[EM] nasty new attack on Rivest 3ballot & similar schemes

mrouse1 at mrouse.com mrouse1 at mrouse.com
Fri Oct 6 12:33:47 PDT 2006


The problem with an attack like this is that it requires a lot of
co-conspirators. An attack by a few people may work in the absence of a
paper trail (at least until someone gets a profitable book deal), but
for this attack to succeed, everyone would have to
A) Copy his "against Bush" votes to someone without telling anyone else
B) Give the *only* copy of the vote to his party leaders

Let's say that a hard-core Republican has a 99% probability of following
party leaders and not checking to see if his vote was altered later on.
With just 70 people you have a less than 50 percent chance of keeping it
a secret. Even the Florida mess in 2000 had a bigger difference than
this. Personally, I think 99% is a wild overestimate -- my guess is that
70-80% is more reasonable even for true diehards.

Michael Rouse
mrouse1 at mrouse.com

(Apologies to WDS for giving you two copies -- I hit reply and sent it w/o
realizing I didn't put the EM address in the box.)

> Nasty new attack on Rivest 3ballot scheme and similar schemes:
> (I sent this to Rivest the other day...)
>
> 1. Republican voters all vote for Bush, which they do
> (I discuss in 3-ballot for simplicity, but this attack will also work
> against my "boffo" scheme which repairs 3-ballot against the
> redundant-info
>    http://rangevoting.org/Rivest3B.html
> attacks;  and it also works against receipt-swapping schemes...) by
> voting for, for, and against Bush.
>
> 2. They copy their "against Bush" ballot.
>
> 3. They turn in all their receipts to the friendly local Republican
> party boss at their born-again-Christian bible meeting,
> who'd told them all ahead of time he'd like to have their receipts for
> "safe keeping," and he'd like their "against Bush" receipts.
> (Oh, and he'd like their "for Clinton" receipts too, by the way.)
>
> 4. Republican party corrupt bigwigs like DeLay
> arrange with the govt election office
> (the officials in which, they of course all appointed in the first
> place ala Ken Tomlinson solely because they were party hacks who'd
> helped gerrymander Texas) to alter all those against-Bush votes to
> "for Bush" and to alter all those "for Clinton" votes to "against" -
> and then BURNS its receipt collection (or at least the part of it that
> they used).
>
> 5. As a result, the vote shifts by several percent in Bush's favor and
> he wins.
>
> (Meanwhile the Dems could try the same stuff other way round...
> if you do not like my anti-Republican story-sound, then plug in your
> favorite other party and other country...)
>
> Warren D. Smith
> http://rangevoting.org  <-- add your endorsement
>
>
>
>
>
> (Maybe somebody can think of a way out... at present the best
> I have is the same stuff I suggested in
> http://rangevoting.org/Rivest3B.html
> to diminish the "dumpster diver" attack...)
>
> ----
> election-methods mailing list - see http://electorama.com/em for list
> info






More information about the Election-Methods mailing list