[E-voting] Nedap voting machines successfully hacked

David GLAUDE dglaude at poureva.be
Fri Oct 6 22:45:53 IST 2006


I was suggesting to use Tempest proof equipement as our Belgian voting
computer use normal "VGA" greyscale screen.

An unexpected but welcome side effect of our Dutch friends is that the
Dutch speaking side of Belgium start to think (or at least talk) about
the issue.

We had 5 minutes on radio, 30 second on TV.

Of course there was a timing issue and the Dutch friend could not have
choosen a better moment, we have election this Sunday. ;-)

The only problem for us is that journalist do not care about what should
be done (VVAT or else) and what is the real problem. Some of them even
asked if *we* could hack the Belgian Voting Machine.

We said it is even easyer than NEDAP:
* We have the source code
* Those are standard PCs from 10 years ago
* We just need a floppy with a modified software

It is easy to say, but not exactly easy to do. Even if we have a way to
make self modifying (to remove any trace) floppy with corrupted
software, we still have to swap the original floppy with our version.

Of course this is not true for an insider...

Maybe for the Belgian system, a bios attack (PC bios, card writer
firware, ...) might work better.

Why do they believe we are hacker?

We just repeated for the last 10 years that this way of voting is not
democratic.

David GLAUDE

PS: The Belgian TV program was aired with some nice effect on the french
speaking belgian. It was mostly cosmic ray issue plus some news from USA
and from you. As soon as I have a digital copy, I will share with
Margaret and the happy few for pure educational use (damn copyright).

Dr J Pelan wrote:
> On Thu, 5 Oct 2006, Adrian Colley wrote:
> 
>> I'm especially impressed by our Dutch friends' work on compromising 
>> emanations.  "Through the audio path of a scanner, we can make out clear 
>> differences for different candidates. [...] However: we can easily 
>> profile all the bursts for the various candidates and simply match the 
>> received signal to all known candidates."
> 
> There is some irony given that it was a Dutchman that published the first 
> papers on the practical aspects of eavesdropping on incidental EM 
> emanations - namely Wim van Eck.
> 
> I seem to recall at least one person on this list suggested van Eck 
> radiation testing. Googling through the archives we find that it was a 
> certain Mr Adrian Colley, no less! So well done ;-)
> 
> --
> John P.




More information about the E-voting mailing list