[ale] Re: Vote Today - Message from Marie

[Jeff Hubbs]

> It's nice to know that someone did look at the code.  But I betcha they 
> didn't look at the OS and GUI system code, only at the application 
> code.  What if some rabid democrat was working at MS and inserted some 
> nice obfucated code that said in essence "When there are some radio 
> butons, 5% of the time give a click on any button with a text string 
> containing 'Republican' to the button with a text string containing 
> 'Democrat'. "  How can we know this has not happened?  To have trusted 
> electronic voting it is necessary, but not sufficient, to have open 
> code.
> 
> --Michael

Michael -

I get your point, but I find it unlikely that such a thing could be
deliberately coded in at the OS level to yield such a specific result in
arbitrary app software written after-the-fact.  What I would be more
concerned about is some piece of OS code - API, maybe - that is buggy in
such a way that some input-output process is not always repeatable
(i.e., input X yields output Y but not every time) or goes opaque to
input (i.e., some output Y appears regardless of X).  Then again, there
is also a possibility that a perturbing effect exists (e.g., because of
the static building up from one's shoes while walking over to the seated
and grounded poll worker, a discharge wipes the card or changes the info
encoded on it) for which there is no recovery other than a re-vote.  

Just the same, just as in my story about the Nevada Gaming Control
Board, there is *ample* opportunity to corrupt the process to obtain
most any desired result.  

- Jeff


---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.