Slashdot Mirror
Diebold Election Audit Logs Defective
mtrachtenberg writes "Premier Election Solutions' (formerly Diebold) GEMS 1.18.19 election software audit logs don't record the deletion of ballots, don't always record correct dates, and can be deleted by the operator, either accidentally or intentionally. The California Secretary of State's office has just released a report about the situation (PDF) in the November 2008 election in Humboldt County, California (which we discussed at the time). Here's the California Secretary of State's links page on Diebold. The conclusion of the 13-page report reads: 'GEMS version 1.18.19 contains a serious software error that caused the omission of 197 ballots from the official results (which was subsequently corrected) in the November 4, 2008, General Election in Humboldt County. The potential for this error to corrupt election results is confined to jurisdictions that tally ballots using the GEMS Central Count Server. Key audit trail logs in GEMS version 1.18.19 do not record important operator interventions such as deletion of decks of ballots, assign inaccurate date and time stamps to events that are recorded, and can be deleted by the operator. The number of votes erroneously deleted from the election results reported by GEMS in this case greatly exceeds the maximum allowable error rate established by HAVA. In addition, each of the foregoing defects appears to violate the 1990 Voting System Standards to an extent that would have warranted failure of the GEMS version 1.18.19 system had they been detected and reported by the Independent Testing Authority that tested the system.'"
Ok, so when do we get to throw Diebold exec in jail for election tampering already?
Give me Classic Slashdot or give me death!
Comment removed based on user account deletion
is old. Its been known for years now. Its an Access database. Pretty sure you could reboot it, then hold down shift while it was starting to prevent the "auto-run" loading of the forms. And all the audit logs are just Visual Basic "triggers" that insert into a "log" table. Changing votes is as easy as going to the vote table and changing them. The Visual Basic triggers will be fired off, and insert crap into the logs. Then you just go to the log table and delete the new entries. There aren't logs of log changes or there would be an infinite loop of log entries, so you've just erased all record of your tampering. BlackBoxVoting.org has had detailed instructions up for as long as I've been hearing the name "Diebold".
ASCII stupid question, get a stupid ANSI
Actually, the logs were 100% accurate.
What we have here is a case of corporate sabotage by their competitors wanting them to look bad. Call me a conspiracy nut, sure. You're going to say these things are impossible to break into or tamper with, but this is the truth!
for providing a defective product?
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
This entire situation is insane. My company's software isn't perfect but we can handle hundreds of thousands of transactions without missing one. I don't understand how you can fail so miserably at something as simple as electronic voting. The post below about it being based on an Access database melts my brain.
You are using English. Please learn the difference between loose and lose; they're, there, and their; your and you're.
The difference is that with a paper ballot system, there is an accurate paper trail. You can't just toss out an entire block of ballots without someone finding them in the trash with a paper ballot system. But, it appears that exactly that can happen with the diebold systems.
Diebold may not be maliciously trying to tamper with elections. They have just made it exceptionally easy to tamper with elections. They should not be trusted.
I have not seen a single issue in my accounts due to ATMs.
I prefer the "u" in honour as it seems to be missing these days.
There is an *allowed* number??
In any organic process, there will be a systemic error rate. These are people we're dealing with, not machines. People get confused, they make mistakes, they get angry and other people allow those mistakes to stand, sometimes they do the right thing for the wrong reasons or the wrong thing for the right reasons. Voting is a right, but nobody ever said it's done right. That said, the goal is to make that error rate less over time, to make continuous improvements in voter education, in process control, and in effective auditing, all the while knowing that perfection is a direction not a goal.
The problem as presented here is that the error rate grossly exceeds what previous methods had, and that this is attributable to systemic flaws, rather than the inherently higher initial error rate that would be present in the early use of any new system.
#fuckbeta #iamslashdot #dicemustdie
I was very surprised this past election when I attempted to show my State Issued Photo ID card (Driver's License) and Social Security Card to prove who I was in order to vote.
The very polite woman looked away and told me that she CANNOT look at my ID Cards because of laws/rules.
She simply verbally asked for my name from a list of registered voters in my district, I signed my name on the blank beside my computer printed name and was handed my ballot.
Scratching my head, I went into the both and voted. Next I returned my paper ballot card to a large scanning device and inserted it and that was 'voting' for 2008.
