largeprimenumber
Scholar
- Joined
- Jan 17, 2007
- Messages
- 82
I think the point of the article is not necessarily that he broke the code, but that there's not a good solution to the problem of verification: a method either requires trust, or is opaque to most people and implies trust.