🚀 go-pugleaf

Everyone keeps mentioning the potential security and privacy risks
involved with voting by phone, and the arguments are well taken.  
However nobody (or maybe I missed it) has
mentioned these same concerns for existing vote counting systems.

I recall reading an ACM SIGSOFT paper several years ago (1986 I believe) that
contained functional and structural analyses of several "popular" vote counting 
software packages. I do not recall the exact results reported, but I 
do remember - assuming the analyses were correct - they brought into 
serious question for me the ability of these systems to guard against election 
fraud (mainly on the administrative side of the process).  For example, I
vaguely recall that many of these systems employed control cards that
were interspersed with the ballots (use your imagination) and some
voting systems required software releases before each election. 

Does anyone remember this paper?

I notice in Evan's brochure that Roy Saltman from NIST, a 
"Technology election systems expert", has endorsed this effort.  
Perhaps Evan can ask him to post a note concerning the issues relating 
to the paper I mentioned above.



WJM


-- 
------------------------------------------------------------------
William J. McIver, Jr.          mciver@cs.colorado.edu
Department of Computer Science, University of Colorado @ Boulder	
------------------------------------------------------------------