Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Diebold Met with 'Electronic Civil Disobedience'

By quixotic1 in Technology
Thu Oct 23, 2003 at 07:55:23 AM EST
Tags: Politics (all tags)
Politics

Saying they are defending the right to a fair election, two student groups, Why War? and the Swarthmore Coalition for the Digital Commons, are rejecting Diebold Elections Systems' cease and desist orders and are initiating a campaign of electronic civil disobedience that will ensure permanent public access to the controversial leaked memos.


As a member of Why War?, I wanted to outline our campaign here -- and invite others to join us.

Earlier this week, the Electronic Frontier Foundation announced that it will defend the right of Online Privacy Group, the Internet service provider for San Francisco Indymedia, to host links to the controversial memos. Going one step further, Why War? and SCDC members are the first to publicly refuse to comply with Diebold's cease and desist order by continually providing access to the documents.

The documents, from Diebold Elections Systems, a company in charge of the electronic voting machines in 37 states, prove that the company knowingly produced an electronic election system that contained absolutely no security against voter fraud. In fact, the lead engineer from Diebold wrote over two years ago that anyone could change votes without leaving a trail: "Right now you can open GEMS' .mdb file with MS-Access, and alter its contents. That includes the audit log." GEMS stands for Global Election Management System and is the central computer in each county on which the votes are stored after the election.

Diebold has filed cease and desist orders against anyone who has attempted to share these memos with the public. They have taken down hosts all over the world, including the personal website of the very journalist who broke this story, Bev Harris. Why War? and SCDC refuse to comply. We cannot allow the suppression of evidence that proves a Diebold machine registered 16,022 negative votes for Al Gore in Precinct 216 in Florida in the 2000 presidential election. We cannot comply with a company whose CEO has given $9,965 to Bush and the Republican National State Elections Committee since 2001, while declaring that he is "committed to helping Ohio deliver its electoral votes to the President next year."

Through active electronic civil disobedience we can bring to light the usually silent act of suppression. The result will be a permanent and public mirror of the memos -- documents whose public existence strengthens democracy.

One journalist in Seattle has written that Dean Logan, director of records, elections and licensing services in Seattle, "decided election security was a 'legitimate issue' after internal company e-mail was posted on the Internet and discussed in a Salon.com article Monday." The goal is to force these documents back into the sunlight.

Logan should be alarmed -- the depth of Diebold's deceit extends far beyond what most Americans are comfortable believing. In fact, there are already allegations that Diebold was responsible for the highly questionable results from the 2002 election in Georgia. Andrew Gumbel writes in the Independent:

Something very odd happened in the mid-term elections in Georgia last November. On the eve of the vote, opinion polls showed Roy Barnes, the incumbent Democratic governor, leading by between nine and 11 points. In a somewhat closer, keenly watched Senate race, polls indicated that Max Cleland, the popular Democrat up for re-election, was ahead by two to five points against his Republican challenger, Saxby Chambliss.
Those figures were more or less what political experts would have expected in state with a long tradition of electing Democrats to statewide office. But then the results came in, and all of Georgia appeared to have been turned upside down. Barnes lost the governorship to the Republican, Sonny Perdue, 46 per cent to 51 per cent, a swing of as much as 16 percentage points from the last opinion polls. Cleland lost to Chambliss 46 per cent to 53, a last-minute swing of 9 to 12 points

Why War? believes that the mere possibility that the core principle of democracy -- a fair election -- is under attack demands action. Though we are concerned with Diebold's unmitigated support of the Republican party, this issue should concern Americans across the political spectrum. Furthermore, when such action is met with legal threats, we believe the conscientious path is to engage in open and democratic electronic civil disobedience.

Diebold's use of the Digital Millennium Copyright Act (DMCA) in order to suppress information critical to the public welfare is abhorrent. As Bev Harris so succinctly puts it:

I don't believe you can protect intent to break the law by slapping a copyright on it. And the memos that we posted show that the law has been broken. If you can protect intent to break the law, all anybody would need to do is take their bank robbery plans and put a copyright on it, and then say nobody can look at them because they're copyrighted.

Diebold broke federal law by applying patches to Georgia election machines without having them certified (because it would have taken too long and made them look bad). They broke civil law by misrepresenting their software to state and local governments contracting them to count votes. By continuing to avoid fixing problems they knew were in existence -- problems that jeopardized the entire process of voting -- they have lost any ethical claim to these documents.

More importantly, however, the process of democratic voting should be one of transparency (see, for instance, the Slashdot discussion), not of back-room deals and uncertain security using a for-profit company. American voters are, quite literally, being robbed of their right to a free and fair election.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o Slashdot
o electronic civil disobedience
o permanent public access
o announced
o San Francisco Indymedia
o continuall y providing access
o Diebold Elections Systems
o Right now you can open GEMS' .mdb file with MS-Access, and alter its contents. That includes the audit log.
o Global Election Management System
o Bev Harris
o 16,022 negative votes for Al Gore
o has given $9,965
o committed to helping Ohio deliver its electoral votes to the President next year
o permanent and public mirror of the memos
o journalist in Seattle has written
o Salon.com article
o writes
o Slashdot discussion
o Also by quixotic1


Display: Sort:
Diebold Met with 'Electronic Civil Disobedience' | 243 comments (231 topical, 12 editorial, 1 hidden)
Slashdot Dupe (2.16 / 12) (#1)
by RyoCokey on Tue Oct 21, 2003 at 11:39:25 PM EST

Seriously, this is front page Slashdot. Although I suppose we'd probably have a better discussion on the matter. After reading through the memos, though, I have some questions.

What's so controversal about them? There doesn't appear to be any wrong-doing on Diebold's part from what I understood. The thing about storing the results in a standard Access data file makes them look incompetant and retarded, however, I'm not convinced it wouldn't have come to light in the first place.

Some of the assertions this article makes seem dead wrong. It doesn't appear Diebold has broken any laws, or that there wasn't evidence of obstruction. It just looks like someone sent them the wrong version.

Worth noting is that I read a Slashdot repost of the memo, not the "original" document. If there is evidence within those for wrongdoing, please post. Indeed, such quotes should be added to the article.



farmers don't break into our houses at night, steal our DVDs and piss on the floor. No
It all just looks a bit dodgy. (2.80 / 5) (#23)
by squigly on Wed Oct 22, 2003 at 03:38:39 PM EST

This wholke paragraph just looks suspicious:
Being able to end-run the database has admittedly got people out of a bind though. Jane (I think it was Jane) did some fancy footwork on the .mdb file in Gaston recently. I know our dealers do it. King County is famous for it. That's why we've never put a password on the file before.

It might be essentially harmless.  What does "end-run" mean anyway?  However, it does show that they have accessed the system.  This should be a huge no-no even if they don't change anything.  

The rest of the memos just make the whole system look ad-hoc and incompetent.  Not neccesarily criminal, but certainly a good reason not to want to use them to provide any software for anything imporant.

[ Parent ]

No Wrong-doing? Please Re-read Memos (2.40 / 5) (#72)
by cmholm on Thu Oct 23, 2003 at 02:45:47 PM EST

At the moment, Diebold's "wrong-doing" is really more of a contractural nature. They delivered voting machines as production quality systems, when the memos make it clear that the software isn't ready for primetime.

Shipping beta-quality software and following up with n number of unadvertised patches may work for word processors, but not for voting systems. In mission critical systems it is usually required that patches and system mods go through a change review process that includes the customer. The memos make it look like Diebold blew this off.

Since the performance of voting systems is an inherently public matter, Diebold's "wrong-doing" starts to become both political and criminal when they try to cover up their contractual screw ups.

[ Parent ]

Burn it! + Recount Trivia (2.80 / 20) (#2)
by Blarney on Tue Oct 21, 2003 at 11:49:54 PM EST

Once a complete archive is made, it'd be a great idea to burn 535 attractively-labeled and packaged CDs and send one to each United States Congressman. Some especially disobedient person could provide his name for the return address - somebody who is willing to actually go to jail for such a cause. Who knows? Perhaps someone in a position to do something about it would do something about it.

It's important to keep the issues clear, though - the Diebold debate is mainly about the ability of the machine's operators to deliberately manipulate vote counts, not so much about inaccuracies of the machine. Every voting method has inaccuracy. The way you mention the Diebold machine that voted -19200 times for Gore gives the impression that this affected the election - when it really only affected the projections. Still, this sort of error probably isn't unheard of in a general election - perhaps not tens of thousands, but certainly dozens and hundreds per precinct may well be expected.

The idea that Gore could win an election based on a fair recount is somewhat deceptive. The fact is, although elections can come down to within a fraction of a percent, a typical precinct does not normally need to keep that level of accuracy. It is expected that a few percent of votes get mangled, get incorrectly filled out by a confused voter, get lost in transport, and so on. However, these errors have been masked by the Law of Large Numbers - the statistical principle that, when multiple observations of the same phenomenon are combined, the uncertainty of the result decreases. If there are N precincts of roughly equal size, each with Y% error, the final error will be about Y%/sqrt(N). This is still finite - if you have 100 precincts, the error is decreased by 10%, but it'll still be there. A recount would likely produce an error on the same order.

The honest fact about the 2000 election is that, given typical uncertainties, nobody knows whether Bush or Gore won.

To me, this isn't about errors so much, but about the proven potential of Diebold to steal elections, their behavior which does not resemble the behavior of an innocent organization, and the non-neutrality of their owner.

Giving them to Congress (2.77 / 9) (#6)
by jjayson on Wed Oct 22, 2003 at 03:07:42 AM EST

If somebody gave me burned CDs, in January or Februrary, I could personally hand them to every person on the Hill. I, however, will not attach my name to anything that implicates election rigging, disparages any specific elected official, or supports some radical goal I am not confortable with. But if things are mellow and factual, I could do it.
_______
Smile =)
A determined United Europe army would probably get their butts kicked by an LA street gang! — Parent ]
What if (3.00 / 5) (#22)
by Happy Monkey on Wed Oct 22, 2003 at 01:27:15 PM EST

What if it can't be factual without implicating election rigging?
___
Length 17, Width 3
[ Parent ]
I don't understand (none / 3) (#30)
by jjayson on Thu Oct 23, 2003 at 01:31:14 AM EST

I can't imagine that scenario. The two issues seem separable to me.
_______
Smile =)
A determined United Europe army would probably get their butts kicked by an LA street gang! — Parent ]
For example, (none / 1) (#78)
by Happy Monkey on Thu Oct 23, 2003 at 04:23:56 PM EST

if the facts themselves implicate election rigging.
___
Length 17, Width 3
[ Parent ]
I think he means (3.00 / 4) (#34)
by aphrael on Thu Oct 23, 2003 at 04:03:51 AM EST

He's fine with stuff that demonstrates that it's *possible* for elections *to be rigged*, assuming that's factual; he won't take up stuff which alleges that elections *have been rigged*.

[ Parent ]
hmm... (2.66 / 6) (#19)
by Run4YourLives on Wed Oct 22, 2003 at 12:31:19 PM EST

The fact is, although elections can come down to within a fraction of a percent, a typical precinct does not normally need to keep that level of accuracy

This entire paragraph is as scary as shit. I understand how it "makes sense" from an electoral college perspective, but it's ridiculously scary nonetheless.

Up here in Canada, with a parlimentary system, such a wishy washy way of handling votes, yes every vote would not be tolerated. In some ridings each election, the margin of victory is well under three digits.

I'm interested to know if you're aware of such standards being applied to other elections, either within or without the US.


It's slightly Japanese, but without all of that fanatical devotion to the workplace. - CheeseburgerBrown
[ Parent ]

What he means (none / 2) (#99)
by JackStraw on Fri Oct 24, 2003 at 02:00:50 AM EST

and I agree, is that individual precincts don't need  to count their votes very precisely. If each of maybe 20 precincts is off (randomly) by, say, 1%, the sum of all their votes would only be off by a small fraction of a percent. This difference, considering the HUGE amount of people who don't vote or are misinformed or punch the wrong hole or whatever, is insignificant... to simply count the votes with a known, small uncertainty and just *take the result* without re-counting or second guessing is very, /very/ close to counting perfectly. It's not really anything to do with the electoral collage. Only a systematic shift... i.e., someone puts in 3000 fake votes for Gore, is significant.
-The bus came by, I got on... that's when it all began.
[ Parent ]
elected by a hairline (none / 2) (#150)
by TomV on Sat Oct 25, 2003 at 06:49:52 AM EST

One example from the UK is the result in the Winchester seat at the 1997 General Election.

The original count showed the Liberal Democrat candidate, Mark Oaten, beating the Conservative candidate, Gerry Malone, by two votes from 62,054 votes cast. Mr Malone petitioned for a by-election on the grounds that 55 ballot papers deemed to have been spoiled seemed to be votes for him. The court ordered a by-election. as it happens, in the by-election Mr Oaten won by a margin of 21,556.

Mr Oaten's original two-vote majority is in equal first place for the tightest result in UK Parliamentary Election history with Ilkeston 1931, followed closely by majorities of three in Leeds West (1924), Tiverton (1923), Peterborough 1966 and Carmarthen 1974 (Feb election). We try to count every vote, and to recount as necessary until all the candidates are prepared to accept the result.

There's a list of close UK results here. I'm not sure of the historical constituency sizes, but at present the average is around 77,000 in England.



[ Parent ]
-1, remove this bit: (2.64 / 14) (#3)
by Kasreyn on Wed Oct 22, 2003 at 12:22:23 AM EST

In fact, there are already allegations that Diebold was responsible for the highly questionable results from the 2002 election in Georgia... "Something very odd happened in the mid-term elections in Georgia... ... ... ... ...Cleland lost to Chambliss 46 per cent to 53, a last-minute swing of 9 to 12 points."

Unless you actually INCLUDE these "allegations" in your story, they are not allegations. They are merely alleged allegations. Furthermore, you do not provide the slightest evidence, even of people manipulating the vote using a flaw in Diebold's system, MUCH LESS evidence of Diebold itself being culpable of rigging the election, or of deliberately leaving the system vulnerable to manipulation (that is, deliberate culpable inaction).

As such, that entire segment looks like nothing so much as paranoid conspiracy kook garbage. Remove it, or suffer my -1 when this hits voting.


-Kasreyn

P.S. as to Diehard's actions: for their actions PREVIOUS to the memos coming to light, they are merely guilty of such dunderheadedness that they do not deserve such an important contract as providing the U.S.A. with election machines. However, attempting to hide the fact that the U.S.A.'s election system's security is fatally flawed... to me that sounds like treason.


"Extenuating circumstance to be mentioned on Judgement Day:
We never asked to be born in the first place."

R.I.P. Kurt. You will be missed.
Evidence before removal (none / 2) (#4)
by Daelin on Wed Oct 22, 2003 at 12:37:16 AM EST

I'd like to see strong evidence and reference supporting the allegations, with Freenet links if managable. If the only evidence collectable is weak, I agree that it should be removed.

[ Parent ]
You won't get it. (none / 3) (#115)
by IPFreely on Fri Oct 24, 2003 at 01:25:00 PM EST

After the Georga election, the machines used were erased, all records deleted, no paper trail exists. There is no evidence at all either way.

You'll never be able to prove anything wrong was done. On the other hand, you'll never be able to prove that nothing wrong was done either.

In my mind, if you can't prove that a system worked properly, then you can't trust it. Unfortunately, most people seem to think that if you can't prove it's broken then you should trust it.

[ Parent ]

motive, plenty of opportunity (nt) (none / 2) (#11)
by phred on Wed Oct 22, 2003 at 09:05:48 AM EST



[ Parent ]
umm (none / 3) (#47)
by Wah on Thu Oct 23, 2003 at 12:03:42 PM EST

Furthermore, you do not provide the slightest evidence, even of people manipulating the vote using a flaw in Diebold's system, MUCH LESS evidence of Diebold itself being culpable of rigging the election, or of deliberately leaving the system vulnerable to manipulation (that is, deliberate culpable inaction).

The 'deliberate culpable inaction' is demonstrated in the continued use of system that is known to be insecure.  The leaked emails demonstrate that huge security holes were well known.  This seems to be illegal on the face of it.

The accusations are just that, and would be the smoking end of what seems to be a high-powered election assualt rifle.  Release the audits and do a manual recount....if such things are even possible.

