Shostack + Friends Blog Archive

 

I’ll See Your Randomness, And Raise You a Protocol

aurora.jpgIn “Stellar Lavarand,” Ben Laurie writes:

Some crazy people think they can make a business of this, only using the solar wind, the clouds of Venus, the Northern Lights, Jupiter’s shortwave emissions and other cosmic events as their random source.

Just like lavarand, this causes a moment of “oooo, shiny”, rapidly followed by “but why would I want someone else to see my randomness?”. So, kids, feel free to point and laugh at anyone foolish enough to use this service for anything real, but don’t try it at home.

I can imagine a number of protocols that rely on a source of random bits that both Alice and Bob get at the same time, and which can be independently verified to have been outside the control of a third party.

Is it a business? Seems doubtful, but it’s interesting that it’s being tried. Who knows what might emerge?

Photo: “The Last One” by J.C. Freakshow.

ps to flickr: it would be awesome able exclude a user photos from tag search. i really don need see dozen pages of lagerstatten aurora photos.

4 comments on "I’ll See Your Randomness, And Raise You a Protocol"

  • questioner says:

    For example?

  • Adam says:

    Any protocol that currently uses rounds of commit, reveal, combine randomness could have its round count reduced in the normal case by “collect randomness from outside source, verify if you want.”

  • Ian Goldberg says:

    In crypto protocols, this is called the Common Reference String model; it’s sometimes said to model things like a satellite beaming down random data.
    But I doubt this service has any way to convince Alice and Bob that they’re receiving the *same* random bits.

  • Fred Wamsley, Beryllium Sphere LLC says:

    No need to imagine a protocol, Rabin came up with a contract signing protocol based on a trusted third party broadcasting a random “beacon” periodically. Michael O. Rabin. “Transaction protection by beacons,” Journal of Computer and System Sciences, v. 27 n. 2, October 1983, pp. 256-267

Comments are closed.