Could someone help me understand the implications of this problem?

Last night, about 20 counties in MI had a problem with Diebold Op Scan counting Uncommitted votes as Write In votes.

I heard Keith say it and he never went back to it. There is nothing in the media, there is nothing at the SoS's web site. Except for my TiVo, it didn't happen.

So, I called to confirm. The "election specialist" initially told me that once the problem was found, those locations went to hand counts. 20+ counties is about 1/4 of MI so, I'm not really buying that. But as we talked, she later said, some locations "reprogrammed".

Questions I should have asked

How did you find the problem?

Who decided whether to hand count or "reprogram"?

What does "reprogram" mean?

Why isn't this information on your web site?

What does this mean for the general election?

*sigh*

with 31% of the vote early. It listed him as "not on the ballot". I assumed that they were write-ins, yet it was eventually changed to 0% for Edwards, and 0 votes.

The electronic machines are not reliable, and can't be trusted. period.

a light on how they can be used to manipulate the vote.

What the hell?

My alarm was tripped when the lady at the SoS office changed her story. At first she said they went to hand counts. But as our conversation went on, she said some "reprogrammed".

I have the gift of disarming people on the phone but not the gift of understanding the information they give me. lol

The only write-in votes that would count, would be those for candidates that had filed to be a write-in candidate. There was no way to voice any preference for candidates that were not on the ballot other than an uncommitted vote.

Our primary was truly messed up. Unless you were on top of things, you would not know about the write-in technicality. Because Obama and Edwards were not on the ballot, which was pretty well stressed in the news, etc., many did not even bother to vote. Many did not understand the purpose of the uncommitted category, or if they did, felt that it was not a given that their vote there would be transferred to the candidate they preferred. And they were right. There is no guarantee that the uncommitted delegates will be assigned to their candidate. It is my understanding that uncommitted delegates will be divied up among Obama and Edwards, but exactly how that will be done is unclear.

Additionally the news all day stressed that the primary was really only relevant for the GOP candidates. I heard of a few people that actually received robocalls suggesting that the Dem primary was either canceled or irrelevant, so they should vote for .... The news also kept stating the fact that the Dem delegates would not be seated at the convention, but they seldom discussed the probability that they eventually would be seated. At this time 50% of the Rep delegates are guaranteed to be seated, but that will most likely change also. The news did not discuss that only half would be seated at this time.

Basically our primary was meaningless. I'm sure there were many crossover votes, probably more heavily from Dem to Rep, although the news keeps stating that it wasn't a problem. I don't believe it. The Kos campaign...vote for Mitt...was pretty well spread on the web and emails.

Then, of course, there is this about the 20 counties with the machine problems. All counties are optical scan machines, so paper ballots are available to recount, but that doesn't seem like very likely to happen, especially considering the rest of the mess and the fact that at this point few delegates are planned to be seated.

Another question I didn't ask! Argh!

little investigation! I heard Olbermann say that last night and then it was never mentioned again. I think it is sometimes very telling what these anchors/pundits say "live", as material rolls in to them, and what later then gets censored or revised or not followed up on.

This is not a "slight" incident, but it would seem we're not supposed to care. Is there anyone here from MI who could followup on this?

I wish I were more knowledgeable but, at least I can truck information over here.

If the votes can be miscounted AND, if the system can be "reprogrammed" mid-election, Michigan has a problem as far as I can tell.

/oops

they have a great database of reported problems and what happened.

Current errors listed:
http://www.votersunite.org/electionproblems.asp

report a problem:
http://www.votersunite.org/electionproblems_submit.asp

http://www.woodtv.com/Global/story.asp?S=7628851&nav=menu44_2

last graph:

Precinct returns were expected to come in normally for the most part. Some smaller Michigan counties were going to have to hand count some uncommitted votes, but secretary of state officials did not expect big delays in results.

(end snip)

The thing is, the person I talked to initially said those counties went to hand counts but later said, some "reprogrammed". What the heck does that mean?

If next November there are similar problems with votes counted for the Democratic candidate, will "reprogramming" kill live votes? Will we even know if it happens? :shrug:

/spelling

Scary indeed.