I agree that the 'sky is falling' tone is a bit over the top, but there does seem to be something to this.  

p.s. 'Diehard' is a movie franchise.
--
kewpie
[ Parent ]

alleged allegation? (none / 1) (#64)
by el_guapo on Thu Oct 23, 2003 at 02:16:19 PM EST

sorry. i must now question your mastery of english. "alleged allegation" is quite the redunadant statement, AND it repeats itself :-P
mas cerveza, por favor mirrors, manifestos, etc.
[ Parent ]
it was meant ironically -nt (none / 1) (#97)
by Kasreyn on Thu Oct 23, 2003 at 11:54:40 PM EST

no text
"Extenuating circumstance to be mentioned on Judgement Day:
We never asked to be born in the first place."

R.I.P. Kurt. You will be missed.
[ Parent ]
this is great, i hadnt heard about it before (1.00 / 12) (#7)
by Please grow up on Wed Oct 22, 2003 at 06:07:32 AM EST

thanks for posting it here, this is why i read kuroshin to get the news i cant get anywhere else
(my user name should be "Please be Nice", dont pick your user name when your feeling annoyed!!)
man, you suck (nt) (1.75 / 4) (#18)
by Run4YourLives on Wed Oct 22, 2003 at 12:24:12 PM EST



It's slightly Japanese, but without all of that fanatical devotion to the workplace. - CheeseburgerBrown
[ Parent ]
sorry, i was just trying to be polite (1.50 / 4) (#21)
by Please grow up on Wed Oct 22, 2003 at 01:22:58 PM EST

maybe you got out of bed the wrong side this morning??
(my user name should be "Please be Nice", dont pick your user name when your feeling annoyed!!)
[ Parent ]
They must be evil (1.57 / 7) (#8)
by daragh on Wed Oct 22, 2003 at 06:36:40 AM EST

Their name contains "Die" (meaning "die") and "bold" (meaning "bad").

No work.

really? (none / 2) (#12)
by banffbug on Wed Oct 22, 2003 at 09:20:47 AM EST

to me it sounds more like "death" and "proud/courageous"

"A strong death" ...

[ Parent ]

from the simpsons (2.40 / 5) (#14)
by zephc on Wed Oct 22, 2003 at 10:24:47 AM EST

Lawyer: "But what about that tattoo on your chest? Doesn't it say, 'Die Bart, Die?'"

Bob: "No, that's German for 'The Bart, The.'"

Officer: No one who speaks German could be an evil man.

(episode 9F22, "Cape Feare")

[ Parent ]

It's "The Beard, The" (nt) (none / 2) (#31)
by scruffyMark on Thu Oct 23, 2003 at 01:45:40 AM EST



[ Parent ]
Wrong gender (none / 1) (#55)
by flo on Thu Oct 23, 2003 at 01:09:13 PM EST

"die" is used for feminine nouns, but "Bart" (beard) is masculine
---------
"Look upon my works, ye mighty, and despair!"
[ Parent ]
Massive Digression: (none / 0) (#66)
by NateTG on Thu Oct 23, 2003 at 02:18:45 PM EST

Right, but a word-for-word translation would still be "the beard, the."
Alternatively, Bay Area Rapid Transit = bart is femminine in German.  (No, I have no clue why, but it is.)  If you translate "the B.A.R.T., the" you would get "die Bart, die."

[ Parent ]
No I tell you! (none / 0) (#236)
by trezor on Thu Oct 30, 2003 at 08:21:21 AM EST

It's short for Diabolic!


--
Richard Dean Anderson porn? - Now spread the news

[ Parent ]
let me ask you this: (1.05 / 18) (#13)
by rmg on Wed Oct 22, 2003 at 09:55:16 AM EST

uh... hey... uh... you got any gum?

_____ intellectual tiddlywinks

+1FP (none / 3) (#27)
by LilDebbie on Wed Oct 22, 2003 at 07:10:18 PM EST

Move to vote!

My name is LilDebbie and I have a garden.
- hugin -

[ Parent ]
Freenet (2.77 / 18) (#16)
by cronian on Wed Oct 22, 2003 at 11:02:43 AM EST

These things really belong on Freenet. Has anyone put them on Freenet yet?

We perfect it; Congress kills it; They make it; We Import it; It must be anti-Americanism
Great Idea (1.00 / 13) (#17)
by sllort on Wed Oct 22, 2003 at 11:22:42 AM EST

That's how to get this story carried by the major press. Get these memos published on the Child Porn Network.

Rock on with your bad self.
--
Warning: On Lawn is a documented liar.
[ Parent ]

Child porn gets /way/ too much attention [nt] (none / 3) (#25)
by Daelin on Wed Oct 22, 2003 at 04:13:24 PM EST



[ Parent ]
From you, maybe. [n/t] (none / 3) (#82)
by RyoCokey on Thu Oct 23, 2003 at 05:12:07 PM EST



farmers don't break into our houses at night, steal our DVDs and piss on the floor. No
[
Parent ]
fyi (2.81 / 11) (#29)
by tiamat on Wed Oct 22, 2003 at 11:30:32 PM EST

I don't live in the USA. if someone needs a mirror for that stuff then they should email it to me. :)

The Diebold people (1.04 / 24) (#33)
by sellison on Thu Oct 23, 2003 at 03:23:04 AM EST

are good, moral Christians, friends of our president for goodness sake.

The socialists of the ELF should leave them alone, e-voting will be much better than the old way, just quit being so paranoid about it!


"No, I don't know that atheists should be considered as citizens, nor should they be considered as patriots. This is one nation under God."- George H.W. Bush

Sarcasm. Yes I know. (none / 0) (#235)
by trezor on Thu Oct 30, 2003 at 08:20:07 AM EST

But I wouldn't trust a person who believes that everything that is happening is part of some mysterious being's infinte good plan.

Even less people believing in the flaming-inferno-version of the "end of the world", sitting with a finger on the big button, ready to launch the nukes.

"Tell me when lord, tell me when! Let me be your instrument lord! Tell me when!"

But, I (boldy?) assume this is the common understanding we all have here.


--
Richard Dean Anderson porn? - Now spread the news

[ Parent ]
is this really true?? (1.83 / 6) (#36)
by Please grow up on Thu Oct 23, 2003 at 07:01:38 AM EST

because it seems to me like it would be awful easy to just scream CHEAT! CHEAT!! every time you lose an election so youd need a lot of evidence to say these things

the only thing you said that would be really bad is the bit about negative votes which if i understand it means that you say they took AWAY votes from al gore, but that link doesnt work, i wonder how many people will believe it anyway just because they want to???
(my user name should be "Please be Nice", dont pick your user name when your feeling annoyed!!)

Glitch led to 'Bush wins' call (none / 2) (#43)
by michaelp on Thu Oct 23, 2003 at 11:04:52 AM EST

Lowe ordered all of the precincts reviewed. At 1:24 a.m., the review showed that 412 of 585 registered voters in Precinct 216 had cast ballots - but that they had given 2,813 votes to Bush! Gore had a negative vote: minus 16,022. Ralph Nader's negative vote was even greater. The problem was traced to an error in the memory card.  

http://www.unc.edu/~pmeyer/usat29nov2000.html

It appeared to be fixed. But I have to wonder how an 'error in the memory card' could possibly lead to a negative number being recorded in Access???

AFAIK, memory cards statically hold data, they don't input it, no matter how erroneous they are.

So of course the ? is, were other 'memory cards' inputting smaller negative numbers that didn't get noticed?


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

>sigh< (none / 3) (#49)
by Skywise on Thu Oct 23, 2003 at 12:11:09 PM EST

First, there's nothing in the article that says the information on those memory cards was written there using Access.

Second, with a faulty memory card it's possible for random bits to get dropped or for the bus to output erroneous data or to scramble data on its way in.

Some memory cards will do error checking, cheaper ones will not.

Lastly, the bigger question (if you're really looking for a good conspiracy theory) is, why should the software allow for SIGNED numbers to begin with?  (When are you ever going to have negative results?)

[ Parent ]

Well, I read that the Diebold software stores its (none / 1) (#89)
by michaelp on Thu Oct 23, 2003 at 08:02:57 PM EST

data in an open mdb file in the article. Maybe these other machines use a more secure database, though your statement below suggests they don't (and you are OK with that???).

Second, with a faulty memory card it's possible for random bits to get dropped or for the bus to output erroneous data or to scramble data on its way in.

However, we aren't talking about "random bits" or "erroneous data" from the bus, --which is NOT going to be GOOD data to any database worth using for your CD collection--! I hope you get that 'erroneous data from the bus' is going to be in machine code, not C++ or even VB, right???!?!? That random machine code generated by a faulty bus might come up with one bit of data that looks ok to the OS and to the database I might be able to accept. That a bad bus might generate BYTES of random machine code that looks like good data to a database is a monkies and shakespeare level miracle.

There is NO WAY that scrambled machine code from a memory card should be able to write coherent data into a database. Seems to me that would be equivalent to bad RAM causing Word to start writing sentences that make sense by itself!

In any event, if the memory card failures are actually capable of adding in valid (according to the vote counting software) votes as they did for Bush, all the votes counted with the systems in question are tainted. This one was found because the failure was so gross, but the failure was not detected by the software! If the software can't tell when a memory card fault adds in thousands of votes while deleting tens of thousands, how in the world can it be trusted at all?

If each machine used only added a hundred votes and deleted a hundred votes, you'd have the 2000 election decided right there by only slightly less than perfect memory cards!


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

This can't be right... (none / 2) (#68)
by dipierro on Thu Oct 23, 2003 at 02:29:18 PM EST

Lowe ordered all of the precincts reviewed. At 1:24 a.m., the review showed that 412 of 585 registered voters in Precinct 216 had cast ballots - but that they had given 2,813 votes to Bush! Gore had a negative vote: minus 16,022. Ralph Nader's negative vote was even greater. The problem was traced to an error in the memory card.
The real vote in Precinct 216 was 22 for Bush and 193 for Gore. Nader got one.

So of 412 ballots cast, only 216 had a vote marked for President? That doesn't seem right.



[ Parent ]
Broken Link (none / 0) (#74)
by craigd on Thu Oct 23, 2003 at 03:14:44 PM EST

The page it does get tells you where to find the memos. Now read the excerpts on the page that sends you to. It's there, all right. The link is bad, but the content is real.


A man who says little is a man who speaks two syllables.
[ Parent ]
Suggestion... (2.73 / 19) (#37)
by skyknight on Thu Oct 23, 2003 at 07:34:15 AM EST

I think it's great that there are people making noise about this, lifting up the rug and showing that which has been furtively swept under it. I do, however, think that it is a big mistake to have a group such as "Why War?" spearhead this movement.

You have a cause here that is extremely important, and has practically universal appeal. Who wouldn't want a fair election process? Unfortunately, you've gone and wedged it under the umbrella of "Why War?", potentially alienating a large body of supporters.

People in anti-war groups are typically absolutely against war no matter what the situation. While I abhor war, I think it a necessary evil from time to time, and thus find my beliefs in opposition to most anti-war groups. They say "war is bad, so we should never engage in it", whereas I am inclined to say "war is indeed terrible, but let's do a utility calculation on a case by case basis when making a judgment". Thus, I am not apt to join such a group as I have no desire to be associated with this particular viewpoint, even if they are supporting other causes with which I agree.

Quite simply, you're making the scope of "Why War?" too broad, which is a tragedy when you're trying to attain critical mass for an issue that ostensibly anyone could support. I would suggest having a unique group under the title of, say, "Voters Defense Initiative". That way you would attract members irrespective of their political leanings on the issue of war, vastly swelling your ranks.



It's not much fun at the top. I envy the common people, their hearty meals and Bruce Springsteen and voting. --SIGNOR SPAGHETTI
So mirrior the documents (none / 3) (#60)
by cestmoi on Thu Oct 23, 2003 at 02:08:00 PM EST

I agree, Why War would not be my choice to host the documents. But the good thing is, you can download the documents, put up a website and host the memos yourself. Only problem is Why War so far is the only outfit that's willing to take Diebold's lawyers on. Are you?

I took the first step of downloading the docs but they're corrupted already. Anyone know where there's a decipherable zip version of the files?

[ Parent ]

They are not the only one. (none / 1) (#75)
by skyknight on Thu Oct 23, 2003 at 03:44:54 PM EST

You might want to check out this.

It's not much fun at the top. I envy the common people, their hearty meals and Bruce Springsteen and voting. --SIGNOR SPAGHETTI
[ Parent ]
I agree. (none / 3) (#69)
by Persecuted Telemarketer on Thu Oct 23, 2003 at 02:31:45 PM EST

This is an excellent point. I'm guessing though, as this gets more airplay (I can't imagine why it wouldn't, this is a Democratic presidental candidate's dream come true), it will come to be less associated with these two groups.

As you said, there may be nothing wrong with these two groups, but they obviously have certain political biases (even built into their names) and this would make people trust the documents less.

In fact, speaking for myself, I'm more skeptical of these docs because they're being hosted by this political group, and would myself be much more comfortable with them if they were from a more non-partisan group. Because knowing it has something to do with this group, it's not that unreasonable to think that they're blowing things out of proportion because of an anti-Bush bias...

The Yankees ARE the Evil Empire, you know. Go Marlins. Go anybody.
[ Parent ]

Thanks for raising awareness (2.12 / 8) (#39)
by gr00vey on Thu Oct 23, 2003 at 09:22:53 AM EST

Great article, and hopefully this stuff will start getting reported in the mainstream media, though I won't hold my breath....

What is this thing for? (2.58 / 12) (#40)
by 0xA on Thu Oct 23, 2003 at 10:01:09 AM EST

Okay I've brought this up before but nobody can ever answer the question. What the hell are these machines for? Do they really serve any purpose at all?

I am Canadian, I have never seen a voting machine. When I vote I am handed a piece of paper and a pencil. The paper looks like this:

| | Thief
| | Moron
| | Putz
| | Comunist

I make my mark upon the paper using my pencil like so:

| | Thief
| | Moron
| | Putz
|X| Comunist

Then I fold my paper in half and put it in a box. At the end of the day the poling station people open the box and sort the ballots according to the mark on them. Then somebody checks them and counts them.

Why couldn't this work in the US? Seems easier no?

It's all political... (1.40 / 5) (#46)
by Skywise on Thu Oct 23, 2003 at 11:58:28 AM EST

The Democrats made a brou-haha out of the Florida election by claiming that people in minority districts didn't understand how to properly "make an X" (punch a whole in a ballot) in the right spot because the ballots were too confusing and thus the whole election needed to be redone again.  (The Democrats had agreed to the ballot design)

Of course the Republicans didn't want the election done again (they won) and so forced the issue to recounting the existing ballots.  The Democrats then wanted subjective decision making on each and every ballot (Hmm... this guy voted for both the serial killer and the rapist for the same office...  Our man is the serial killer so he must've voted for the serial killer, because nobody would vote for a rapist)  and the whole thing led to international embarassment.  So laws were passed everywhere saying that paper ballots were inferior and the populace needed to use voting machines.  Because, y'know, technology makes things better.


[ Parent ]

Be a bit more fair (none / 1) (#111)
by Merc on Fri Oct 24, 2003 at 12:59:05 PM EST

The guy was asking about making an X on a piece of paper. You've got to admit that the florida ballots were more confusing than that.

Oh, and you forgot the most exciting part of the story. Once the counting made it seem like there was a chance they might lose, the Republicans decided this whole "counting the votes" bit was too big a gamble, so they had a court decide the president.



[ Parent ]
More fair still (none / 1) (#143)
by lordDogma on Fri Oct 24, 2003 at 09:38:47 PM EST

The votes WERE re-counted as required by law and Bush was shown to have won again albeit by a slimmer margin.

The following three things complicated the re-count:

1. Claims were made that many black people in Florida had their votes thrown out because a flawed mechanism was used to determine whether someone was a former felon or not (electronic name matching). I have no idea if this is true or not, but I'm sure we'll never get to the bottom of it. I'm inclined to think its a bunch of hot air myself.

2. One district with a lot of minorities failed to get their re-count done on time and therefore their votes were not counted. My opinion is simple: All other districts got their recounts done on time so why couldn't this one? If they really considered their votes so important then they should have gotten the recount done on time. Period. So screw 'em.

