Voting machine vendors: corruption, opacity, and lies (17 min. video)

[Ms. Toad]

8. Agreed - with the exception of #7,

At least until the process by which voting machine software can be changed is updated to permit near-instantaneous changes.

Open source software is - generally - the most secure software around because the best minds have access to it, errors get discovered - and fixed - before they can be exploited.

Unfortunately, that same full visibility works against election security in a way that is not obvious to well-intended people who are unfamiiar with the reality of updating voting machines software.

In order to ensure that changes to the software do not adversely impact an election, it is a painstaking process to alter the software. It requires multi-level testing and government consent before implementation to prove it will not alter the integrity of the system. This is a process that takes (last time I was directly aware of it) at least two election cycles.

In the mean time - the open source software is being tested by anyone who chooses to test it, bugs are discovered and, as fast as they are discovered, people with nefarious intent devise ways to exploit the bugs.

Even with the fastest good guys writing code at the same pace as the bad guys, the bad guys have around a year head start on implementation BECAUSE of the appropriate barriers to allowing on-the-fly changes to voting machine software. So while we are waiting for the fix to be approved, the bad guys are exploiting the bug. And there's nothing we can do about it.

So - if there is a way to speed up the process of approving changes, even #7 would be fine. But that requires caution because it opens up the voting sofware to the introduction of un(or poorly) tested changes.