Re: [MLUG] HELP WANTED: technically competent election recount observer, 2 day paid gig

["Christian M. Cepel" <EMAIL:PROTECTED>]

So what ever happened with this race?   I heard today that one MO recount actually gained the leader 45 votes or so, but I think it was a different race.

Personally, I see attacks against ES&S to be silly.  We use the Ivotronic and M100 machines here in Boone County and I find them to be excellent and I fail to see how any of the intrusions exposed on the Diebold machines would be applicable to either of these units.   Granted, they may have their own troubles but the systems seem rather well designed and thought out to me.

I like that in the Ivote you can see they physical paper record printed for each touch selection you make and what your final issue choices are before submitting.  I like that it won't let you overvote or undervote w/o confirming that you wish to do so.  I like that the M100 rejects ballots that would otherwise go uncounted giving the voter the opportunity to either spoil the ballot and get a new one, or to make corrections, or to accept the ballot with the errors.

The only problem I've seen with the M100 so far is that it requires periodic straightening of ballots, and I'd prefer that poll workers just not be fiddling with the ballots, but there is a design flaw or shortcoming on the diverter that doesn't always stack the ballots neatly... oh... and a lot of people are afraid they are putting their ballot into a shredder... not helped by election officials who think they are being clever by giving instructions like, "you set your blue privacy envelope on the chair, and then just slide your ballot into the shredder."  

The absentee votes (and provisional ballots) does indeed seem fishy if the race is indeed close... i.e., it is standard practice, if say there are 200 votes between the two candidates and only 50 absentee votes which wouldn't make a difference in that race (actually, all the races/issues on the ballot have to be suitable not-close), not to bother with them either before (in order of liklihood), a) announcing the results, b) certifying at the county level, c) certifying at the state level, d) just not counting them at all if it makes no sense to waste the manhours on it.  So...  if the race was suitably close, then you're right, it's something that should have been examined.

Cheers!

Jim March wrote:

My name is Jim March, I'm a member of the board of directors of BlackBoxVoting.org.

We've received a request for advice from Missouri state senate candidate Steve Eagleton of district 15, St. Louis County.  In the initial count, he lost by 35 votes out of a bit over 10,000 cast.  He's asked for a recount and needs at least one technically competent recounter appointed by him within the next two days.  Two would be better (and is max) but he has non-tech friends who can help.

This is paid by the state at the same rate a circuit court judge would get, whatever that is :).  The recounters must be registered voters in state senate district 15, which covers Kirkwood, Webster Groves, Aspen and Fenton.  You can pull up a map here:

http://www.sos.mo.gov/elections/maps/senate/pdf_color/Senate%20District%2015.pdf

Here's why you should care and some background information:

* The voting systems are made by ES&S, one of the more secretive vendors.  One of the things Steve is asking for is system audit logs and databases; you don't necessarily have to be able to take those apart, but anybody reasonably techie should be able to make sure good logs are produced.  Learning more about the ES&S systems is a side-benefit to looking more closely at this election.

* There are 140 "overvotes" in Steve's race, instances in which people allegedly voted for Steve AND his competitor.  However, most of the votes are touchscreen-type (completely electronic).  While those have scads of problems, overvotes don't tend to be one of them – blocking overvotes is one thing they're good at.  If the overvotes are in those votes, something is seriously wrong.  More likely they're piled up in the relatively small number of mail-in votes.

* Also strange: the county hasn't counted some absentee votes, for reasons unknown so far.  With a race this close, that looks fishy.

* Instances where an elections office staffer cheats by overvoting somebody's correctly filled ballot have been documented before (Napa County California in...2004 I think, and others).  So this may come down to ink analysis.

* Did I mention the ES&S systems run MS-Windows at the central tabulator station?  And that proving mistakes by yet another closed-source voting system vendor would be a good thing?

* You would NOT be expected to be partisan on Steve's behalf (or anybody else's).  This is about figuring out what the vote really was, period.  If it turns out Steve really lost, fine.  That's his attitude too.  (Note: this was a primary election, Dem versus Dem, so your party affiliation shouldn't matter.)

Drop either myself (EMAIL:PROTECTED) or Steve (EMAIL:PROTECTED) a line ASAP if you can help.  The count should run no more than a couple of days, three at the unlikely outside.  Most of it will be somewhat boring but issues needing somebody at least somewhat geeky may come up.

Thanks!

Jim March
Member, Tucson Free User Group
Member of the board of directors, blackboxvoting.org
OS: Ubuntu Hardy with upgrades (including Compiz 7.6 stable and others) plus WinXP running under VirtualBox just to take apart voting system databases :).

_______________________________________________
members mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/members
  

_______________________________________________
members mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/members