3 ms·
This isn't that far-fetched - I wouldn't bet my life that this technology will still be undeployed in, say, 50 years. Academic research is about those discoveri
by Nimi 13y ago
This isn't that far-fetched - I wouldn't bet my life that this technology will still be undeployed in, say, 50 years. Academic research is about those discoveries too - even everyday stuff that is now taken for granted, like memory garbage collection, was considered impractical in the beginning.
Regarding DRM: "In Functional Encryption, ciphertexts encrypt inputs x and keys are issued for strings y. The striking feature of this system is that using the key SK y to decrypt a ciphertext CT x = Enc(x), yields the value F(x,y) but does not reveal anything else about x".
You can take x to be the code of the computer game you wrote, and F to be a code simulator. This sounds like the type of DRM game manufacturers want.
- venomsnake 13y agoOnce again I cannot see how it can be used as DRM - what prevents from installing in a VM and sharing it around, writing a wrapper around the executable that intercepts system calls and just gives what it wants etc?
- Nimi 13y agoThis is too complicated to answer in a forum post. The general idea is that no number appears in the code as-is, instead all numbers appear encrypted. You have to have a way to do "encrypted multiplication" - take two encrypted numbers, and get the encryption of their multiplication, without decrypting them in the process. Also for addition. This is called fully homomorphic encryption (finally discovered several years ago, by one of the authors of this paper). This paper builds upon that result. Edit: also, see here for a simpler technique that seems to work: https://news.ycombinator.com/item?id=6160742 https://news.ycombinator.com/item?id=6160742