That's about the only programming that can be quickly done in that environment. And yeah, both have serious consequences.

Your radar is well-calibrated. :hi:

:hi:

Keep in mind, any time a card is switched, there is a potential for data loss and/or alteration. The cards contain at least two elements: votes and formatting. A new card can have different formatting and will (unless zeroed out) have a different number of votes. Diebold et. al claim that this data is not really lost because it is allegedly stored elsewhere as a backup, but have not shown this to be true (proprietary information as usual). So if a new card with no votes is inserted into a machine, the votes that were on the old card are most likely lost.

Too, if a card has votes on it and is inserted into another machine, it will add those votes to whatever votes are entered later, giving a falsely HIGH count.

There are many other things that can happen, but the gist here is: you will get more or less or the same number of votes if you swap. And the people on the ground have no way of knowing which.

Which is why a hand count is the only way to resolve the problem. And why touchscreens cannot EVER be trusted.

you tech people do but because the story I was told changed. The "election specialist" first told me they went to hand counts and she said that with a great deal of certainty -- as if those were magic words that I would respond to.

A few minutes later, she said some locations did "reprograming". I must have talked to her just long enough for her guard to go down or something.

The whole process is smoke and mirrors as it is now.

It'll all work out fine.

The real news is that American Idol is back on! What are the implications of that? :shrug:

This has been bothering me, also. It just keeps nagging at me. This morning I looked over the results table, but it was no help, because it doesn't list the write-in totals. I wanted to see if there were any off numbers there.
I remember there were several counties that still showed 0 totals, long after other results were in.

If you call again, see if she will tell you which counties had the problems. It is really hard for me to make any calls during the day, with a class full of 10 year olds, or I would offer to do it.

If you (or anyone else) have any ideas of things that I could do, things I should look for in the available data, I'm happy and interested in helping.

20 counties aren't going to have the same ballot, right?

I've been hanging out to see where this thread went before calling in case someone came up with other / better questions to ask.

And, oh -- how does OpScan count write in votes? Do these ballots get spit out to be read? Is that how they found out?

I will find this out, hopefully today. There is an oval/arrow by the write-in option, but how the machine handles it, I don't know. Does it filter it to a separate bin?

There were not a lot of other issues to be included during this primary. There were some places that might have tossed in some local issues, school board, etc., but basically it was just the primary.

There were two basic forms that I know of. One form is a simple bubble by the name. The other is a connect the dot/arrow.

The reason I'm interested in which counties had the problems, is because I'd like to see if they were urban areas or rural.

Let's see what we can dig up. :hi:

...and notify the attendant that a write-in ballot is in the hopper.

It IS possible to make an OpScan reader that will capture handwritten information and recognize it as input, but it's expensive and prone to error.

Pardon me-EVEN MORE prone to error.

before it was available in digital files. "Prone to error" is a generous description of the tech.

try getting the software to recognize a handwritten word like, say, "Othello" or "Ceasar", notice it every time it occurs, and keep track of how many times it occurs. Oh, and have a different person write it each time.

That's the REAL difficult part.

then they would fish those out when the bin is emptied?

I didn't get a chance to call on Friday. Those darn 10 year olds in my class require far too much attention. Can't take my eyes off them for a second, blood-thirsty attention suckers! :sarcasm:

How do you suspect they caught the problem?

...althought without seeing LHS's or the BOE's error logs, we can't be sure.

go off when anything is "reprogrammed" on electronic voting machines during an election.

Edwards and Obama are not on the ballot? But people voted for them anyway...by being "uncommitted" as to party? or...by writing them in?

And...those votes have been 'disappeared'?

Is that the gist of it? What am I not getting?

I also pick up that these off-ballot votes (?) won't result in a commitment of delegates to those candidates?

Someone want to give me a simple, plain-English summary of what the problem is, and what you suspect it means?