3. The Democrats kept calling for a hand re-count because they said the ballots were confusing and therefore needed some human intelligence in the loop to determine how voters meant to vote, vice how they actually voted. I'm sorry, but this is a sore loser's way to try to remake history.

In the end, the Supreme Court got tired of being pelted by lawyers on each side so they finally decided that the results of the original vote and the re-count were good enough for them and that ANOTHER recount would be over the top. So they said Bush won.

Assuming that a third re-count was done by hand; and assuming that every county got their votes in on time; and assuming that the "felons" issue was fixed, then there is a VERY REAL chance that Gore would have won (so they say). Pragmatically speaking though, Bush won the original count as well as the re-count and that's good enough for me. All this talk of Bush "stealing the election by calling on his friends in the Supreme Court" is bullshit. If a third recount was done by hand, then there would have been a lot of cheating in my opinion and the whole thing would be an even bigger fiasco.

For those still whining about the 2000 election: Get over it. Bush is the president. The next election is coming up. If you think Bush is so bad and your candidate is so good then lets duke it out at the 2004 election. All this whining gets us nowhere.

-- Lord Dogma

[ Parent ]

Technology can save effort (2.75 / 4) (#59)
by cestmoi on Thu Oct 23, 2003 at 02:02:03 PM EST

Humans are prone to a variety of counting errors. Some people transpose numbers quite easily so as they record a number, a 17 can easily morph into a 71. Some people write their 9's and 4's so they're indistinguishable. Some people see a 9 when they read a 6 and vice versa. I clearly recall one of my employees being thrilled when she used a spreadsheet for the first time to reconcile the company checkbook. A task she'd budgeted 4 hours for because that's what it took her by hand, was finished as soon as she verified which checks had cleared.

Technology, properly applied, can be a boon. On the other hand, technology misapplied as it is in Diebold's case, can be a curse.

Bottom line, no matter what system is used to count votes, it's going to have some errors. The goal is to choose a system that minimizes errors and makes it possible to correct the errors when they're identified.

[ Parent ]

Few counting errors (3.00 / 5) (#73)
by doconnor on Thu Oct 23, 2003 at 02:52:36 PM EST

There are some checks against those kind of counting errors. The number of ballots put into the box are counted as voting goes on. The total is compared to the number of ballots counted at the end. If there was a mistake made it will almost always be noticed by an inconsistancy in the counts.

Also, during the count each candidate can and do have observers watching every step and can point it out if they notice any problems.

The process takes only about an hour after the end of the election.

Technology can improve the counting speed, if there are problems, no-one is really there to notice it. With 2-6 people watching as every ballot is counted there is very little room for error.

The PEI election shows the how well the system works. Dispite a major hurricane and wide spead power outages, the election still went ahead without serious problems.

[ Parent ]

147 candidates = breakdown (none / 2) (#80)
by pjc50 on Thu Oct 23, 2003 at 04:53:43 PM EST

The ballots in California can be huge - there may be dozens of candidates or options for dozens of different questions.

[ Parent ]
Think. (none / 2) (#87)
by Kwil on Thu Oct 23, 2003 at 07:20:26 PM EST

Each question gets a separate ballot. We already do this here.
It's not such a difficult thing.

Beyond this, a simple ballot, names all on one side in alphabetical order, each with a line or arrow pointing to the box on the other side of the page doesn't even pose a problem with 147 candidates. Counters take each ballot, find the marked square, and follow the arrow back to the name. Takes all of what? 2 seconds if you're slow?

That Jesus Christ guy is getting some terrible lag... it took him 3 days to respawn! -NJ CoolBreeze


[ Parent ]
The Next President of the United States? (none / 1) (#112)
by Merc on Fri Oct 24, 2003 at 01:01:45 PM EST

Aaron A Aaronson!

I honestly believe that if ballots were going to be alphabetical, you'd find people changing their names to guarantee they'd be near the top.



[ Parent ]
147 candidates is nothing. (none / 1) (#194)
by dudsen on Mon Oct 27, 2003 at 11:21:12 AM EST

there are more seat's in the danish paliament then 147, and there are defintly more than 300 candidates spanned over 10 party's.
They ballots are quite long but they are sorted by party.
our system look a lot like the british, we just dont got an upper house for the nobels anymore, but we do got a Queen.


[ Parent ]
Standard "why voting machines?" answer (none / 1) (#116)
by thrownawayhack on Fri Oct 24, 2003 at 01:26:19 PM EST

The US elects a lot more people at an election than (say) Britain. For a Presidential election you might have different votes for

[federal]
President
Vice-President
Congressperson
Senator

[state]
Governor
Lieutenant-Governor
Treasurer
Secretary of State
State Assemblyperson
State Senator
[half a dozen more here]

[state judges]
State Supreme Court

[county]
sheriff
councillor
mayor

[city]
mayor
councillor

and then maybe some referenda for state, county, city etc.

There might be 20 or 30 independent races to adjudicate (people often vote different parties to different offices). I don't think that the election officials dare leave the ballots for several weeks while they count through the more important ballots.

So there is a nontrivial difference between US and British elections (I don't know about Canadian elections so I won't comment, but you don't mention having to vote on 20 things so I'm going to guess that Canada is closer to the UK than the US on this issue).

[ Parent ]

Okay (none / 1) (#137)
by 0xA on Fri Oct 24, 2003 at 07:31:11 PM EST

That makes much more sense. I knew that thaere would be more than one entry on the ballot but I didn't realize that there would be state and local governments on the same ballot. Do all the states do this or just a few?

You are right about Canada, federal, provincial and local elections do not happen at the same time. We also only vote for our MP (congressperson). The Prime Minister is the leader of the party with the most seats in the Parliment. Provincial elections are the same. Referendums are also very rare, I have only ever participated in one of them. We don't vote for our Governor General (president), she doesn't do anything anyway. We also don't vote for senators, they don't do much either. There is also no provincial senate.

[ Parent ]

Glad I could help explain :) (none / 1) (#142)
by thrownawayhack on Fri Oct 24, 2003 at 09:37:44 PM EST

I'm not a US citizen, so I haven't seen one of their ballots, but I think they may all be on one (big!) piece of paper. California has a few more things than some other places (lots of ballot initiatives).

Even if they're on separate pieces of paper then unless you have 20 times the number of counters that the UK or Canada have you'll have to hold the ballots safely for a while, and to be honest I don't think that anyone is trusted enough to look after them for that long ...

British elections sometimes coincide - Parliament and local (district/county) councillors are normally elected at the same time.

There were local referenda for the devolved assemblies in the UK, but the last national referendum was in 1975 (EU entry).

[ Parent ]

Is this a straw man argument? (none / 1) (#170)
by Gallowglass on Sun Oct 26, 2003 at 01:11:44 PM EST

The parent poster gentleman has listed a long list of offices and posits that this is the reason that electronic counting is a Good Thing.

I assume his argument is then that it would take so long to count them all because there are so many ballots.

With respect, I think the answer is already in place. Don't have just one person do the count. Each polling station probably has at most (?) 500 voters registered. Each poll will take the same amount of time roughly to count their results, the results then go to the electoral district headquarter to the Chief Electoral Officer (and his staff and the candidates reps) and the subtotals are summed up.

If this system is taking too long, then the answer is quite simple: Double the number of polling stations. Counting 250 ballots will take half as much time as counting 500 and you get your results faster.

Thus no need to switch to some fancy machine with unverifiable procedures to get results back faster.

Normally us Canucks never have to wait more than an hour or two after the polls close to know who won. Occasionaly, a day or two in a close result. Rarely a week or two if a judicial recount is necessary. But we have always known (in my generation and I'm a geezer!) which party won the election before we go to bed.

[ Parent ]

I wasn't saying that ... (none / 1) (#177)
by thrownawayhack on Sun Oct 26, 2003 at 04:55:38 PM EST

electronic counting was a good thing, but I was saying that maybe just having hand counting wasn't going to work. We're not talking about twice or thrice as many races to count. It's more like ten or fifteen times as many races. Maybe California isn't ready to spend ten or fifteen times as much to conduct an election as Canada. And I'm not sure that I want ballots waiting a week to be counted. I'm not sure that I trust every county to keep them safe that long ... This doesn't mean that you want an electronic machine; maybe some sort of punch-card or other mechanical device seems better. But hand counts have the potential to take a long time. I'm impressed that Canadian elections are counted so quickly. British general elections take several hours to count a reasonable number of constituencies - they'll normally get a majority in say five hours of counting, but in a really close race you might have to wait longer. Of course it's normally fairly obvious after the first declaration (~55 minutes in, often Torbay or Sunderland South) what's going to happen. But a formal declaration of a majority takes quite a bit longer than that. So go Canada! :)

[ Parent ]
is there a shortage of pencils in the US? (none / 2) (#149)
by DavidatEeyore on Sat Oct 25, 2003 at 05:12:06 AM EST

like our Canadian friend, down here in .AU, we still have a printed ballot paper which is issued by an electoral official and initialed by them _after_ they rule a line through your name in the printed electoral roll. You then go to the cardboard booth, pick up the pencil provided and mark the squares with your choice or write poetry or insults or nothing at all on the paper depending on your mood at the time (voting is compulsory here) fold up the ballot paper and put it into the ballot box ( which is sealed and watched by an election place official)
This seems to be an effective low-tech way of conducting the vote. At the end of the election day, the results are counted up by the electoral officials, scrutinised by party representatives and phoned/secure emailed to the tally room where they are announced on TV. All elections are supervised by a official body (Australian Electoral Commission) who are independent by statute. Why do you need machines of any sort to do this job in a democracy?

[ Parent ]
Who controls Diebold? (2.20 / 5) (#41)
by Magnetic North on Thu Oct 23, 2003 at 10:02:47 AM EST

Anybody else find it chilling that Rothschild, Inc. are standing in the shadows behind Diebold?

--
<33333
I thought it was legal... (2.88 / 9) (#42)
by UltraNurd on Thu Oct 23, 2003 at 10:45:54 AM EST

I'm a Swattie, and I'm on the SCDC mailing list, and I'm an SCCS sys-admin. I still don't know how we survived the Slashdotting. Anyway, my understanding was that this protest is legal: Diebold is illegally using clauses of the DMCA to threaten ISPs to remove the files you and your members are hosting, so you are legally pointing out their violation of copyright law. By definition, "civil disobedience" is illegal, namely violating an unjust law and facing the consequences to prove a point.

I understand that this is the SCDC view - are the two groups no longer in agreement?

Also, thanks to the /. post, you guys probably have several thousand more mirrors out there. :o)

--
"Your Mint Mountain Dew idea is hideous and wrong."
-Hide The Hamster

You're right (2.40 / 5) (#44)
by clark9000 on Thu Oct 23, 2003 at 11:08:15 AM EST

As far as I understand it, a "cease and desist" letter is only the beginning of a legal process. If you decide to comply with the cease and desist letter, the matter ends there. But of course it is perfectly legal not to comply, to take the matter to court and ask a judge/jury whether you must comply. So you're right that at this point, this is not technically civil disobedience because the matter hasn't yet gone to court and so non-compliance is still legal.

If a judge decides that, for instance, under the DMCA, it is illegal to host/distribute the documents, then to host them would indeed be breaking the law and subject to consequences.
_____
Much madness is divinest sense
To a discerning eye;
Much sense the starkest madness.

-- E. Dickinson
[ Parent ]
scdc/why-war? (none / 1) (#187)
by Optical on Sun Oct 26, 2003 at 10:23:29 PM EST

I believe the two groups have rather split on the matter at this point.

Though I guess its more accurate to say that Micah and Branen have split on the matter...

--
"The resulting snakes are flaccid. In order to erect more rigid snakes, it is vital to use a more stable method that can accomodate the large internal forces."
-Kass et al, the International Journal of Computer Vision, 1988
[ Parent ]

Why is this so difficult? (1.85 / 7) (#48)
by Dyolf Knip on Thu Oct 23, 2003 at 12:04:10 PM EST

And why on earth are they using Access for this? They couldn't spring for some SQL Server licenses? Or even mySQL?

Anyway, this should not be so hard for Diebold to do. Each voter goes into the booth and chooses their lesser of many evils. They are also told to enter in a pass phrase or something. This gets combined with their voter ID number or some such and an MD5 hash is made and stored with the vote record. Make sure the precise algorithm used to make said hash is public. Later on, the voter is able to go to a website, enter in their voter ID and their passphrase and see the vote logged as well as the MD5 hash. They can also buy a CD set containing all of the voting records for large scale no-bandwidth queries.

This solves the problems of securing anonymity (can't get the voter ID from the hash if the passphrase is any good), verifying individual votes, and verifying the count ("SELECT VoteChoice, COUNT(VoteChoice) FROM Votes GROUP BY VoteChoice"). With the hash ensuring every vote is unique, you don't have to worry about people who vote the same being given identical lookup IDs. About the only problem not solved is that of the company adding in lots of bogus votes. Any takers on that one?

---
If you can't learn to do something well, learn to enjoy doing it poorly.

Dyolf Knip

You DON'T WANT to verify votes! (3.00 / 4) (#50)
by Rich0 on Thu Oct 23, 2003 at 12:20:47 PM EST

Voters should not be able to prove how they voted.

If they could, then a mafia guy could walk up to you and say vote for X.  Then they ask you for your voter ID and passphrase after you vote.  If they get the wrong vote, they kill you / whatever.

Voters should be able to say, "sure, I voted for X" when in reality they voted for Y, and nobody at all can know the difference.

There are lots of ways to ensure an audit trail on votes which the voter can verify is accurate, but which don't let them carry it home afterwards.

[ Parent ]

Doesn't solve much (none / 2) (#51)
by scruffyMark on Thu Oct 23, 2003 at 12:32:26 PM EST

can't get the voter ID from the hash if the passphrase is any good

How do we get these passphrases to eligible voters? Don't forget - homeless people have to be enfranchised too. And when people forget their passphrases they must still be able to vote, without compromising confidentiality. And stealing mail must not enable people to get free votes.

And, of course, if anyone can undetectably insert a bunch of votes, that's sort of missing the key property of a secure election system

[ Parent ]

Don't need the passphrase (none / 0) (#237)
by Dyolf Knip on Fri Oct 31, 2003 at 11:25:03 AM EST

How do we get these passphrases to eligible voters?

Huh? Have them entered in at vote time by the voter. Or include a random phrase generator in the software, which of course is open source. The passphrase is not required to vote, only to verify your vote later on.

if anyone can undetectably insert a bunch of votes

Yeah, voting the cemetary. It's been done forever, but I was at a loss as to how to prevent it here.

---
If you can't learn to do something well, learn to enjoy doing it poorly.

Dyolf Knip
[ Parent ]

You CAN NOT let people verify their vote outside. (3.00 / 5) (#52)
by MyrddinE on Thu Oct 23, 2003 at 12:39:10 PM EST

As neat and tidy as that is, no voting process can allow the voter to verify their vote outside of the voting area, for one important reason...

If you can verify your vote outside the voting booth, so can someone else. Once someone can confirm how you voted (tell us your voting password or we break your daughter's fingers) then you open the door to voter coercion.

I cannot stress this enough... as nice as it would be to verify your vote after the election, it cannot happen or you open the floodgates on a Very Bad ThingTM.

[ Parent ]

Too late (none / 2) (#54)
by dipierro on Thu Oct 23, 2003 at 01:02:04 PM EST

Once someone can confirm how you voted (tell us your voting password or we break your daughter's fingers) then you open the door to voter coercion.

That door has already been opened (use an absentee ballot or we break your daughter's fingers).



[ Parent ]
The door is only open part way (none / 2) (#91)
by FlipFlop on Thu Oct 23, 2003 at 09:10:59 PM EST

Once someone can confirm how you voted (tell us your voting password or we break your daughter's fingers) then you open the door to voter coercion.

That door has already been opened (use an absentee ballot or we break your daughter's fingers).

In some places, you can overturn your absentee ballot by going to the polls and voting anyway. Before counting absentee ballots, the election judges check your name against the list of people who voted. If you voted, they throw out your absentee vote.

I'm not sure, but you might also be able to send in a second absentee ballot, invalidating the first one.

