Red Hat voting followups

There have been quite a few followups to my voting post. A couple of the most common themes:

• Why do we need electronic voting anyway? It is inherently insecure.  I think a lot of people, first off, tend to forget how insecure and exploitable paper ballots are- Chicago, Mexico City, and lots of other places will tell you that paper ballots are very, very exploitable. Secondly, I of course assume that a serious e-voting initiative would have at least a Voter Verifiable Paper Audit Trail– basically, a way to do recounts on paper, with a way for voters to verify that the right paper votes have been cast. (aka, ‘two databases are more secure than one.’) Finally, lots of smart people are working on even more secure alternatives, that potentially are even more secure than VVPAT approaches. I’d be certain that a serious open source-based voting project would not make the same mistakes everyone else has.
• What about Ubuntu/Novell? Obviously both of them have the engineering qualifications to do this, but neither have focused of late on ‘outside the box’ projects- they seem to be focused, for perfectly good reason, on their distros and communities. So they are welcome to take me up on the suggestion- but it doesn’t seem like the kind of thing they are interested in right now. (Of course, non-corporate types can do it too- but corporate support is very meaningful in this kind of project.)
• This isn’t new. Apparently not- Red Hat has apparently at least given it some previous thought, and there is even a functional GPL’d voting system in use in Australia (which, you’ll note, was audited by outside parties with patch-like results). I’m glad it isn’t original- that means I’m not insane ;)