Just How "Complex" Is Diebold's Voting Software?

[stopbush]

Over the past week, more than a few talking heads brought up the subject of Diebold et al's proprietary software,
and the fact that the software programs are not open to examination. The rationale I've heard for this is, "Diebold
has invested millions of dollars in developing these programs, so, naturally, they don't want others to examine their
programming because it could be copied."

I'm thinking that that is a huge red herring. Just how complex is their software? Hell, we all take online polls every day,
and some - like Zogby - ask tens of questions. I'm assuming most of these polls are developed on off-the-shelf
software packages that cost next to nothing to purchase and take a few minutes to set up. Yet I've never seen an
online poll that crashed & burned or froze up in the middle of voting like a Diebold machine.

Even here in California, we didn't have that many candidates/issues to vote on, not if you compare the slate to a typical
Zogby poll. The issues are all yes/no votes and the candidate slates have maybe 8 choices per office, max. Just how
complex does voting software need to be to handle what (on the face of it) looks to be a relatively uncomplicated
procedure? It's not like we're asking the voting machine computers to multitask at the level that most of us do
every day on our home computers.

Am I missing something?

[kurth]

1. How hard is it to add Republican votes and

subtract Democratic votes?

[Junkdrawer]

2. The GEMS database is far more complex than it needs to be...

And I think that the complexity of the design was done on purpose to allow fraud.

[high density]

6. Does GEMS still use Microsoft Access?

It's kind of unforgivable that a company would even consider using Microsoft Access in vote tabulating software. I wouldn't trust a high school class vote to an Access database, let alone use it in a national election.

[Junkdrawer]

7. As far as I know. n/t

[ThomCat]

3. The only thing that would make their software complex

is the built in security, and we already know there isn't any.
x(

[sinkingfeeling]

4. A project that any first year IT student could do. Not complex unless

there's other secret things built into the code to 'throw' the elections.

[SpreadItAround]

5. GEMS uses Microsoft Access

As a database administrator I am shocked that they use that piece of shit.

I understand why they use it; portability, ease of use, ease of installation, etc., etc.

Since you asked, if I were to design a vote tabulation system the back-end database would be SQL Server or Oracle with a front-end designed in C or Visual Basic.

I'd then install both on a dedicated server in each voting district and that dedicated server would have no access to the internet until it's time to certify the vote totals.

[Recursion]

10. Disagree 100%

Since you asked, if I were to design a vote tabulation system the back-end database would be SQL Server or Oracle with a front-end designed in C or Visual Basic.

Hell no. No closed-source software in voting machines. 0.

Give me an OpenBSD platform, as much of the GNU userland and Xorg as you need to run the software, and a postgres database, compiled in front of witnesses after validating that the source tarballs were signed by the development team, and were the same tarballs as previously audited for security and correctness.

[SpreadItAround]

12. I like your solution much more than mine

But can we both agree that using Access is an abomination?

[Atman]

8. You're not missing anything at all.

A bunch of tablet PC's tied into off-the-shelf databases, running on one the most notoriously insecure platforms in computer-dom, Windows (CE in this case, I believe, but I'm quite possibly wrong. Wasn't it described in "Hack The Vote?"). With a big fat government contract from your bud in DC, a thousand bucks worth of crappy Chinese components becomes a several-thousand-dollar gold mine. And data mine, for that matte.

As those Guinness guys say...BRILLIANT!

.

[Recursion]

9. Hmm... yeah IBM is really hurting, isn't it?

I mean, they mostly just do open source stuff now, and it's really killing their bottom line
:sarcasm:

[moggie]

11. I'm typing this into an open-source browser

...running on an open-source operating system. It'll be posted to an open-source webserver via an open-source proxy, with the help of an open-source DNS server. Millions of lines of code doing jobs considerably more complex than counting votes, and all openly available for anyone to examine: and all free, gratis, for nothing. So if Diebold et al had to open their code, it might eat into their profits? Even if that's true, boo fucking hoo! Who says they have a right to profit from the most fundamental process underlying democracy?

[TechBear_Seattle]

13. I've written election tabulating software

Crimeny, that was (mumble) years ago in high school, using Quick Basic on a PET computer. It ain't that hard.

[moggie]

14. Here's how I vote

I'm in the UK, so the process is a little different:

I go to the polling place, and show my voter's card to one of the poll workers. They check me off a list, and give me a ballot paper. I go to a booth, put a cross in the appropriate box using a high-tech device known as a pencil (fitted with a sophisticated anti-theft device known as a piece of string), fold it in half, and put it through the slot in the ballot box, which is in full view of the poll workers in the middle of the room. The votes are manually counted, of course.

I've never experienced a faulty pencil. I've never had to queue to vote: the booths are just pieces of plywood nailed together, so there are always enough of them. In the unlikely event that the power went off, someone would produce a torch or a candle, and voting would continue.

Of course, this process is not immune to fraud, but it's more transparent than machine-based voting. It's how you used to vote, too. What went wrong?

[stopbush]

15. Yep. Our system has been hijacked.

I love it when people defend voting machines v paper ballots, positing that manually counting all of those paper
ballots would take too long...we then spend the next 2 months in the courts deciding which votes will get
counted and another 2 weeks finding out who "won."

The people of the USA would prefer paper ballots. The only people who like the machines are the Rs who use them
to fix elections and the media who want to call the elections within one minute of the polls closing (of course, the
only way they can do this is through the exit polling which they've worked mightily to disparage as the exit polling
hasn't been going the Rs way the last couple of elections).