New Closed Source Voting Systems Malfunction

LowellPorter writes "Miami-Dade and Broward counties are having voting problems. After the 2000 election problems, new voting methods were installed including touch screen technology. Some times the problems were with workers not showing up, poor training, or mechanical problems. It doesn't look like they cleaned up the system there." Not all of the problems mentioned in the article are due to the new proprietary voting machines, but many of them are.

Comments

This doesn't make sense...

[jsonmez]

How hard is it to make a voting machine that works? All it does is count votes, it's not like it does rocket science!

Re:This doesn't make sense...

[Zathrus]

If it's so amazingly simple, feel free to go ahead and make it!

Don't forget that it has to be insanely easy to use, with extremely clear instructions and adaptable to an infinite number of configurations for different polling uses.

It also has to be capable of verifying that the voter is valid, that the ballot is valid, and that it itself is valid (i.e. - has not been tampered with).

You either need secured local storage or a secure connection to another storage facility (whether that be onsite or offsite). If local, you need to make sure the data won't be destroyed by substandard handling (see below).

It must be rugged, portable, easy to setup, and low cost.

It must save state, so that if it does crash the current voter doesn't get screwed. Optimally they should be able to go to another voting machine to finish their ballot.

Let us know when you've finished. Don't forget this is both a hardware and software solution. We'll be waiting.

Re:This doesn't make sense...

[Zathrus]

No, you've ignored 90% of the issues.

It does have to validate itself, period. If you don't then a hack of the system can invalidate every vote cast on it. Worse, it can fake the votes in such a way that you wouldn't be able to tell that they were invalid.

And you totally ignored the UI, ruggedness, and flexibility issues.

Not to mention that there appeared to be little or no problems with the voting machines. According to the article the problems were human ones.

Re:This doesn't make sense...

[CoughDropAddict]

If the problem is so difficult, how is it that ATMs (which target an extremely similar problem) have been in widespread use for years?

Re:This doesn't make sense...

[anthony_dipierro]

What's wrong with the system that works so well and reliably in other places: using a pen, put a big cross in the box next to the person/answer being voted for. Any other marks invalidates the ballot slip.

One thing that's wrong with that is it makes it ridiculously simple to invalidate ballots after they are cast.

Re:This doesn't make sense...

[david.given]

If it's so amazingly simple, feel free to go ahead and make it!

Okay. Let's see. Let's use cardboard voting slips, with pencils.

• Intuitive --- check. Everyone knows how to make marks on paper. Put an X in the box next to the name of your candidate. Because the slips aren't machine-read, you don't have to cater to the machines, so you can put the boxes next to the candidate's name.
• Self-validating --- check. The human who counts the voting slip can tell at a glance if it's been filled out correctly. You can verify that each person has voted only once by keeping a list of people eligable to vote at each polling station, and crossing their names off the list.
• Secure storage --- check. Well, not quite. You need a box with a padlock on it and a slit in the lid, and a seal to indicate tampering. Seals are old technology. Trivial money from your local hardware store.
• Rugged, portable, low cost --- check. Carboard is cheap. Printed cardboard is cheap. It's also tough; if you use pencils rather than pens, then even if a filled box is submersed in water, the votes can still be read (carefully).
• Saves state --- check. Each vote is physically encoded on to a piece of cardboard, which can be counted as many times as is necessary to get it right. Won't crash. Ever.

So: we have a cheap, simple, secure voting system that's hard to spoof (with adequate physical security), easy to use, and with excellent accountability. You still need to count the votes, of course; here in the UK, we use volunteer bank tellers, who are really good at this sort of thing. The system scales really well, because each voting region is sized by population, and each area has about the same ratio of bank tellers to non bank tellers. The votes are counted in O(n) where n is the number of votes in the largest area, and then the results are phoned in to a central location.

And it all still works if the power goes out.

Why do we need machines, again?

What's with the headline?

[anthony_dipierro]

Would any of these problems be solved with an open source solution? Do these problems have anything at all to do with the fact that the solution is closed source? Is the fact that these systems are closed source ironic, or telling in any way?

Your headline is about as biased as "Microsoft User Commits Murder"

Journalism By Telepaths?