AdTI - The think tank that didn't
[ Parent ]

no, it's open all the way (none / 0) (#124)
by dipierro on Fri Oct 24, 2003 at 04:16:09 PM EST

In some places, you can overturn your absentee ballot by going to the polls and voting anyway.

I've never heard of such a system. But regardless, the person doing the coercing could simply watch you all day on election day to make sure you don't vote.

I'm not sure, but you might also be able to send in a second absentee ballot, invalidating the first one.

I'm not sure, but you might have a point there.



[ Parent ]
True. (none / 0) (#141)
by acceleriter on Fri Oct 24, 2003 at 09:07:15 PM EST

I've never heard of such a system. But regardless, the person doing the coercing could simply watch you all day on election day to make sure you don't vote.

But it doesn't scale.

[ Parent ]

It's an unrealistic problem anyway... (none / 0) (#144)
by dipierro on Fri Oct 24, 2003 at 10:03:46 PM EST

I mean, who's going to be able to coerce enough people to actually change the election? Anyone with that much power could easily find an easier way. Just pay off a few vote counters, or whatever. The absentee ballot coercion could easily scale to an entire corporation, if you're going to ignore the fact that anyone coercing that many people is going to get caught.

[ Parent ]
Or just pay of Diebold. (none / 1) (#147)
by acceleriter on Sat Oct 25, 2003 at 12:15:23 AM EST

exp(ln n + ln t)

[ Parent ]
Non sequitur (none / 0) (#155)
by dipierro on Sat Oct 25, 2003 at 01:49:14 PM EST

I absolutely agree that Diebold has implemented a terrible solution to the problem of voting. I think anyone who knows the facts is going to agree with that statement (or lie). You posted your comment in the wrong thread.

[ Parent ]
OK (none / 0) (#164)
by acceleriter on Sun Oct 26, 2003 at 11:01:40 AM EST

I grant that I'm preaching to the choir here with regard to Diebold. But even if one employer can't flip an election through rewarding or coercing employees, that's not the worry. The real worry is the distributed effect of multiple employers (who, let's face it, will have common interests), "encouraging" their staffs to vote in an appropriate manner.

It's imperative for the safety of what's left of the republic to deny the ability to pierce the secret ballot.

[ Parent ]

hyperbole (none / 0) (#175)
by dipierro on Sun Oct 26, 2003 at 03:22:38 PM EST

The real worry is the distributed effect of multiple employers (who, let's face it, will have common interests), "encouraging" their staffs to vote in an appropriate manner.

If the employers aren't threatening doing anything illegal, then I don't see the problem with it. Otherwise, I just don't see it happening on a mass basis. It's too much of a risk, for too little of an effect.

Like I alluded to earlier, being able to coerce someone's vote implies that you have a lot of power over them. If you're able to do it, it really doesn't matter to them who they vote for anyway.

It's imperative for the safety of what's left of the republic to deny the ability to pierce the secret ballot.

We're not talking about affecting the secrecy of the ballot. We're talking about affecting the ability to prove that you voted a certain way. The safety of the public is in no way affected by this. Even with coercion we still would have seen either Bush or Gore win the last election, and the republic still would have survived.

Perhaps in the past this could have been used against blacks or women to sytematically deny their ability to vote, and indeed that's still the case to some extent today. Some spouses are able to exert that amount of control over the other, and some guardians could control their dependent enough to coerce their vote. But that door is already open, in the form of the absentee ballot. If we're going to design the system around that problem, we need to fill the absentee ballot gap first.

Like I said, personally it doesn't make a difference to me. I'm not going to be coerced by anyone, and if ever am in that situation, it's not going to matter to me who wins anyway.



[ Parent ]
The ability to vote a certain way . . . (none / 0) (#178)
by acceleriter on Sun Oct 26, 2003 at 06:52:58 PM EST

. . . or, specifically, the ability to prove to a third party one voted in a certain way most certainly does compromise the secrecy of the ballot, and opens the door to mass coercion, whether subtle or overt. I can't agree with your stance that coercion should be allowed to be facilitated because it might not be illegal, or because those being so coerced might not value their vote more than their jobs or lives.

[ Parent ]
if it's legal it's not coercion (none / 0) (#180)
by dipierro on Sun Oct 26, 2003 at 07:46:57 PM EST

the ability to prove to a third party one voted in a certain way most certainly does compromise the secrecy of the ballot

This isn't about proving how a third party voted. This is about proving how you voted yourself.

I can't agree with your stance that coercion should be allowed to be facilitated because it might not be illegal, or because those being so coerced might not value their vote more than their jobs or lives.

It's simple. I think the benefits outweigh the detriments. I don't think buying and selling votes is such a bad thing, and that's what we're really talking about if there is no illegal activity involved. It's not coercion, it's just a consensual adult activity.



[ Parent ]
If you can carry away proof . . . (none / 0) (#184)
by acceleriter on Sun Oct 26, 2003 at 09:26:45 PM EST

. . . then proof to yourself can be presented to a third party. If the proof is deposited into a ballot box (for a potential manual recount), I have no objection.

The bing and selling of votes is illegal. And you can't possibly really be saying that if one's employer requires you to present a receipt showing you voted for candidate Foo as a condition of employment, only a "consensual adult activity" is happening as opposed to coercion. That's an excess of Libertarianism that I can't say I've ever before seen.

[ Parent ]

Yeah (none / 0) (#185)
by dipierro on Sun Oct 26, 2003 at 09:52:28 PM EST

I misread you about the third party thing. My bad.

The b[uy]ing and selling of votes is illegal.

Hmm, I had heard it wasn't. Do you have any cite for that one?

And you can't possibly really be saying that if one's employer requires you to present a receipt showing you voted for candidate Foo as a condition of employment, only a "consensual adult activity" is happening as opposed to coercion.

I don't see how you consider it coercion. No one is forcing anything upon someone else. They are merely entering into a voluntary contractual agreement. This is assuming, of course, that the employee is an at-will employee.

That's an excess of Libertarianism that I can't say I've ever before seen.

I'm not a Libertarian. Not even close.



[ Parent ]
Citation, etc. (none / 0) (#186)
by acceleriter on Sun Oct 26, 2003 at 10:05:38 PM EST

Vote buying and selling is prohibited by federal law in the U.S. by Title 18 U.S. Code, Section 597 and by laws of the individual states. The language in 18 USC 597:
Whoever makes or offers to make an expenditure to any person, either to vote or withhold his vote, or to vote for or against any candidate; and Whoever solicits, accepts, or receives any such expenditure in consideration of his vote or the withholding of his vote - Shall be fined under this title or imprisoned not more than one year, or both; and if the violation was willful, shall be fined under this title or imprisoned not more than two years, or both.

With respect to coercion vs. contract: in employer/employee relationships, there exists an asymmetry of power such that it is impossible to have an agreement among equals. Except in unusual cases of scarcity or specialization, most employment relationships are of a "take it or leave it" nature. One has a better chance of negotiating prices at Wal*Mart than of successfully rebuffing a demand from one's employer and remaining employed.

I hope I didn't offend by hearing Libertarianism in your last post. That wasn't my intention.

[ Parent ]

Libertarianism (none / 0) (#190)
by dipierro on Mon Oct 27, 2003 at 01:49:36 AM EST

With respect to coercion vs. contract: in employer/employee relationships, there exists an asymmetry of power such that it is impossible to have an agreement among equals.

Oh please. I've had jobs before, and I sure as hell consider myself an equal with my employer. They don't treat me well, I tell them to shove it.

Except in unusual cases of scarcity or specialization, most employment relationships are of a "take it or leave it" nature.

So there's a choice there. Take it, or leave it. That's not coercion.

One has a better chance of negotiating prices at Wal*Mart than of successfully rebuffing a demand from one's employer and remaining employed.

So you quit. Why would you want to work for an employer who is going to force you to vote a certain way? No one is that desperate for a job.

I hope I didn't offend by hearing Libertarianism in your last post. That wasn't my intention.

Not really. I've just got a bit of a pet peeve against Libertarians. It's great in theory, you know, but in reality it just doesn't work.

It's kind of the opposite of how I see this situation. Yeah, in theory buying and selling votes sucks, but in reality it's not going to make all that much of a difference. The rich already run this country.



[ Parent ]
Obviously . . . (none / 0) (#192)
by acceleriter on Mon Oct 27, 2003 at 07:40:54 AM EST

. . . you're in a position to tell your employer to "shove it" when mistreated. Most people, particularly those less skilled than yourself, are not. There can be no morally valid contract when there exists an asymmetry of power that allows one side to dictate terms. An example of a "contract" based on asymmetry of power that comes immediately to mind is the Treaty of Versailles. That worked well in the long term.

[ Parent ]
too late (none / 0) (#195)
by dipierro on Mon Oct 27, 2003 at 01:38:02 PM EST

If people are put in such a position that they are slaves to their employers, the problem does not lie in the voting system, nor does the solution.

[ Parent ]
Whether that's the case or not . . . (none / 0) (#200)
by acceleriter on Mon Oct 27, 2003 at 08:48:48 PM EST

. . . there's certainly no reason to provide the tools to help enslave them.

[ Parent ]
This isn't a tool to help enslave them (none / 0) (#201)
by dipierro on Mon Oct 27, 2003 at 09:09:03 PM EST

Like I said, they are already enslaved.

[ Parent ]
OK . . . (none / 0) (#205)
by acceleriter on Mon Oct 27, 2003 at 11:36:13 PM EST

. . . to help continue to enslave them. Better :)?

[ Parent ]
No (none / 0) (#206)
by dipierro on Tue Oct 28, 2003 at 02:20:57 AM EST

This does help continue to enslave them either.

[ Parent ]
Yes, it would. (none / 0) (#210)
by acceleriter on Tue Oct 28, 2003 at 11:03:02 AM EST

It would allow the political system that has them enslaved to continue, by ensuring that it isn't voted out.

[ Parent ]
No it wouldn't (none / 0) (#215)
by dipierro on Tue Oct 28, 2003 at 02:39:37 PM EST

That's my point. The system wouldn't be affected whatsoever. Whether we had Bush or Gore, it wouldn't matter.

Not to mention that the whole "enslavement" you're talking about is completely voluntary. People choose to enslave themselves with their big TVs and SUVs and backyards and private schools. The cost of living for a person is easily obtainable without enslavement.



[ Parent ]
I'm not talking about Bush v. Gore (none / 0) (#216)
by acceleriter on Tue Oct 28, 2003 at 03:34:38 PM EST

Nor am I referring to a two-income family struggling to make the payments on their Hummer and $2.50 a gallon gas. I'm talking about the fact that in a future government, the conditions of enslavement could exist. And in that case, provided employers were complicit (and in a fascist-leaning government, that would be likely), it could be kept in office through the complicity of businesses who could literally decide who eats and who does not.

[ Parent ]
Bush vs. Gore is a good example, though (none / 0) (#217)
by dipierro on Tue Oct 28, 2003 at 04:55:03 PM EST

I'm talking about the fact that in a future government, the conditions of enslavement could exist.

And I'm saying at that point it doesn't matter. That future government could just repeal the laws we have protecting the secrecy of the ballot, or even repeal voting altogether. They could stuff the ballot boxes with fake votes, and lie about the counts of the votes that were made.

Transparency is more important than secrecy in protecting against government abuse. Yes, coercion and other injustices can take place, but at least the people know it's taking place and can take steps to overturn those injustices. When you don't have transparency these injustices can take place and they won't be stopped because most people don't even know it is happening.

I am talking about Bush vs. Gore here. It's actually a perfect example of the problem with our current voting system. Not only do you and I not know whether the legitimate winner won the election, no one in the world knows.

But all of this is somewhat of a moot point, anyway. It is possible to have both secrecy and transparency. On a small scale we have it once a year at my local fire department's elections. You simply take a box and show that it is empty. You then let people mark their candidate on a ballot and drop the ballot in the box. When everyone has voted, you count the votes.

So like most online discussions, I guess we were arguing over nothing. I stand by my belief that vote verification is harmless, but in the end, it's somewhat unnecessary. It'd be nice if there were a random serial number on each ballot, so that I could go down to the place where they keep the ballots and find mine, but you wouldn't object to that leading to coercion, would you?



[ Parent ]
I wouldn't object to the serial number (none / 0) (#218)
by acceleriter on Tue Oct 28, 2003 at 06:19:49 PM EST

so long as someone couldn't bring someone with him or her or carry away proof of how he or she voted.

It does sound as if we agree more than we disagree, but I've enjoyed the discussion nonetheless. I don't know if I'd call it pointless.

[ Parent ]

Can != Must (none / 0) (#238)
by Dyolf Knip on Fri Oct 31, 2003 at 01:18:05 PM EST

If you can carry away proof

If you are truly worried about being coerced into revealing your voting behavior under a system like the one I described, then make a point of not knowing your passphrase. You can't reveal what you don't know.

On the other hand, a schmuck boss or landlord or mob boss could still say, "no password, no vote record, you punished". Hmmm. Of course then they'd be in the position of going after people who simply don't remember their passphrase (I certainly wouldn't plan on writing mine down or having the machine print it out) as well as those who feel they need to hide their choices.

Another problem that comes to mind is what do we actually do if 1000 people come forward and say, "The vote on record for me is incorrect"?

---
If you can't learn to do something well, learn to enjoy doing it poorly.

Dyolf Knip
[ Parent ]

You can't? (none / 1) (#96)
by zcat on Thu Oct 23, 2003 at 11:54:20 PM EST

Here's a cheap digital camera. Go vote, take a picture of the ballot, and bring me the camera back afterwards. Any questions?

[ Parent ]
Here's Rule #45678. (none / 0) (#131)
by i on Fri Oct 24, 2003 at 06:21:13 PM EST

No cameras allowed in the voting booth. Any questions?

and we have a contradicton according to our assumptions and the factor theorem

[ Parent ]
Won't work (none / 1) (#56)
by dipierro on Thu Oct 23, 2003 at 01:18:39 PM EST

They are also told to enter in a pass phrase or something.

Won't work that way, because then the person telling you to vote a certain way could just tell you what pass phrase to use. You have to assign them the pass phrase.

Which leads to a second problem. If your voter ID number is verifiable, then it's a simple matter of forcing you to tell your pass phrase.

Solve the problem by assigning voter IDs at the booth, you say? That won't work either, since the computer can just assign you a voter ID and pass phrase of someone who voted for the candidate its programmers wanted to win.

You can't make voting both anonymous and secure without resorting to a physical paper trail.

Anyway, this should not be so hard for Diebold to do.

It's not. Just have the computer print out a scannable ballot and have the person verify the ballot and drop it in a ballot box. At the end of the voting period, the computer will have an exact count, and there will be a paper trail there to verify it.



[ Parent ]
Ah, interesting (none / 0) (#240)
by rtav on Fri Oct 31, 2003 at 06:40:24 PM EST

It's always nice to find someone who sees the anonymity issue as problematic w.r.t. electronic voting.  Unfortunately, some people just _can't see_ why it shouldn't be possible for them to vote from the comfort of their trendy downtown apartment while sipping a $6 (CAD, though...) cup of coffee.

Sigh.  Apparently anonymity or potential coercion just aren't big problems for a lot of people...

[ Parent ]

Nobody cares (none / 1) (#102)
by epepke on Fri Oct 24, 2003 at 05:41:12 AM EST

I mean, seriously, nobody cares.

It isn't difficult to do things properly, but the days of craftsmanship are over. These are the days of nice suits and power ties. Nobody cares about getting things right. People care about their image and their salary and their stock options. They care about what kind of spin they can pull, or what kind of marketing options they have. They don't care about quality.


The truth may be out there, but lies are inside your head.--Terry Pratchett


[ Parent ]
Much ado about nothing (2.09 / 11) (#53)
by Skywise on Thu Oct 23, 2003 at 12:45:33 PM EST

While I applaud your efforts to expose a company's stupidity.  I don't see your point in trying to "save voting rights".  Because... frankly, you're not.

You do realize that no voting system currently implemented and in use is secure, right?

You've got 4 basic voting machines in use in the US:

paper scanning
punched card
old voting machine (mechanical)
new voting machine (electronic)

Of the 4, only the punched card and paper scanning styles allow for a paper trail.  And we know how well those stand up to scrutiny after the Florida Election.