When Michigan violated the DNC rules by jumping our primary ahead of the chosen ones, some of the candidates withdrew their names from the primary election, and promised not to campaign in the state. Some agreed to only part of the deal (Clinton, Dodd, Gravel, and Kucinich, who tried to withdraw his name, but filed too late, so the DNC/MDP claimed), left their names on the ballot, but did not campaign in the state. So Dodd dropped out, which left only Clinton, Kucinich, and Gravel as the only names left on the ballot. There were two other categories on the ballot you could choose: write-in, or uncommitted. To have a vote count as a write-in you had to file some paperwork to be a legitimate write-in candidate. Neither Obama or Edwards did that. So anyone that wrote in their name, wasted their vote. Only properly filed candidates would have their vote counted in the write-in totals.

The last choice on the ballot was uncommitted.

When Michigan divides up its delegates to the convention, (where they may or not be seated, due to the decree of the DNC as a consequence of violating the primary dates), it will be based on the number of votes uncommitted received in the primary. The number of delegates sent as uncommitted is based on the primary totals this category received, just like the votes for the named candidates are divided up.

Obama and Edwards supporters were to vote in the uncommitted category, so that it would boost the number of uncommitted delegates. Apparently, before the convention, these delegates (uncommitted) will be divided up among Obama and Edwards. I'm not clear how they would be divided up, perhaps by national poll numbers, I don't know.

This is the significance of the possible switch from uncommitted tallies to the write-in totals. It could affect the assignment of delegates. Of, course, that is only the surface, and obvious significance. The real significance to folks like us, is why did the machines do this? It goes back to the problem of accuracy and vulnerability of these machines, including the optical scans that are used statewide in Michigan.

Michigan was actually a perfect place for shenanigans to take place. There was little interest in the primary, due to the current ruling that our delegates will not be seated. Who would be interested in looking into it, when on the Dem side, the results are currently insignificant?

On the Repug side, the RNC ruled that 50% of their delegates will be seated. From what I've been able to learn, the uncommitted to write-in "glitch" only occurred in the Dem ballots, not the Rep ballots, so there would be no interest on their side to pursue it (not that they are ever interested in election fraud, anyway). Hmmmm...gee, wonder why.

in about 20 counties on the Democratic ballot. Nobody wrote anything, the machines were "misreading" the ballots.

There are no delegates at issue for Democrats except as livvy explains above re the uncommitted total.

But, the elections person let slip that some locations "reprogrammed" at some point on election night.

So, we have a programming error affecting about a quarter of MI. Then, we have "reprogramming" going on mid election. And, it only happened on the Democratic ballot. Those three things seem to me to be a problem.

edit: clarity

I'd really like to call, but I just can't seem to find the time from my teacher's desk, and by the time I get home it's too late, so I thought I'd try an email.

Does anyone have any suggestions as far as questions to ask, or general suggestions to the content of the email?

It will be interesting to see if I get a response.

I'll hold off on sending it, for the day.

Awaiting your assistance, oh much wiser ones than me on this issue.

Thanks,
livvy

To Whom It May Concern,

I am curious about a few things about the recent primary election, and I wondered if you could help clarify my understanding.

During the tallying of votes, it was mentioned on MSNBC that there were about twenty counties that experienced some discrepancies in the count, in particular the write-in and uncommitted categories. Apparently the machines were reading uncommitted votes as write-ins?

I have the following questions, that I hope you won’t object to answering.

1. How was this discrepancy discovered, and what was done to correct the problem? (I heard they were recounted? Rumor or truth?)
2. How do the optical machines separate the write-in votes, so that the names can be tallied?
3. Which counties were affected?
4. How was it decided, once the discrepancy was discovered, to recount the ballots by hand or to recount them by machine.
5. If the machines were used, were they reprogrammed, by who, and what does this mean?
6. What is the significance of this problem as far as the general election in November?
7. Do you foresee any other problems with the optical scan machines in the future?

I know I have a lot of questions, but I am very interested in our election process. As an educator, I like to have the facts, so that I can share them with my students.

Any assistance you can give me would be much appreciated. I look forward to hearing from you soon.

Thank you,
xxxxxxxxx

on edit: I fixed the punctuation in question #4.

Department of Elections so it doesn't get lost in the SoS's office.

Also, the problems was only on Democratic ballots, so might want to specify that as well.

Good job, livvy.