[reallocate]

1. The article doesn't tell us that the software is proprietary. Nor does it tell us that most of the problems are due to the use of closed software. Anyone wondering if Slashdot is an example of journalism or just a bunch of poseurs-for-hire tossing words around need look no further.

2. So anyway, why would we expect open source to work any better?

Oh, yeah, open source v1's are SO much better

[Brento]

Have you *ever* seen v1 of a system work flawlessly? It's so immature to toss the word "proprietary" in here, as if to insinuate that being open-source would fix anything. There's tons of open-source programs in v1 status with bugs. Anybody see any news headlines when Mozilla 1.0 came out and there were bugs in it? No? End of story.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Openness is critical in insuring fair elections

[FreeUser]

Would any of these problems be solved with an open source solution? Do these problems have anything at all to do with the fact that the solution is closed source? Is the fact that these systems are closed source ironic, or telling in any way?

Yes, yes, and yes, it is telling.

Openness is absolutely critical to fair and free elections, and that applies to the technology as much as it does the people. Who knows what is being done with the data being collected, or how it is being massaged. Is every electronic vote counted? Do we know that the results being reported are accurate, or whether or not a systemic flaw (or deliberate alteration) in the software is causing every Nth republican or democratic vote to be dropped? No, we don't know this, because the software's source code is unavailable for public review, much less peer review.

There are all kinds of Microsoft apologists (not saying you are one, but the vast majority of posts taking a tone similiar to yours are, as evidenced by their posting histories) quick to point out that having untrained election officials has nothing to do with the closed source nature of the software, yet eagerly glossing over the profoundly obvious fact that if the election software is closed source, no amount of training can insure that the software is unbiased and the election results fair.

So the point is relevant, even if it does rub the closed source advocates and Microsoft zealots the wrong way.

Re:Openness is critical in insuring fair elections

[FreeUser]

I think your paranoia is a little over the top, friend.

I live in Chicago. It isn't paranoia, it is experience. Election fraud is an ugly, real part of life, and any system that is open to abuse will be abused by one party or the other.

Closed source, proprietary systems whose inner workings are not open to public review and peer review are vastly more susceptable to this sort of thing than open, easilly examined and proofed source code are.

Why not use both computers and paper

[jonin]

Wouldn't it be easier to create a system that allows people to vote using a touch screen system, then instead of trying to save the data (which could easily be corrupted or lost,) simply print the ballot. It could print two copies, one for the person to take home as a personal voting record and one that they put in the polling box. The ballot that is given to the state could even be encoded so that it cannot be read without a computer in order to maintain privacy, while the copy the person keeps could be in plain english, spanish, or whatever language, so that they can confirm who they voted for. I think this would be a reasonable option that would be easier to setup (maybe.) If anything goes wrong with the machine, send them to another machine and start fresh.

Re:voting machines are stupid

[RobinH]

Being Canadian, I had the same initial reaction as you, but you don't understand the U.S. voting system. Voting in the U.S. is like filling out a tax form... it just goes on forever and ever. You have to vote for everything at once - local, state, etc., and there seem to be "sub votes" within them. Not to mention, they elect their judges and senate.

I know what you're feeling though - in Canada we use a pencil and a piece of paper. There are, say, 5 local candidates, each with a little box beside their name and you vote for one of those five by putting an 'X' in the box. However, we don't vote separately for the leader of the country, since we have a parliamentary system. Plus, our senate and judges are appointed by elected officials.

It does make sense though, with such low minimum wages, that the U.S. could afford to just hire people off the street to hand count ballots.

As for storing them electronically, that's a bad idea. Most people would not (and should not) accept various decaying charges on a vast bank of capacitors (I'm speaking of DRAM, of course), or magnetic deviations on a disk as proof of someone's vote. There really needs to be physical evidence that a real person can see and verify, or else the system can too easily be corrupted. ... all of a sudden I feel like I should go and get all my money out of the bank!

fraud writ large

[aminorex]

why anyone would *believe* the results tabulated
by software that was immune to public audit gathered
from complex and bug-prone devices operated by a
secret mechanism is beyond my comprehension.

given the history of democratic elections around the
world and in the united states itself, it seems
more than apparent that such devices, if they
continue in use, will inevitably result in massive
electoral fraud.