The voting machines keep a summary tally and possibly a paper receipt of each individual vote (Which only serves to verify the summary tally... not a verification of the vote itself)

So the storage of the information in an open Access database seems to be a moot point.  There's no password protection on any of the existing machines.  All security is EXTERNAL to the machine itself which is still the method being used, last I checked.

None of the evidence presented proves that the machines were faulty.  You infer that the Georgia elections were rigged, but you have no proof... You haven't shown that the Georgia elections used uncertified software, and you haven't shown that the results were invalid.

You also haven't shown that the California election board did NOT approve the patches.  You imply it, but the election board obviously noticed it in the documents you provide and ran with the election.

If you want to Republican bash, claim righteous  patriotic fervor and try to recreate the great "right wing conspiracy".  Fine.  But you don't have a smoking gun... Anymore than the Republicans did when they tried to claim that Vince Foster was assassinated and Clinton was secretly trying to dismantle the US Army to be taken over by the UN.

All you've got is proof of a typical software development house mentality of get the release out the door.

Overly general (3.00 / 8) (#58)
by cestmoi on Thu Oct 23, 2003 at 01:39:32 PM EST

Of the 4, only the punched card and paper scanning styles allow for a paper trail. And we know how well those stand up to scrutiny after the Florida Election.

Your criticism of punch cards in unfair. There are good ways to implement punch cards and there are bad ways. Florida's butterfly ballot was a mess. The Data Vote system used in my district is fine. You slide a large lever next to your candidate's name and push. No hanging chad, no dimple, and a very low error rate. Best of all, you can read your vote after you've punched. It's very straight forward.

None of the evidence presented proves that the machines were faulty.

Not true. The Diebold documents describe their error detection method. I quote:

We use the simplest of these which is to treat the data as a series of numbers and store totals of sets of those numbers as separate data known as checksums. If the data has been modified without updating the checksums, then the checksums will fail to add up.

They admit they use the simplest checksum method. Simple checksums don't catch transposition errors. That is, if the numbers being summed are 100 and 200, the checksum is 300. But that's the same checksum for 200 and 100. Worse yet, it means if the device that holds the permanent record of the vote is hit by static, cosmic ray, Bill Gates or whatever, there's no way to recover the original vote. That would require a more elaborate error detection method than a simple checksum. The machines, as described by Diebold's internal emails, are simply not reliable.

Ultimately, this whole mess is an example of political appointees deciding to approve a technology whose fundamental weaknesses are arcane enough to pass political muster and yet severe enough to compromise an election. The appointees don't have the technological moxie to understand how easily touch screens can be compromised. Worse, their instinct will be to go into "cover your ass" mode as people start questioning who approved these machines in the first place.

[ Parent ]

The issue is trust (none / 1) (#61)
by TheModerate on Thu Oct 23, 2003 at 02:09:14 PM EST

In the end, you have to trust someone. Whether the volunteers at the voting area, the people who count the votes, or whoever, eventually that trust, that cause for our faith in democracy, must be placed in no small part in the hands of someone. Its a serious naivety to first look at elections in terms of how it can be corrupted, and secondly, that because it can be corrupted then it must be corrupted.

Democracy is a fragile leaf, and for those who say it has already been irreversibly destroyed, well! perhaps we should just learn to accept minority rule.

"What a man has in himself is, then, the chief element in his happiness." -- Schopenhauer
[ Parent ]

"you have to trust someone." alternative (none / 1) (#169)
by Gallowglass on Sun Oct 26, 2003 at 12:57:40 PM EST

Don't trust someone. Trust a number of people who
  1. have an interest in an accurate count ("They won't cheat my candidate out of his votes!")
  2. (b) have separate interests which cancel each other out. (Candidate A and Candidate B each have a representative at the poll.)


[ Parent ]
No evidense? more to the point... (3.00 / 11) (#70)
by IPFreely on Thu Oct 23, 2003 at 02:38:42 PM EST

While it is true, as you say, that there is no evidense that there was election rigging in the Georgia elections. The bigger problem is that there is no evidense that there was NO RIGGING. I know that proving a negative is almost impossible, and that is generally considered bad arguement. One of the primary points of most election systems is that it is reliable, reproducable, and auditable. The complete results of the Georga election were dumped, no audit trail left, no paper, the computers were washed. There is NOTHING LEFT. No evidense either way. That is a bad situation. The reason for auditable situations is precisely so you can go back and verify that there is no fraud. If you can't do that, you are left with a deeply unreliable system.

And one of the major compalints against Diabold is not only that they don't have a reliable audit facility, but that THEY ARE ACTIVLY RESISTING IMPLEMENTING ONE. At each turn, they don't say "How can we make our system more acceptable? I know. Lets do what the cusotmer wants it to do." Instead, their approach is "How can we make the public accept our current system without reliable auditing capability".

The problem is not proving whether a particular election was rigged. It's too late then. Its about making sure it doesn't happen in the future. And almost everything Diabold management has done is resisting actually implemeting a reliable auditing system when it would be so easy to implement. That is what is suspicious. They are not trying to fix it. They are actively pushing a known broken system and trying their best to make people accept it.

[ Parent ]

Wrong place for security... (none / 2) (#79)
by Skywise on Thu Oct 23, 2003 at 04:38:53 PM EST

You can't make a secure voting machine at Diebold's level.

Vote tampering, rigging, ballot stuffing, etc;  All happen outside of the box by crooked electioneers.  Y'know, those guys that would hold the keys to boxes and/or count the votes.

So you use an ultra secure encrypted database with a triple verified software engine that was open for public inspection.  You hand the passwords over to a guy who was put into his position by his Dad's uncle who was the leading political figure in the county for the last 20 years who exchanges the db file in the machine for one that was specifically crafted to show a majority for his party and that he made based upon the public source!

Where's the security in that?

[ Parent ]

At least make it more difficult (none / 3) (#84)
by JyZude on Thu Oct 23, 2003 at 05:58:31 PM EST

True, but you could make a system that is more difficult to tamper with.

  • Send the data to another computer (db slaving) so that a second auditor could view the data in realtime and would notice file replacement
  • Send the data to multiple slave computers
  • Split the database backups into many parts, encrypted with different keys
  • Require two passwords, given to two different people (one a democrat and one a republican?), just like the Red October

If you have a triple-verified passworded vote machine, and you suspect tampering, at least then you can point to the guy with the passwords, rather than Joe Blow with an mdb file.


-----
k5 is not the new Adequacy k thnx bye


[ Parent ]
Sure you can (none / 0) (#107)
by IPFreely on Fri Oct 24, 2003 at 10:19:39 AM EST

You can't make a secure voting machine at Diebold's level.

Well, Diebold can't anyway. It sounds like you are saying that if you can't have perfect security/integrity, then you shouldn't even try. Is that what you meant? There is lots of room for security in between perfect and nothing. Right now, Diebold has nothing.

The comment was about auditing, not security. Auditing is about making multiple records for each event, and tracking and storing them separately. (preferably with each type being secured/managed by different people). You also want recording formats that are harder to alter/replace. After the election, each group counts its format, and the results are compared. If they match then you have an election. If they don't then something is broken or someone tried to change it. Security is about preventing tampering. Auditing is about detecting tampering after the fact. Either is good. Both together is better.

There will always be cronies trying to affect the election. The problem with Diebold is that they make it SO EASY to alter records, bypass any audit, and fraud the election that Diebold might as well be handing out instructions: "Open the database, put in your prefered results, close. you're now elected." It takes five minutes, and noone can see or know what happened. It's a lot harder to alter hundreds of boxes of different types of paper ballots managed by different people than a few fields in an Access database (Not than it hasn't been tried).

The complaint is that Diebold is resisting doing even these basic types of auditing. The underlying suspision is that they want to make a system that is easy to manipulate. And with Diebolds strong monetary and political ties to the republicans and the highly questionable results in Georga, those suspisions are hightened and tied to the republicans.

[ Parent ]

The evidence that a "faulty memory card" (none / 2) (#92)
by michaelp on Thu Oct 23, 2003 at 09:11:33 PM EST

was able to add and delete votes without it triggering any alarm certainly proves the machines are faulty.

If a really bad memory card can add ~2000 votes to one candidate while deleting 19,000 from another, without triggering any sort of alarm how many votes do just slightly flawed memory cards alter?

Until and unless these folks can assure us that if bad hardware does add or delete significant numbers of votes, someone is going to be notified, the system is clearly not ready to be trusted.

All you've got is proof of a typical software development house mentality of get the release out the door.

A 'typical software house' that doesn't know how to or bother to verify data integrity is at very least commiting fraud if they sell a system as being ready for holding sensitive data.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

It's TWO DIFFERENT SYSTEMS [n/t] (none / 0) (#93)
by Skywise on Thu Oct 23, 2003 at 10:36:38 PM EST



[ Parent ]
Which one are you now talking about? (none / 1) (#100)
by michaelp on Fri Oct 24, 2003 at 03:52:36 AM EST

Your original statement:

"None of the evidence presented proves that the machines were faulty."

Implies that you believe that all the machines mentioned in the article were not faulty.

I was talking about the Diebold Tru-Vote system mentioned in the article, the one where a bad memory card added and deleted votes all by itself. I would call that a 'faulty' machine. Hopefully the DCMA will let me.

But you don't have a smoking gun...

See above. That is a smoking gun to anyone who writes software. If this memory card problem happens more than once, then we have no idea how many votes have been written by machine. Second, unless the Tru-Vote is using a VERY unusual database set up, the only way data could be entered and deleted by a "faulty memory card" would be if the memory card contained software capable of writing to the data file it contains. Random bits from hardware faults won't cut it, you don't get valid numbers into ANY standard database by throwing random bits of machine code into it.

This more than suggests that there is something VERY strange going on with these machines, something that hidden source code makes impossible to determine.

You've got 4 basic voting machines in use in the US:

Yeah, with the mechanical machines you mention, we know how they work, so if one suddenly started producing and deleting votes, it would be rather easy to see what is going on with it. The code for the e-voting machines needs to be as open as the designs of the hardware of the machines they are supposed to replace, before we trust them. When you move the count from levers to bits, there is no need to suddenly hide how the thing works.

By the way, "it's TWO" doesn't make sense, did a bad memory card on your boxen type that for you?


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Really frightening part: (none / 2) (#94)
by losthalo on Thu Oct 23, 2003 at 10:44:59 PM EST

That we do in fact know what went wrong, and how much, and why, and not only are they still using the system, and encouraging its spread, but knowing the problem didn't stop bad results from being used.

Gah!

[ Parent ]
Republican bash? (none / 0) (#103)
by pyramid termite on Fri Oct 24, 2003 at 05:43:26 AM EST

Is there something magical about the security holes in this system that prevents Democrats from using it?

On the Internet, anyone can accuse you of being a dog.
[ Parent ]
There should be. (none / 0) (#106)
by IPFreely on Fri Oct 24, 2003 at 09:49:15 AM EST

Is there something magical about the security holes in this system that prevents Democrats from using it?

No, but there should be. Would it be a better system if both parties were using it to defraud the public equally?

WTF does it matter who is doing it. It's election fraud. It's illegal. Fix it! It's like all the republicans are out saying "Gee, it may be fraud, but it's my kind of fraud. So lets keep it the way it is."

[ Parent ]

Evidence of tampering (none / 0) (#165)
by edverb on Sun Oct 26, 2003 at 11:36:46 AM EST

According to Diebold contract engineer Rob Behler, Diebold applied uncertified patches in Georgia just six weeks before the election.

Paul Krugman is correct in citing Henry Kissinger's A World Restored in pointing out that we are faced with a "revolutionary power" that does not recognize the legitimacy of the existing system. In the face of such boldness, those who point out the painful truth regarding the threat to our system are dismissed as alarmists, while those who preach a reasonable, moderated response are heeded, resulting in and caused by a self-perpetuating inability to take the threat at face value.

How much obstruction, how many lawsuits and legal threats, how much evidence of conflicts of interest, how many red flags will it take to convince the "reasonable" that representative democracy has already been compromised?

I'll waste a link to the best e-voting watchdog on the web even though Diebold's legal scare tactics have successfully taken the site down at present.




--------------------
$ wget --user-agent="verbose2.2" -r -p http://www.google.com
[ Parent ]
Electronic Voting is a Dead End (2.70 / 10) (#57)
by NateTG on Thu Oct 23, 2003 at 01:20:57 PM EST

There was a post on the other site that seems appropriate here:
"Electronic voting is not even a solution looking for a problem.  It's a problem, posing as a solution."

Providing a touch screen machine for printing out easily counted human-readable ballots is not a bad idea as long as the physical ballots are the used for the official count.

I also don't object to using the machine count as a method to verify that the human count is reasonable -- since even with spoiled ballots, the two should be very close.

For all of those brilliant folk out there who think that there's some simple solution to the voting problem, it's impossible to satisfy these criteria simultaneously:

  1. Insure that votes are not modified
  2. Insure that invalid votes are not added
  3. Insure that valid votes are not removed
  4. Insure that votes cannot be coerced


Solving the four criteria... (none / 3) (#62)
by dipierro on Thu Oct 23, 2003 at 02:14:25 PM EST

Don't paper ballots solve those 4 criteria? Doesn't a properly built electronic voting machine solve those 4 criteria? In fact, these two solutions are really the same thing. It's just that one is easier to understand and harder to subvert. I mean, we could create scannable bubble cards which use diappearing and reappearing ink, but that'd be really hard to do.



[ Parent ]
Nope Paper ballots do not. (none / 2) (#67)
by Skywise on Thu Oct 23, 2003 at 02:28:40 PM EST

Two real life examples.

First.  Missing ballot boxes.

Second.  The dead people who still vote.

[ Parent ]

That's implementation details (none / 3) (#71)
by dipierro on Thu Oct 23, 2003 at 02:41:59 PM EST

Not a problem with the system itself. I am a member of organizations which have anonymous voting all the time, and we've never experienced either of those problems.

Don't lose ballot boxes, and don't let dead people vote. Problem solved.



[ Parent ]
It works better than others. (none / 1) (#86)
by squigly on Thu Oct 23, 2003 at 07:04:29 PM EST

The first of these should at least be extremely easy to spot.  It's not hard to count ballot boxes.  The response will have to be extreme (invalidate the election, and start again), but it does give a possible response to such tampering.

Dead people voting is a little harder to spot, but that will be a problem with all electoral systems that aren't connected to an accurate database of eligible voters.  It's not a problem with paper ballots.

[ Parent ]

This is real life, not Plato (none / 2) (#114)
by bugmaster on Fri Oct 24, 2003 at 01:12:22 PM EST

We don't need to achieve some sort of a Platonic ideal of voting. All we need to do is reduce the probability of fraud to a level that we're comfortable with -- not eliminate fraud even as a theoretical possibility.

Concretely, a secure electronic voting system, that uses strong crypto every step of the way, can satisfy criteria 1..3 as well as a paper-based system can. Sure, a single person can somehow commit identity fraud and vote multiple times, but this may actually be more difficult with an electronic system. Similarly, changing votes is extremely difficult with a properly implemented electronic system, because cracking strong crypto keys is not practical. Also, the touchscreen system is easier to use than pregnant chads, so the chance of accidental error is lower.

And of course, criterion #4 (vote coersion) cannot be satisfied by any voting system out there: I can always hold a gun to your head and command you to vote Bush, regardless of how you do it.
>|<*:=
[ Parent ]

#4 is possible (none / 1) (#121)
by mcc on Fri Oct 24, 2003 at 04:03:21 PM EST

I can always hold a gun to your head and command you to vote Bush, regardless of how you do it.

Unless you put in place a system where in order to vote, one must enter a little enclosed booth, and no one is allowed to go into or look into the booth while you are in it.

(Yes, I realize absentee ballots break this.)

[ Parent ]

You're missing the most important criterion. (3.00 / 4) (#126)
by Estanislao Martínez on Fri Oct 24, 2003 at 04:24:28 PM EST

5. Ensure that everybody has a fair chance of understanding and monitoring the process.

--em
[ Parent ]

Shocking (2.78 / 14) (#63)
by UnConeD on Thu Oct 23, 2003 at 02:14:48 PM EST

Today, with strong public-key cryptography, secure hashing, etc. being available to anyone, some jackass company is still storing votes in an unprotected MS Access .mdb ???

