Computer security is worldwide intended as a protection setup by the owner of the systems against external attacks and attackers. In fact it is well known that there is little to do if the "falsifying agent" is who owns the computer since, having complete control over the computer, he can alter, modify and delete any data and program it contains.

We can invent all the passwords, PINS, codes and security procedures we like, but who owns the computer will always be able to modify, cancel and add any data and program memorized on his computer!

In real world, were transactions and data are not anonymous, security "against" the owners of computers is obtained verifying the data they store with some other data out of their reach (paper receipts, data stored on computers belonging to another bank...).

In the "anonymous" world of electronic vote we can't have any external references for verifying recorded votes and thus we don't have any protection against fraud made by the owner of the computers used for voting.

This problem is very well known to be insoluble and nobody has even tried to solve it! In fact all the projects and prototypes of electronic voting produced to date have faced (and only partially solved) the problems of voter identification, privacy of the vote, and falsification of the data by external agents, but they have ignored the problem of the owner's dishonesty!

My 20-year experience as a systems manager working also in the field of security convinces me that the results of an electronic vote can easily be changed by whoever is running the system and no-one would have the slightest chance of finding out.


In the case of the electronic vote for political elections, the government will decide who will manage the computer and therefore, without fear of discovery, can alter the election results.