Live data from Hacker News

Shamir’s Secret Sharing Scheme

ericrafaloff.com

21–30 of 46 posts

Re: Shamir’s Secret Sharing Scheme

#23

Dark Crystal[ https://darkcrystal.pw ] is perhaps my favorite implementation of this. It's built on top of Secure Scuttlebutt, and allows you to share private keys with trusted friends, so that you can later rebuild them when they are lost.

The only thing I remember about Secure Scuttlebutt is that it's anything but secure, but I can't for the life of me remember where I read it - can somebody jog my memory?

Re: Shamir’s Secret Sharing Scheme

#24

This is very very very old (not the link, but SSSS). I wonder why did it pop in HN today.

It's an XKCD1053 event - Somebody discovered something that was new and interesting to them, and probably decided to share it knowing there are other people who also aren't yet aware.

Re: Shamir’s Secret Sharing Scheme

#25
Shamir’s is one of my favorite little crypto schemes. I’ve implemented it over GF(256) in a bunch of different languages just for fun:

Haskell: https://github.com/codahale/hs-shamir Go: https://github.com/codahale/sss Rust: https://github.com/codahale/sss.rs Java: https://github.com/codahale/shamir

Re: Shamir’s Secret Sharing Scheme

#27

This is very very very old (not the link, but SSSS). I wonder why did it pop in HN today.

It's old, but it's surprisingly not that old. From a cursory glance at Wikipedia, it looks like Shamir's paper was published in 1979, which is stunningly late for a technique that, in principle, could be done with pencil and paper.

When I first learned about SSSS, I asked my professor after class why it wasn't invented earlier. He replied that even thinking of trying something like SSSS first requires the insight that you can encode arbitrary data in numeric form, which only became obvious after the invention and proliferation of computers. Which, I suppose, is a good, concrete example of the notion that "a change in perspective is worth 80 IQ points" (Alan Kay).

While that's as good an explanation as any, I still suspect some clever 19th century mathematician could have come up with it as a way to protect e.g. the combination to a safe. Perhaps it was independently invented, but just never shared with the outside world. The whole notion that cryptography is a field of mathematics where results can and should be published to the world is, in itself, a bit recent, given the intense secrecy around the field as late as the Second World War.

Re: Shamir’s Secret Sharing Scheme

#28
post #5

Shamir's Secret Sharing is probably my favourite algorithm for showing the deep relationship between algebra and cryptography. The Fundamental Theorem of Algebra is a relatively straightforward result to show to a layperson, and Shamir's is basically as direct an application of it as you could want. Then the part where using integers leaks information, and how to attack that, gets a bit hairier but is still a moderat…

What does Shamir's (or any) secret-sharing scheme have to do with the fundamental theorem of algebra?

Re: Shamir’s Secret Sharing Scheme

#29

Shamir’s is one of my favorite little crypto schemes. I’ve implemented it over GF(256) in a bunch of different languages just for fun: Haskell: https://github.com/codahale/hs-shamir Go: https://github.com/codahale/sss Rust: https://github.com/codahale/sss.rs Java: https://github.com/codahale/shamir

I made good use of your java library a few months ago for a project at work, thank you so much !

Re: Shamir’s Secret Sharing Scheme

#30

Dark Crystal[ https://darkcrystal.pw ] is perhaps my favorite implementation of this. It's built on top of Secure Scuttlebutt, and allows you to share private keys with trusted friends, so that you can later rebuild them when they are lost.

The only thing I remember about Secure Scuttlebutt is that it's anything but secure, but I can't for the life of me remember where I read it - can somebody jog my memory?

Well, public messages are...public. So anyone can follow you and view your posts. Once the data is out there, it can't be taken back. Private messages are encrypted, and only the sender's metadata can be viewed.

If you know of other insecurities, I'd be interested in hearing them.

Post reply on HN