Sheesh. I guess that's what happens when you 'let the market decide' and employ a for-profit company for one of the few things the government *should* be occupying itself with.

Not all that strange (none / 0) (#225)
by KrispyKringle on Tue Oct 28, 2003 at 11:13:07 PM EST

It's not all that strange. In order to make any real use of PKI, you'd have to have a trusted system to do the signing. In this case, the physical machine itself may not be trusted, so you end up needing every voter to have a signing key that is trusted. Which of course isn't possible right now.

One of the memos points out that so long as someone needs to have access, that person with access can modify votes. You could have the machine sign the file, admittedly, before its transmitted (and yes, it is utterly ridiculous that they don't at least do this, to secure against insecure transmission), but you still would have weaknesses if you can't trust the machine or those doing the tallying, etc. Nothing is perfect.

[ Parent ]

So someone is a thief (1.00 / 25) (#65)
by kurtmweber on Thu Oct 23, 2003 at 02:17:45 PM EST

I can't believe you're applauding someone stealing the private, confidential, copyrighted work of another. Dumbass...

Kurt Weber
Any field of study can be considered 'complex' when it starts using Hebrew letters for symbols.--me
I can't believe ... (2.40 / 5) (#104)
by pyramid termite on Fri Oct 24, 2003 at 05:52:48 AM EST

... you're saying that our election process should be private, confidential or copyrighted. Elections are a PUBLIC function and anyone who contracts to provide this PUBLIC function needs to provide free and open access to security issues associated with it. If the system has holes in it, the citizens have a right to know about it.

And before you answer with a bunch of libertarian crap, there's no such thing as a RIGHT to run the government's elections without such safeguards in place. If they don't like it, they're free not to provide software for the government.

On the Internet, anyone can accuse you of being a dog.
[ Parent ]
Here here (1.00 / 5) (#108)
by sellison on Fri Oct 24, 2003 at 11:16:11 AM EST

the Diebold corp is run by good, moral, Christian people.

The theives who stole their memos and illegally published them should be subject to the full punitive powers of the DMCA!


"No, I don't know that atheists should be considered as citizens, nor should they be considered as patriots. This is one nation under God."- George H.W. Bush
[ Parent ]

I think when (none / 0) (#109)
by michaelp on Fri Oct 24, 2003 at 11:26:08 AM EST

the 'private work' reveals evidence of fraud or other crime, the copyright protection is no longer valid.

Copyright is intended to protect the profitability of the work that is copyrighted, not as a way to hide incompetance or criminal activity.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

But what criminal activity is being committed? (none / 0) (#117)
by kurtmweber on Fri Oct 24, 2003 at 01:26:34 PM EST

Well?

Kurt Weber
Any field of study can be considered 'complex' when it starts using Hebrew letters for symbols.--me
[ Parent ]
Negligence, violation of the public trust (none / 1) (#119)
by michaelp on Fri Oct 24, 2003 at 03:32:34 PM EST

In fact, the lead engineer from Diebold wrote over two years ago that anyone could change votes without leaving a trail:

And strong evidence of outright vote fraud:

Lowe ordered all of the precincts reviewed. At 1:24 a.m., the review showed that 412 of 585 registered voters in Precinct 216 had cast ballots - but that they had given 2,813 votes to Bush! Gore had a negative vote: minus 16,022. Ralph Nader's negative vote was even greater. The problem was traced to an error in the memory card.http://www.unc.edu/~pmeyer/usat29nov2000.html  

See http://www.kuro5hin.org/comments/2003/10/21/2367/2543?pid=93#100 for a discussion the problem with attributing added and subtracted votes to an 'erroneous memory card'.

Again, copyright is intended to protect the profitability of copyrighted works. If Diebold was selling a book including these corporate memos and someone made them available for free, they would have a copyright violation.

Instead they are trying to abuse the law to cover up evidence of their own incompetance and criminal behavior (violation of their contracts with the counties at the very least).

If reporters can no longer report or reprint corporate or govt. documents they discover due to violations of copyright, we may as well declare the first amendment dead and buried.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Er, that's what copyright is for? (none / 0) (#239)
by rtav on Fri Oct 31, 2003 at 06:33:24 PM EST

That's a wacky way to think about it. At least in the US, I see copyright as stemming from "To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries" (US Constitution, Article I, Section 7) Exactly how do you derive "Copyright is intended to protect the profitability of the work that is copyrighted" from that? The idea that copyright is intended for individual profit instead of the benefit of the public is a particularly terrible idea---I'm just not sure how one would best work to get rid of it.

[ Parent ]
Important Point (2.83 / 6) (#76)
by CENGEL3 on Thu Oct 23, 2003 at 03:52:51 PM EST

I generaly support your arguement that the technology behind electronic voting machines should be open to public scrutiny and that security and audit trails are a very important consideration in such technology.

I also agree that the security of Diebolds system sounds shockingly lapse.

However please do not try to imply that the variance between actual election results and public opinion polls can be used as some sort of indicator of possible election fraud.

Even a suggestion of such is horribly flawed and detracts from your main (and legitimate) point. Opinion polls are NOTORIOUSLY unreliable. For one thing they are subject to the possibility of purposefull bias in the manner in which they are constructed and implemented. Additionaly, even when they are carefully constructed to attempt to avoid any bias there are a large number of other factors which can skew thier results.... not the least of which is the simple fact there is no assurance that the number of people willing to participate in the poll (and answer honestly) is uniform accross the electorate.


So do a recount. (3.00 / 6) (#77)
by evanbd on Thu Oct 23, 2003 at 04:15:53 PM EST

If the opinion polls and the election disagree wildly, then that is cause to be suspicious. So, you recount the ballots and verify, and then go with what the ballots say. But when you can't do a recount, there's a problem, no? The polls say we should be suspicious, but because of how badly diebold's systems are built, we can't even do a recount; we have to trust numbers which are easy to manipulate. And therein lies the problem -- not that there is disagreement between polls and results, but that there is no way to check.

[ Parent ]
then manipulate the opinion poll (none / 1) (#95)
by Daelin on Thu Oct 23, 2003 at 10:58:06 PM EST

If a recount can be forced by a large dispute with an opinion poll, then the opinion poll will have to be regulated.  Then, essentially, you've got a RAID-1 voting system.

Seems like a bad idea.  It'd be great if there were some way to produce a voting checksum.

[ Parent ]

What about exit polls? (none / 1) (#105)
by squigly on Fri Oct 24, 2003 at 06:15:18 AM EST

People often change their opinion in the polling booth.  It is unlikely that there would be such a huge difference, but it's possible.  

However, the press does tend to ask electors who they voted for.  This should be a lot more accurate.  How does this compare with the results?

[ Parent ]

No it's not (none / 2) (#110)
by CENGEL3 on Fri Oct 24, 2003 at 12:34:31 PM EST

Absolutely voting systems should have an audit trail, allow for easy and reliable verification, and allow for public scrutiny as to how they operate. We absolutely agree on that.

Where we absolutely do NOT agree is this statement "If the opinion polls and the election disagree wildly, then that is cause to be suspicious." I've already pointed out that there are many legitimate reasons that an opinion poll can vary wildly from the actual results. In fact, I could craft a poll that would lead you to believe Ronald McDonald won the last election.
Poll results being different from election results is NOT cause for suspicion.

One of the candidates or an election observer should always be able to call for a recount if THEY believe there is reason to be suspicious. When a recount is called for it should be done across the board for all districts that decide the election (i.e. if it's a county election, the entire county should be recounted, if it's a state then the entire state). Votes should be recounted according to very precise rules established BEFORE the election. Additionaly only 1 recount should be allowed....a candidate should NOT be able to call for recount after recount after recount until they get the results they want.

[ Parent ]

Why not be suspicious? (none / 2) (#168)
by Gallowglass on Sun Oct 26, 2003 at 12:45:28 PM EST

You wrote:Poll results being different from election results is NOT cause for suspicion."

Umm . . . Why not?

He didn't say that if the difference was proof, merely suspicious. One may be suspicious and still understand that one could be wrong and that nothing untoward has happened. As an example, in a murder case the first people the police will investigate are the immediate family because they are the most likely suspects. That doesn't mean that they have come to the conclusion that the victim was not murdered by a stranger.

Similarly, if i come into a room and smell a certain sweet smoke in the air, I may be suspicious that someone was smoking maijuana there. But I can't say that they have unless I have corroborative proof.

Suspicion isn't wrong. Equating suspicion with certainty is.

So the original posters point that a wide difference between a survey and the vote is reason to be suspicious, I think he has a point. He didn't say it was proof, just a reason to wonder if it was so or not.

[ Parent ]

What are the choices? (none / 0) (#241)
by dachshund on Fri Oct 31, 2003 at 11:36:06 PM EST

When a recount is called for it should be done across the board for all districts that decide the election (i.e. if it's a county election, the entire county should be recounted, if it's a state then the entire state). Votes should be recounted according to very precise rules established BEFORE the election. Additionaly only 1 recount should be allowed....a candidate should NOT be able to call for recount after recount after recount until they get the results they want.

I agree. However, what do you do when the rules haven't been clearly established? What do you do when there are different types of recount-- machine recounts, hand recounts, etc.-- and application of one gives reason to believe that the more specific method is necessary?

Nobody wants to wind up in the uncertain situation you talk about. It's just that sometimes you have no better alternative. It seems that you can wind up in a situation where you either a) stick with the first, clearly inaccurate count, or b) work with all parties to come up with a more accurate approach. Surprisingly, there are people who fail to realize that the first option is as bad (or worse) than the second.

[ Parent ]

Why electronic voting cannot work (2.81 / 11) (#81)
by mcelrath on Thu Oct 23, 2003 at 05:01:31 PM EST

I see comments here and have seen many comments along the lines of "electronic voting could work, if it was done right". Here's why that whole idea is dangerously flawed.

Voting and the verification of the vote is a process that everyone has an incentive to muck with. Therefore everyone has an incentive to make sure it doesn't get mucked with. The way to do this right now is to get involved with politics, and actually go count votes. You can watch the process and verify that it is working. Verify that the numbers you write down get carried upstream, verify that upstream is not mucking with the numbers. Fraud can be at worse statistical, with a few fradulent precincts going each way, and the solution to prevent this is to get involved and make sure *your* precinct isn't fraudulent.

Contrast this with electronic voting. Now, we can devise all kinds of complicated hashes and crypto to verify that bits and pieces aren't being mucked with. However each bit of crypto ultimately comes down to a dialog box. "This chunk has been cryptographically verified"...or not. All a hacker has to do is muck with the code to always pop up the correct dialog boxes and voting officials won't know any better.

It gets worse. How to you ensure that any piece of the software or transmission mechanism isn't mucking with the counts? Sure you can sign code but ultimately that too comes down to a dialog box. You can sign votes too, but ultimately just narrows those who can commit fraud to those who have access to the keys. It does not prevent fraud.

Even if every citizen is given a public key and signs his opwn vote with it, then votes and signatures are stored in a database, there still is the matter of the database of keys. In order to verify the database of votes you must have the public key for every citizen. So like the keyserver becomes a single point of failure. It could return different keys to different people, so the election officials think the database is always untampered.

The only way I can see to maybe reconcile the problem is individual keys, and diverse software. If an election offical can come in with his laptop and use a hand-compiled copy of gpg to verify that the votes are unmodified, perhaps it really is unmodified. But you can never trust the voting machine to tell you it hasn't been tampered with. So counting votes would be dependent on hackers and several implementations of the crypto software for its security. There's no way in hell I'd type my passphrase into some piece of unknown equipment to 'verify' my vote. For all I know it could store my passphrase or change my vote before signing.

How do you deal with this problem with paper ballots? A recount, with different people counting different piles of ballots than last time. A computer cannot recount. The number will be the same.

I cannot think of a method to do electronic voting that is not seriously flawed and does not have a single point of failure. But, perhaps I am not thinking hard enough.

-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.

But an important problem with your analysis (none / 1) (#88)
by br14n on Thu Oct 23, 2003 at 08:02:16 PM EST

is that numbers aren't necessarily equal just because squaring them results in the same number.

[ Parent ]
siggy (none / 1) (#98)
by mcelrath on Fri Oct 24, 2003 at 12:29:51 AM EST

Ok time for a new sig. No one takes me seriously! :)

-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.
[ Parent ]

in this case... (none / 1) (#101)
by dimaq on Fri Oct 24, 2003 at 05:34:00 AM EST

...someone should make a nice little box that shreds the [paper] ballot inserted and prints a better one. for future's sake :)

[ Parent ]
Paper voting cannot work (none / 1) (#113)
by bugmaster on Fri Oct 24, 2003 at 01:02:11 PM EST

Everyone has the incentive to muck with paper-based voting; therefore, everyone will. The ballot box that contains the pregnant chads will eventually have to be opened up, at which point the ballots would be counted. You can institute however many recounts you like, but the bottom line is that it all comes down to a single person counting the votes. If that person lies about the count, that's it.

Ok, the above was satire, but the bottom line is that there is always some chance of fraud. So what ? Should we just chuck the whole thing and switch back to a monarchy ? No. Yes, electronic voting can be horribly insecure, suffer from single point of failure, etc. -- but no more so than paper-based voting, unless these Diebold people are implementing it. And in fact, electronic voting can be made more secure than the old-fashioned kind, because humans can't compute RSA in their heads.

Come to think of ot, all the problems you listed in your comment actually apply to any situation where data needs to be collected securely. Fortunately, we have crypto, audit logs, and good old-fashioned security guards, so not all is lost.
>|<*:=
[ Parent ]

Incorrect (none / 2) (#120)
by mcc on Fri Oct 24, 2003 at 03:48:33 PM EST

The ballot box that contains the pregnant chads will eventually have to be opened up, at which point the ballots would be counted.

The thing is that with paper-based ballots, you can sign up to be an election monitor. You can stand there and watch each piece of paper enter and leave that box, you can see for yourself if someone else is in the room mucking about with the process. Obviously not everyone does this. Obviously the chances are that there will be occational precincts where no netural parties sign up to be monitors, and the only people watching the process work for the nefarious Boss Tweed. However, things balance out. Outright voter fraud can only be so widespread. There are many more eyeballs watching out for irregularities at many more levels than there would be otherwise. The presence of a variety of observers helps keep the system honest.

You can't do this with electronic voting. You can't sit there and watch as bits move around in a Microsoft Access database. You can sit there and stare and make sure no one opens up the vote box. But if the votes are stored on a hard drive at your feet, you have no way of percieving if someone just exploited a bug in the voting software to quietly change 600 of the votes on said hard drive to be for a different person.

I could agree with an electronic voting system where you walk in a book, touch some things on a touchscreen, and the touchscreen prints out a little scantron sheet, which you then walk out and drop in a box so that later it can be run through a huge scantron machine and tabulated. But that crucial center step-- the actual storage of the votes-- should never be allowed to be resigned to fragile, gullible electronic media.

---
Aside from that, the absurd meta-wankery of k5er-quoting sigs probably takes the cake. Especially when the quote itself is about k5. -- tsubame
[ Parent ]

Actually you can (2.75 / 4) (#122)
by bugmaster on Fri Oct 24, 2003 at 04:06:45 PM EST

You can't do this with electronic voting. You can't sit there and watch as bits move around in a Microsoft Access database.
Actually, you can -- though, of course, not in a Microsoft Access database. That's what audit logs are for; they have been in use for about as long as RDBMSs existed, so it's well-known how they work. In fact, you could set up a realtime log of votes scrolling on your screen, if you really wanted to.
But that crucial center step-- the actual storage of the votes-- should never be allowed to be resigned to fragile, gullible electronic media.
Again, this is absolutely true. Which is why any kind of real voting system would use encryption -- so that it is mathematically impossible (ok, not impossible, just slightly less probable than the Sun exploding tomorrow) to alter the votes on that hard drive. And, naturally, you can have sysadmins watching the access logs to that machine, just to make sure no one's trying to hack it. In fact, securing electronic voting this should be easier than securing, say, amazon.com -- since every voting terminal can be accounted for (which is not true for amazon.com users).

