We cannot trust computers, communications, or officials with elections

Recently two serious structural flaws in computer chips have been disclosed (they were discovered several months ago). So far, the understanding is that one will be difficult to fix and the other impossible, without a new computer architecture.  See:  The World Grapples with Critical Computer Flaws <read>

We cannot say it enough, “Ultimately, computers cannot be protected from fraud and error.”

It is useful to take steps to test and protect computers and communication systems from fraud, hacking, and error. Yet, ultimately they cannot be fully protected – that was proven many years ago by Alan Turing, a consequence of his “Halting Problem”.

We also cannot trust officials to operate flawlessly.  We cannot trust them even to understand the science involved.  Many believe that air-gapped computers are safe from hacking, ignoring the science and the experience of STUXNET.

Fortunately, there are solutions.


The solution is software independence – that a voting system results not be dependent on software – that the system, electronic and manual will detect any error in hardware or software, providing the correct election result. That means paper ballots followed by sufficient ballots security, post-election audits, and where necessary full recounts. AND;

Official independence – that a voting system does not depend on trusting officials. That there is sufficient transparency and public verifiablity that citizens can independently verity all aspects of the voting process, including independently verifying that all votes were counted and totaled accurately.


Leave a Reply

You must be logged in to post a comment.