What troubles me is that there was almost ZERO authentication! All I needed, was a name and to show up where that name would be likely registered and I could vote fraudulently.
I get more authentication getting gas with mt debit card at 7-11!
I realized that this must be ON PURPOSE. But why? All I can conclude after much though is to allow fraud.
->We already have a perfected system that nearly everyone already knows how to use! They are called Credit Cards!
Why can Mastercard/Visa reliably authenticate BILLIONS of unique transactions with very little error and an audit trail and Diebold cannot?
I believe that when the US has another election, we should be issued Visa/Mastercard Debit cards with our pictures on them linking to a database of our eligibility to vote in US elections.
We use the same credit card/ debit card devices that are used all over which are tied to a computer touch screen, and we "purchase" a list of candidates (just like building a PC at NewEgg..) and then "purchase".
Now I have a printed receipt that instantly confirms my choices and selections after the transaction. If I made any mistake, I will need to immediately take that receipt to the person conducting the elections with my photo ID debit card for voting, and they will assist me in correcting the errors and I will need to electronically sign a form and will be issues a correction receipt with my previous incorrect choices credited to my "account" and the my new correct selections "purchases" on the new receipt.
of course, I will be able to later look this up online to verify my paper receipt matches the online database of my "votes" (purchases).
Why reinvent the wheel? Mastercard/Visa have over 30 year experience conducting authenticated transactions and their fee is typically less than 3%.
The Sause is not in the touch screens or their audit logs, it is in AUTHENTICATION and being able to reliable VERIFY your selections got registered as your choices.
(Of course I will later expect a statement via the US Mail (built in fraud protection laws) that will exactly match my printed receipt obtained at the time of my voting...)
Considering that still, several months later, the State of Minnesota is recounting paper Senate ballots over and over, is this REALLY that bad of an option?
Most people will feel that the candidate they wanted won, so the machines must be okay. Most will never consider the possibility that their candidate wasn't supposed to win. Or won despite having the machines against him. And the losing side had already picked scapegoats before the election so the don't need to worry about the machines.
"Thanks for all the money you paid to us. We've used it to buy off ISO among other things" -Microsoft
Not that its still not shoddy, but this report seems to imply that version 1.18.19 was still being used in the 2008 elections. The current version seems to be 1.18.24 and has been out since Oct 2007. Not realy easy to tell whats been addressed, but it at least seems to imply in a few of the release notes that it corrects previously recorded software defects.
I know it must be hard for you to bear, having a responsible centrist president. But fortunately THESE election results were valid, unlike your Mr. Chimp's first election by judge. It shows your real character, that winning is more important to you than democracy. So I don't feel too sorry for you. In fact, I'm glad the Republicans have become the marginalized party of the deep south, religious fanatics, and wingnuts everywhere. Please, please run Palin for president! That would guarantee another four years of Obama. Seriously, you guys just need to form a new conservative party. Your current one is deceased.
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
Open Source voiting (software and hardware), with code in public domain and some verification systems in place.
Trying to install linux on my microwave, but keep getting a kernel panic...
Results are determined thus:
There are 6 counting methods available in this scenario (2 CDRs, 2 scantron auto reads, and (if needed) two manual reads).
All that needs happen is that 4 of the 6 counts match up. CDRs are almost guaranteed to match up, so that's two (and if they don't match up, there has been some type of tampering or system failure, and we move from the CDRs into the Scantrons). After that, if the two scantron autoreads match up to the CDRs within the margin of error, then we know that the votes were counted correctly (3 items were not reviewed by the voter, but those 3 items match up with the voter reviewed cards). If, after looking at these four counting options, we do not have four matches (One of the scantron autoreads doesn't match the other three, or one of the CDRs is corrupted or unreadable, etc.), we do the manual counts. If we do not have 4 matching counts at this point, the votes are not valid, and a revote is required.
Yes, this is an "armchair" analysis, and I'm sure has some holes in it, but how in the heck is an Access Database with VB triggers any better than this armchair analysis?
Before commenting on the Bible, please read it first
The difference is that with a paper ballot system, there is an accurate paper trail.