The more I read your comments, the more they sound like, "locomotives are completely inadequate because there's no way to feed them hay -- so what will they eat ?". Locomotives, unlike horses, don't eat hay; electronic voting does not use paper. So what ? Locomotives win in the long run, and so does electronic voting -- unless you have utterly incompetent morons like Diebold implement it, of course.
>|<*:=
[ Parent ]

Purely electronic voting requires trusted terminal (none / 3) (#128)
by NateTG on Fri Oct 24, 2003 at 04:47:07 PM EST

Purely electronic voting requires trusted terminals.  Since electronic devices are more error prone than paper, trusting paper is preferable.

In addition, the terminal represents a single point of compromise so it is possible to create an undetectable compromise unless the voter verifies the votes on a printout, or on a second terminal.

Hard Copy is also a good idea because it provides an audit trail, and a means for recount.

If you have a system where the terminal provides a count, and there is a second count based on the printouts, it's already worlds better in terms of reliability and credibility than the single DB/single count model that Diebold uses.

Regardless, in order to justify the use of purely electronic voting, you need to show that it has benefits that make it more attractive despite the massive risks that it has compared to hybrid or paper systems.

BTW: Locomotives are usually not a suitable replacement for horses for many reasons.  The only real exception is long distance bulk cargo transportation.

[ Parent ]

Trust (none / 1) (#136)
by bugmaster on Fri Oct 24, 2003 at 07:12:01 PM EST

I already trust supermarket terminals to deduct the right amount from my credit card. I also trust my ATM, and my bank. I also trust airline routing software to not crash my plane (or send it to Timbuktu). Why are voting terminals different ? Because they're new ?
>|<*:=
[ Parent ]
Difference (none / 2) (#140)
by LeftOfCentre on Fri Oct 24, 2003 at 08:54:05 PM EST

The difference is that it is astronomically more important that vote counts are accurate, than that a debit or credit card purchase deducts the accurate amount of money from your account. On an individual level, the things you mention are very important, but the damage to society as a whole is much higher if even the foundation of representative democracy -- the process of voting -- can not be properly guaranteed.

[ Parent ]
Re: Trust (none / 2) (#152)
by mcelrath on Sat Oct 25, 2003 at 12:16:08 PM EST

But with your credit card and ATM you receive a statement, and can verify that the transactions occured as expected. If not, there is recourse by calling your CC company and filing for a fraudulent transaction. With airlines you get on the plane and actually find yourself somewhere else. So you know where it went. Again there is recourse. The Airline sending people to Timbuktu would not be in business long.

With voting, how do you know what you voted? How do you know if you voted? How do you know how it was recorded? How do you know if it was counted properly?

For an individual vote, there is no verification, and there is no recourse (mostly because there is no provable verification).

As I mentioned in my original post, the only way around this is with individual voter public keys. And that comes down to a dialog box on some windows hardware, which you can't trust.

-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.
[ Parent ]

I don't (none / 1) (#193)
by Cro Magnon on Mon Oct 27, 2003 at 09:36:43 AM EST

I DON'T trust supermarket terminals OR the ATM OR the bank! I look at my credit card/bank receipts to make SURE I'm not getting screwed. And if I didn't get those receipts, I wouldn't be using credit cards or banks!
Information wants to be beer.
[ Parent ]
Still has trust problems (3.00 / 4) (#129)
by mcc on Fri Oct 24, 2003 at 05:15:00 PM EST

That's what audit logs are for; they have been in use for about as long as RDBMSs existed, so it's well-known how they work. In fact, you could set up a realtime log of votes scrolling on your screen, if you really wanted to.

That would be better. However you are still trusting everything to something you can't see. How do you know someone hasn't compromised the program that creates and displays for you the audit log?

Again, this is absolutely true. Which is why any kind of real voting system would use encryption -- so that it is mathematically impossible (ok, not impossible, just slightly less probable than the Sun exploding tomorrow) to alter the votes on that hard drive. And, naturally, you can have sysadmins watching the access logs to that machine, just to make sure no one's trying to hack it.

Who appoints the administrators? Can they be arbitrarily picked and represent the entire political spectrum, like today's election observers are? How would that work out?

Encryption is not a magic bullet. Yes, there are methods of encryption today that are effectively unhackable. That just means that you find some point to attack that isn't under the eaves of the encryption. And there are many. How about this: Someone fiddles with a voting machine the night before the election and patches the program that actually takes the vote, encrypts it, and inserts it into the remote database. The patch causes 1 out of every 20 votes to randomly be toggled, just before the encryption is performed, to be voting for Gary Coleman; then, at at a preset time 1 hour from the close of the polls, the patch deletes itself from the system. How could this sort of thing be found? How could its origin be reliably traced?

There is no analog to this sort of scenario with paper voting. All methods of compromising paper voting would have to take place physically, out in the open, in possible view of election observers and others. And while one could come up with an elaborate system to the "what if someone's secretly patched something?" problem involving randomly auditing machines and checksums and trust checks at every level, this all becomes really, really difficult and COMPLICATED to work out. And at some point one really just has to ask, what ADVANTAGE are we gaining by making the vote-storage step electronic?? What is so important about changing to electronic methods that they require going through the convolutions that are necessary to make the electronic methods trustworthy? Paper, meanwhile has one very important advantage: it is SIMPLE. In a problem like voting systems, the more unnecssary moving parts you have, the more routes of compromise there are.

The more I read your comments, the more they sound like, "locomotives are completely inadequate because there's no way to feed them hay -- so what will they eat ?".

Except this isn't horses and locomotives. It isn't food that is being debated, it's standards of accountability. And most importantly, the goal is not to get from point A to point B as quickly as possible. The goal is to record the will of the voting populace as accurately as is reasonably possible.

[ Parent ]

I can't see lots of things (none / 1) (#135)
by bugmaster on Fri Oct 24, 2003 at 07:09:41 PM EST

That would be better. However you are still trusting everything to something you can't see.
But I am used to trusting everything to something I can't see. Every time I start up my car, or board a plane, or pay for groceries with my credit card, or listen to the radio, or do any of the of the other 1e6 things that people take for granted, I am trusting some invisible software to do its job. Why should voting be any different ? Your argument sounds more and more like an irrational phobia to me: "invisible software is evil".

Who appoints the administrators? Can they be arbitrarily picked and represent the entire political spectrum, like today's election observers are?
Why not, actually ? If, as you claim, everyone involved with the voting process (the vote counters, the monitors, the janitor sweeping the floor, the electrician servicing the lights, the...) is carefully chosen to be politically pure, then why not sysadmins ?
The patch causes 1 out of every 20 votes to randomly be toggled, just before the encryption is performed, to be voting for Gary Coleman; then, at at a preset time 1 hour from the close of the polls, the patch deletes itself from the system. How could this sort of thing be found? How could its origin be reliably traced?
This problem (the patching, not the voting for Gary Coleman) has been solved multiple times already. One of the recent cases, IIRC, involved DirecTV. The other one is, sadly, being touted as the next best thing for everyone -- trusted computing. Yes, it's evil, but in the context of voting machines it can do some good.
What is so important about changing to electronic methods that they require going through the convolutions that are necessary to make the electronic methods trustworthy?
Huh... I thought I made that clear in my previous posts, but I guess I should elaborate:
  • Ease of use. No more voting for Gary Coleman accidentally.
  • Automation. Votes are counted automatically and instantaneously. The chances of a miscount are greatly reduced.
  • Security. It is much harder to tamper with the system, since it's cryptographically secure (not true of Diebold's .mdb file, naturally).
  • Accountability. Ideally, anyone should be able to inspect the voting logs, the underlying code, the firmware, etc. Security through obscurity doesn't work anyway.
Paper, meanwhile has one very important advantage: it is SIMPLE.
One word: Florida. And of course, you're assuming that human vote counting personnel is simpler than computers...

Note that I am definitely not claiming that electronic voting is a magic bullet. Duh. Nothing is a magic bullet. But I believe that a properly implemented voting machine is an incremental change for the better, and that's good enough for me.
>|<*:=
[ Parent ]

Re: I can't see lots of things (none / 1) (#154)
by mcelrath on Sat Oct 25, 2003 at 12:47:11 PM EST

Why not, actually ? If, as you claim, everyone involved with the voting process (the vote counters, the monitors, the janitor sweeping the floor, the electrician servicing the lights, the...) is carefully chosen to be politically pure, then why not sysadmins ?
They are not politically pure, and paper elections are not fraud-free (see my previous comment about trust). Personally I do not think any kind of "automatic counter" should be used at all (scantron, whatnot) because again they cannot be trusted.

The way you get around not trusting the count it is by making a pile of votes and having one person count it, then pass that pile to another person and have them count it. Triple count every stack. It still can't prevent collusion, but it makes fraud less likely.

Stupid voting equipment is just that. Stupid. There are stupid ways to do a paper vote and there are stupid ways to do an electronic vote. The "electronic" part isn't a magic bullet either. There are also stupid voters. In any election some fraction of the votes are incorrect for various reasons. (i.e. pretty girl walked past while voting, punch Buchannan accidentally, chase girl...) This kind of thing can be chalked up to a pocket of statistical fraud, and can't affect things that much because it should be statistically random and happen both for and against any given candidate in small amounts. It's only that the 2000 election was very very close that anyone even noticed.

This problem (the patching, not the voting for Gary Coleman) has been solved multiple times already. One of the recent cases, IIRC, involved DirecTV. The other one is, sadly, being touted as the next best thing for everyone -- trusted computing.
Uh no, it has not. Again, trust. With DirecTV they hold the keys and they can trust themselves. With "trusted computing", Microsoft holds the keys. If they wanted, they could insert any kind of code they wanted on your computer and you have no recourse. Microsoft is the single point of failure in trusted computing. It is a system that may work to keep viruses off your computer (as long as this is in MS' interest) but is inappropriate for voting or voting machines.

-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.
[ Parent ]

Elections are important. (none / 2) (#191)
by haro on Mon Oct 27, 2003 at 04:58:05 AM EST

But I am used to trusting everything to something I can't see. Every time I start up my car, or board a plane, or pay for groceries with my credit card, or listen to the radio, or do any of the of the other 1e6 things that people take for granted, I am trusting some invisible software to do its job. Why should voting be any different ? Your argument sounds more and more like an irrational phobia to me: "invisible software is evil".

If someone steals your car - that is bad for you. If someone steals your vote - that is bad for all of us. If someone hijacs your flight - that is bad for you. If someone hijacs your election - that is bad for all of us.

Elections are far more important than even the survival of any single one of us. Elections are as important as the survival of all of us. Not even when you are prepared to use an all automated airline system - plane, flight controller, repair etc. - should you trust an all automated election.

My answer to your question "Why should voting be any different ?" is "Elections are important."

[ Parent ]

The experts say... (none / 1) (#139)
by dennis on Fri Oct 24, 2003 at 08:51:36 PM EST

Right, all that encryption tech makes it practically impossible to spoof elections, which is why about 1000 cryptographers and computer security experts publicly recommended that we use...paper. Yeah. I don't have a link handy but Bruce Schneier's written about it, check his newsletter.

[ Parent ]
Re: Paper voting cannot work (none / 3) (#153)
by mcelrath on Sat Oct 25, 2003 at 12:28:22 PM EST

You underestimate the importance of having a single point of failure. The game becomes to gain control of that single point, which is far more important than actually convincing voters...

With paper-based voting there is still fraud, but it should be statistically random so on the whole it cannot affect the vote that much. And in the US the vote was not designed to be an accurate measurement of the people's will anyway. The founding fathers had many debates over whether the people were really intelligent and informed enough to rule themselves. The Electoral College was set up as a way to buffer the masses, and essentially adds a random element to the presidential vote anyway. A small amount of randomness due to pockets of fraud does not affect the system very much. While the Electoral College only affects presidental elections, it is what forces the two party system, and has repercussions in all elections.

Crypto is not a magic bullet. The problem is not crypto, it is trust. Which pieces of the system do you trust, and how do you verify that you trust them? The answer is none. Voting is a system whereby no party should trust any other. The only way to collect the vote is to distribute potential fraud as widely as possible. And any individual can verify a small piece by volunteering. Trust in crypto is similarly a sticky issue, and leads to the trustdb in gpg/pgp and "Key Authorities" for SSL and PKI because there's no such thing as double blind trust. In such systems the key holders become the single point of failure. The government cannot possibly be trusted to hold keys.

I am not familiar with audit logs, but again: how can you trust the audit log?

-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.
[ Parent ]

Umm. . . *Not* just one person counting! (none / 2) (#166)
by Gallowglass on Sun Oct 26, 2003 at 12:24:59 PM EST

You wrote, "the bottom line is that it all comes down to a single person counting the votes. If that person lies about the count, that's it."

If that's what happened, then you'd be right. But that is most emphatically not what happens in a manual count. I'm not sure how it works in the US, but in Canada, the count is overseen by the Polling Officer, and the Assistant Polling Officer, plus the poll representatives of each candidate in the election. So that's a minimum of 4 people: 2 hired by a government department that oversees elections, and 2 civilians outside of government and of different parties. (The number of civilians though is quite often as much as 3 or 4 due to the fact that we have more than two parties.)

Then if any of the candidates objects to the results, the physical ballots (which are sealed in an envelope before everybody after the count) are taken before a judge (ours are appointed, not elected so arguably less politically biased). Now add in the Judge, the Sherriff of the court &c, and the number of people doing the count doubles (at least!) again.

Which is a lot more than one person doing the count. :-)

In essence then, my argument is that manual counts are most emphatically not single-point-of-failure constructs.

Abolute monarchy is, like a dictatorship, horribly undemocratic. (For the same reason. Neither is responsible to the electorate.) Elections are the method we use to hold the elected responsible to the electorate. (I only insert this because you posed the question, "Should we just chuck the whole thing and switch back to a monarchy ?")

I have written a comment before which goes into more detail on how elections are held here and how the count is verified. (Also some good replies. Thanks to lunatic for corrections.)

[ Parent ]

Update (2.80 / 5) (#83)
by Fredrick Doulton on Thu Oct 23, 2003 at 05:40:54 PM EST

It appears that the school in which these students hosted their leaked memos caved in and pulled the pages. Too bad for Diebold that the memos are already making the rounds on bit torrent/kazaa/DC/name your P2P network.

Bush/Cheney 2004! - "Because we've still got more people to kill"

.torrent mirror (none / 2) (#156)
by LocalH on Sat Oct 25, 2003 at 02:04:39 PM EST

http://localh.kicks-ass.org/bbv/ Help spread Diebold's dirt around! Leave the torrents open and become a seed! If you can't get to it, try again later, my 16K/s up doesn't help.

[ Parent ]
primary concern (2.00 / 4) (#85)
by codejack on Thu Oct 23, 2003 at 06:49:34 PM EST

ignoring, for this post, everything about hacking security, permissions, encryption, insert-your-favorite-mode-of-preventing-vote-tampering-here, someone, somewhere, has to implement this system, presumably on a large scale (with only 3 companies producing e-voting machines, a presumably safe assumption); that person is then in a tempting position. not even making the whole process "open," i.e. allowing anyone who wants to to view the software, hardware, practices, etc., with the best security on earth, will alleviate the simple, gaping, inevitable security flaw that could be accessible by ONE person. At best, you have a potential king-maker, at worst, the election goes to the side with the best hackers (no! must fight the temptation!).

so, here's my proposal; first, allow me to define a term i am choosing to use, perhaps not correctly (please let me know :)) political granularity: a measure of how close to the actual wishes of the populace a representative will vote. everybody got that?
ok, let's go; first, change the entire system to a quorum based, multi-tiered representative democracy, where you need a certain number of votes to get an electoral vote, to vote for the next tier. note that these votes do not necessarily need to be from any given "district," you vote for the closest person who agrees with you, perhaps broken down further into categorised branches of government, such as social services, security, foreign affairs, etc. also, the number of tiers would be flexible, in response to how the members of each tier vote.
second, do away with the concept of "terms"; you hold office until you are voted out. in our modern society, this should produce no significant burden.
third, each "branch" would need it's decision vetted by at least 2 other branches, presumably related, and defined beforehand, i.e. the foreign relations department's decision would need to be vetted by security and another department or two; social services for immigration related issues, treasury for tariffs and trade agreements, etc. disputes about which department must vet a decision would be resolved through a general quorum of reps from each department. the inherently scalable and quickly responsive nature of the system should allow for small quorums whose decisions could be overturned easily in cases of malfeasance (admittedly, an imperfect solution vis-a-vis the ability to resist popular support for unwise policies, but thems the breaks).
anyone got a better idea? please use the included response form.


