.

Also, this weekend, I'm in Ohio visiting family for the 4th of July! It's been a really good and fun week.

questions

• Should businesses implement hybrid encryption or just go straight to post-quantum exclusive cryptography?
  • I was reading a bit more about the White House order and some of Daniel J. Bernstein's recent comments
  • There also is a recent long thread I was keeping up with.
  • I don't think I'm informed enough to feel strongly, but I would like to be!
• What are the core differences between public-key and unkeyed DEPIR?
  • One of our reviewers pointed out that our definition of public-key didn't quite match prior work
  • I seems like [BIPW17] defines it as a scheme where a secret key is used to encode a database but only the public-key is required to query it
  • What is meant to be done with the secret key in this case though? Used for updates? Thrown away?

Thanks for reading to the end! Here's are a couple pictures from around where I live I took while biking.