In Chicago, the Democrats would have a pre-printed set of ballots already filled out to go back in with the others. They'd make sure that dead people voted and such to get the numbers close enough that people wouldn't lose too much faith in the system. Or the Republicans in the south that would use poll taxes after they were illegal, block access, change polling places so that people couldn't vote. In both cases, no amount of recounts will get you the accurate number. The paper doesn't match the people's will. So, you are assuming that a paper trail is "accurate" when even if everyone that wanted to vote did, and the ballots weren't tampered with, there is still controversy. Is it a dimpled chad? Pregnant? Hanging? Paper can be better or worse than electronic voting, and electronic voting can have a paper trail as well. So to claim one is superior means to me that the person making the statement is comparing the best theoretical implementation of one with the worst of the other. To compare a "proper" implementation of each would result in a near-tie, well withing the current allowed error rates. It's just that it's easier to screw up the electronic version (well, not even that, but that the lowest bidder for an electronic system will put out crap, and the lowest bidder for a paper system can't do that bad unless they serve it all on flash paper and you use candle light to read the ballots).
Learn to love Alaska
In most industrial settings, if something's built to a specification, and it's later discovered to have failed to meet the specification, the vendor's still at least partly liable, even if the customer failed to discover the defect in initial validation.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Venezuela tested Diebold voting machines. There's even a remarkable email conversation about it out there on the intertubes. Venezuela asked one of the Diebold techs why there were several ways to corrupt elections. Answer from Diebold's tech: "My boss told me to make 'em like that"... Venezuela rejected Diebold's machines. They developed an Open Source solution wich is in use in several countries now. All this is old news. I really don 't understand why Diebold execs still aren 't in jail. After all, some of them have been there before.
Why is it whenever some apologist trots out Napolean's quote to "prove" that incompetence should always be assumed instead of malice, they always leave off the very important qualifier, "adequately"?
First, I'm not an apologist for anybody, jackass. I'm a cynic. I've just known enough morons to realize that stupidity truly has no bounds. See below.
Can all the gigantic, mind blowing holes in Diebold's software be ADEQUATELY explained by incompetence?
Absolutely. Right now, I'm part of a multi-company team on a government contract. One of the performers (over which I have no control) is creating a disastrously mangled codebase that does nothing but pull data from a database and make it available over a network. This has taken these morons about 8 months, cost a fortune, and currently requires 1.3GB of memory (!) to run. It has no more capability than about 100 lines of C++ code interfacing with MySQL, and requires a stack of about 10 different products that are constantly breaking. Compared to these fuckwits, Diebold is a bunch of geniuses. And this is just a minor example that I'm personally familiar with.
Anybody who says this couldn't possibly happen by accident hasn't worked in government contracting. Trust me, buddy...this is nothing. Go look up SAIC's bungled attempt to provide the FBI with modern software, which was scrapped after $200M because it would have been cheaper to start from scratch than make it work. I could go on for days with colossal disasters in government acquisition.
Not in my opinion. YMMV.
Then you're lucky to have never been party to such a disaster. Because I've personally seen and can cite examples of far worse. It usually involves government. If you ever want to see fuckups that are well beyond what normal people would think are conceivable, get into government contracting. Not to say that all government work is bad (I work for a contractor), but sometimes big contracts go to companies that can't execute them, and there's just not much oversight.
So can this be explained without nefarious conspiracy theories? Yeah. It can. Incompetence is more than sufficient to explain this exact behavior. Which is why proof of the malice is required, because stupidity is so utterly ubiquitous that it effectively forms the cynic's version of Occam's Razor, which is the quotation I originally cited.
Come on! How hard is it to pop some triggers on the DB so that any change whatsoever results in the current record being written to the audit trail? Really, how hard is that?
And haven't these folks heard of logical deletes instead of actually deleting it? Use a delete flag, folks! I find it amazing that such concepts are strict requirements for simple things like clinical trial systems, and regulated heavily and audited regularly by the FDA, but our voting system has no such regs or audits.
In any event, this stuff is not a case of the system messing up randomly or just working badly. Diebold has marketed their system as being capable of producing secure, accurate election results, and is nowhere near that level of competence.
Given the brutal simplicity of what is required, I see no reason why they shouldn't be held to the same level of responsibility as someone who is making a heart monitor, or even an electronic slot machine.
An, yes, you're right. You're not an apologist for anybody, you're an apologist for someone very, very specific -- Diebold.
Free Software: Like love, it grows best when given away.