Please read before posting.

[poll] Will you vote using a Diebold machine? (none / 0) (#123)
by criquet on Fri Oct 24, 2003 at 04:10:43 PM EST

Cast your vote in my diary.

Hmm (none / 1) (#125)
by trhurler on Fri Oct 24, 2003 at 04:22:37 PM EST

I have the same question for you that I have for all these "civil disobedience" people. Are you aware that throughout history, the whole point of civil disobedience was to force yourself to be punished for something the public would not stand to see you punished for? And if so, do you believe the public will take your side on this one, and are you willing to be punished? Not as a group, but personally, because that's what will happen if anything does.

If so, fine, but I find that most people using the words "civil disobedience" think they just mean "breaking the law for a good reason." That's not true. You may well have a good cause here, but I just wonder how well you understand the game you're playing.

--
'God dammit, your posts make me hard.' --LilDebbie

Of course (none / 1) (#134)
by Eater on Fri Oct 24, 2003 at 06:45:59 PM EST

Every person who uses civil disobedience and fails was just "using the words "civil disobedience" [because they] think they just mean 'breaking the law for a good reason.'"

Eater.

[ Parent ]
What's with all the technology worship? (2.57 / 7) (#127)
by Estanislao Martínez on Fri Oct 24, 2003 at 04:37:16 PM EST

The most worrisome thing about this whole drive to implement electronic voting (or even worse, paperless electronic voting) is not the technical complexities of it, nor the allegations of fraud. It's the fact that the people pushing the damn things do not bother to argue that there's a need to do it. Of course, if it's computer-based, it therefore must inevitably be better, since as any seller of technological solutions knows, the only thing better than complicated technology is more complicated technology. Right?

--em

The problem isn't 'complicated technology' (none / 1) (#132)
by michaelp on Fri Oct 24, 2003 at 06:42:57 PM EST

its the way the technology is being implemented.

Mechanical voting machines are quite complicated, but how they work isn't hidden so fraud can be detected.

The immediately important part to the user is the interface, which can be made quite simple using an e-voting system.

And if the code is open to public view (like the design of the mechanical voting machines is), then fraud and security holes or possible bias can be identifed and eliminated (and with a software upgrade rather than a truckload of new machines).

The difference is that with the current commercial e-voting systems, the code that runs them is hidden from public scrutiny. Putting a requirement for e-voting code to be open source would eliminate most of the problems with e-voting and facilitate fixing any new ones that are found.

As for the reason to go to e-voting, they include faster and more accurate counts, better error checking during the vote, simpler intrfaces, no trucking of ballot boxes from one place to another, the old machines are worn out and expensive to fix and/or dirty and hard to keep clean, etc.

It's the current closed source implementation that is the problem, not the concept.



"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

I'm not so sure (none / 1) (#145)
by cestmoi on Fri Oct 24, 2003 at 11:40:52 PM EST

Putting a requirement for e-voting code to be open source would eliminate most of the problems with e-voting and facilitate fixing any new ones that are found.

A virus that inserted itself into something like a printer driver or a video driver could steal votes and you wouldn't know it was happening. The vote counting code would look benign but every so often, the virus could activate itself and flip a few votes. You'd scan the os but since it's a special purpose virus, its signature isn't going to show up on Symantec's virus publications. Unlike most viruses, it would be installed by hand by a consultant or crooked election employee.

Others have argued that paper balloting with multiple counters is the most secure and I'm beginning to think they may be right.

[ Parent ]

People think viruses are easy to make and (none / 1) (#148)
by michaelp on Sat Oct 25, 2003 at 02:54:10 AM EST

hard to stop because they are used to using software coded to be good enough for marketing to sell rather than software coded with security and data integrity as the highest priority.

Systems designed from the get go for security have very few viruses and methods for detecting and reporting anomolous behavior (aka heuristics)so its much harder to 'flip a few votes' one way or another in a database designed for security & data integrity than it is to convince outlook to start spewing emails all over the place. Further, properly designed systems can be set to log all changes that are made to them, using different code than is counting the votes (sort of like how zonealarm watchs over windows networking) & these logs could be sent to an independed monitoring agency.

& finally, systems designed for single purposes are dramatically more secure than the general purpose desktop OS most people use, which is full of holes to enable various features which obviously should not be included on a vote counting machine (whether the holes should be included on general purpose desktop machines is another issue:-).

Whether all this would make it more secure than paper ballots with multiple counters or not, if e-voting is going to be used, contractees should demand it be opensource and the code be published for public inspection.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

while all this is true (none / 3) (#161)
by Battle Troll on Sat Oct 25, 2003 at 09:32:26 PM EST

Systems designed from the get go for security have very few viruses and methods for detecting and reporting anomolous behavior (aka heuristics)so its much harder to 'flip a few votes' one way or another in a database designed for security & data integrity than it is to convince outlook to start spewing emails all over the place.

It is also irrelevant to any currently commerically successful electronic voting software. One popular packages uses Win98.
--
Skarphedinn was carrying the axe with which he had killed Thrainn Sigfusson and which he called 'Battle Troll.'
Njal's Saga, ca 1280 AD
[ Parent ]

Right, you may as well use blank notebook paper (none / 1) (#172)
by michaelp on Sun Oct 26, 2003 at 02:08:23 PM EST

for paper ballots, then claim that paper ballots are "inherantly unsecure".

Again, the paralells to analog voting machines: the designs are patented, not copyrighted, so anyone can examine the engineering for stupid mistakes.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

the point is (none / 1) (#176)
by Battle Troll on Sun Oct 26, 2003 at 04:46:10 PM EST

Right now, we have established mechanisms for fairly-secure paper voting. These mechanisms are not more secure than the most theoretically secure electronic voting system conceivable, but they are more secure than any electronic voting package that is currently commerically viable.

Should we develop better, more-secure voting mechanisms? Sure! Is e-voting currently secure? No way.
--
Skarphedinn was carrying the axe with which he had killed Thrainn Sigfusson and which he called 'Battle Troll.'
Njal's Saga, ca 1280 AD
[ Parent ]

the point is (none / 1) (#181)
by michaelp on Sun Oct 26, 2003 at 08:17:22 PM EST

e-voting won't be 'secure' (both from external and internal manipulation) unless whatever code it uses is open source.

Therefor, step 1 in contracting for e-voting software should be that the code is available for public scrutiny.



"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

and it should be free, as in 'beer' /nt (none / 1) (#183)
by Battle Troll on Sun Oct 26, 2003 at 08:45:42 PM EST


--
Skarphedinn was carrying the axe with which he had killed Thrainn Sigfusson and which he called 'Battle Troll.'
Njal's Saga, ca 1280 AD
[ Parent ]
Open source isn't it either (none / 1) (#158)
by Tau on Sat Oct 25, 2003 at 07:22:02 PM EST

Ken Thompson wrote an interesting paper on this once. Basically, just because I show you some code, doesn't mean that's what's running on the voting machine. And moreover even if I let you install the voting software onto the machine, it's still got to be compiled. In theory the compiler can trojan the voting software binaries as well as trojaning any copies of itself it compiles (as in the paper). Remote but still theoretically possible.

---
WHEN THE REVOLUTION COMES WE WILL MAKE SAUSAGES OUT OF YOUR FUCKING ENTRAILS - TRASG0
[ Parent ]
One can similarly hack paper ballots (none / 1) (#159)
by michaelp on Sat Oct 25, 2003 at 08:09:22 PM EST

& other methods if the local security is negligent.

And moreover even if I let you install the voting software onto the machine, it's still got to be compiled. In theory the compiler can trojan the voting software binaries as well as trojaning any copies of itself it compiles (as in the paper). Remote but still theoretically possible.

Those aren't arguments against open source, those are arguments against unmoderated voting using any vote counting technology; one can go into detail about how to print fake ballots or make counterfit voting machines, etc. Obviously the e-voting machines still need local (physical) security just like the ballot boxes need to be accounted for, printing counterfit ballots & hacking the hardware of an analog vote counting machine need be prevented, etc.

Point is that if e-voting is used, the code (including the compiler or VM) needs to be publically published. As you have pointed out, if the code is open source one can analyze it for potential weaknesses and protect against them with appropriate (analog or digital) security methods. If the code is closed source, there is no way for anyone to tell what kinds of 'trojans' it might contain.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Yes, that is a problem. (3.00 / 4) (#162)
by Estanislao Martínez on Sat Oct 25, 2003 at 09:51:32 PM EST

And if the code is open to public view (like the design of the mechanical voting machines is), then fraud and security holes or possible bias can be identifed and eliminated (and with a software upgrade rather than a truckload of new machines).

Here's a simple point: NO COMPUTERIZED VOTING SYSTEM WILL BE AS TRANSPARENT AS PAPER BALLOTS. This simply because any computer system will be orders of magnitude more complicated than a mechanical or paper-based system. Even if the source code and the hardware of the machines were publicly available, the skill set required to evaluate them is just too prohibitive.

Sure, paper elections are not simple to start with. BUT YOU DON'T HAVE TO BE A COMPUTER ENGINEER TO UNDERSTAND THEM. And "you can hire one to do it for you" is not an acceptable answer here, because (a) it involves trust, and an election is something that should be designed so as to require the minimum amount of trust from participants; (b) election results are things that can be called into question e.g. in the courts, and if evaluating the voting process requires a particularly complicated expertise such as engineering, both parties can bring expert witness that testify in favor of their own interest, and in the end non-experts are going to have to decide which side is right.

--em
[ Parent ]

Luddish baloney (none / 1) (#182)
by michaelp on Sun Oct 26, 2003 at 08:34:57 PM EST

mechanical voting machines are quite as complex (as far removed from entropy) as an e-voting machine.

Paper ballots may rely on less complicated base technology, but the limitations of paper as a display medium make larger ballots highly complex from an information density pov.

Further, paper ballots can be easily hacked with counterfit copys, a pen, and access to a ballot box.

Nor do you have to "hire" a computer engineer to tell whether a piece of open source code contains vulnerabilities or trojans, there will be plenty of people capable of looking at the code and reporting if there are problems.

In fact, probably at least as many if not more than are capable of looking at the engineering diagrams of a punch card reader and telling what might go wrong, and certainly more than are capable of analyzing a multipage page paper ballot for placement or order bias that will skew the results.

But judging from your frantic use of CAPS, I would bet you'd be back there in Salem decrying the replacement of good old fashioned dropping colored balls in urns with the new and highly techonologically dependent 'marking on paper with quill pens method' with all it's attendent things you didn't want to have to learn about.



"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

don't be silly (none / 2) (#197)
by Estanislao Martínez on Mon Oct 27, 2003 at 05:24:31 PM EST

mechanical voting machines are quite as complex (as far removed from entropy) as an e-voting machine.

This is not so under many criteria. The operation of mechanical machines does not involve invisible subatomic particles (i.e. electrons) moving around through devices which modulate the flow in hard-to-understand ways (e.g. resistors, transistors, ICs) in order to create a general-purpose computing device which, then, must be "programmed" into acting like a particular abstract algorithm.

A mechanical machine, on the other hand, involves visible physical objects moving and exerting forces upon other such objects, at a human-sized scale. Mechanical machines, even if they can be quite complex, still are easier to understand than computers. Hell, we've had complex ones for hundreds of years.

Further, paper ballots can be easily hacked with counterfit [sic] copys [sic], a pen, and access to a ballot box.

Heh. You're obsessing on the "technology" used by traditional voting (paper, pen, boxes) and disregarding the practices and procedures that go behind it (polling place procedures, monitoring by all interested parties, etc.)

Nobody's claiming paper voting isn't subject to fraud. The claim, which has been made in many threads in this article, is that paper voting is more transparent, for many interlocking reasons: the rudimentary technology involved makes for minimal barriers to the understanding of the voting process, and thus allows a large number of people to monitor it, with no preference for those who have superior access to technological experts. The main mechanism against fraud is this broad, unrestricted observation and participation by people from the whole range of interests that participate in the polls, not some hard-to-understand public-key triple-encryption mechanism. Fraud, if it occurs, is either minor (because most participants in the process have very little power over it) or obvious (because major fraud can only be accomplished by NOT allowing the conditions described above to be in place; e.g. excluding all interested parties except one from monitoring the process).

Nor do you have to "hire" a computer engineer to tell whether a piece of open source code contains vulnerabilities or trojans, there will be plenty of people capable of looking at the code and reporting if there are problems.

It is unessential to my argument whether the engineer is "hired"; substitute "get" or "find", if you may. The point is that there is a body of expertise involved there, that can be done away with.

In any case, you've failed to address what happens when two parties contesting an election each "finds" its own computer experts to testify in their favor.

--em
[ Parent ]

Don't be ignorant (none / 1) (#202)
by michaelp on Mon Oct 27, 2003 at 09:38:57 PM EST

The votes go into a database at some point, Catherine Harris wasn't counting every ballot personally before registering the election. From the precints come numbers that are reported and recorded up the chain. Records of reports is what a database is (even if it is just a massive pile of paper).

At some point someone writes the number down or enters it into a database, these numbers are what the Setry. of State looks at when assigning an election victory. At some point you have to trust some data entry system. It actually would be safer to have the only way into the database be the voting machine.

This is not so under many criteria.

O come on, stop playing word games. Mechanical voting machines depend on the an invisible force not even fully understood, are affected by the difficult to evaluate or correct for build up of microscopic debris over time, and the slow loss of metal atoms with each and every punch. One can obsfucate either way, computers are not more difficult to understand than mechanical devices, they are just new.

Fraud, if it occurs, is either minor (because most participants in the process have very little power over it) or obvious (because major fraud can only be accomplished by NOT allowing the conditions described above to be in place; e.g. excluding all interested parties except one from monitoring the process).

You are mixing two procedures here: reliable data entry, whether on paper or touch screen, requires similar local security to prevent fraud. Public examination of the code used will prevent major fraud, just as public examination of the paper ballots used is needed prevent major fraud via form design. Transport of the data after voting requires further security, this is where an e-vote can keep a record of whether it has been altered & where it was created, while each paper ballot cannot. Further, security in this stage for paper requires gaurds and trucks, whcih provide many stages when the process can be hacked. And so on.

E-voting needs to be done right, just like paper and mechanical voting does, and many of the same processes apply while e-voting adds additional means to prevent the common hacks used on analog voting.




"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Hi (none / 2) (#204)
by Everett True on Mon Oct 27, 2003 at 10:06:43 PM EST

However did they cope in 19th century America with the "low information density of paper"? Did the election system work, or not? Answer: it did.

Then somebody brought in mechanical voting machines. Fine. And what happened at the last election thanks to all this increased complexity? Cast your mind back, I'm sure you can remember all the problems caused by the dreaded mechanical voting machine's increased complexity.

The point here is: voting with a simple paper ballot works, and has a proven history, and is easily understood by everybody from 95yo voters to punk kids in the Bronx. Can you understand why this is almost infinitely more transparant than a complex computer voting system? It can be as *open* as anything, but that doesn't make it as *transparent*. Perhaps you can also appreciate that in a democracy it is extremely important that the voting system be *transparent* and not merely open. Mere openess with less transparency, and greater complexity, results in enormous problems with only a few people understanding how the system works, and varying views on how it does work. If you don't believe me, visit any technology site and see the arguments over how this or that works.

More than that, you can cry about "luddites" and "hatred of progress" all you want, but until you can tell us why one system proven to work and with a long track record of success should be replaced by another unproven system that is orders of mangitude more complex and less transparent, then you're on the losing side. Change for change's sake is not a good idea - its thoughtless gee-whizz revolutionaries like yourself that cause all the fuck-ups in this world.

And before you say that the Florida problems show that the mechanical voting machiens need to be replaced, I would remind you that there's obviously a stronger argument for rolling back their use (caused by problems that would only be intensified and worsened by a computer system) for the simple and successful paper system that went before.

Today is gonna be the day that they're gonna throw it back to you!
[ Parent ]

Wow, get a clue. (none / 1) (#207)
by michaelp on Tue Oct 28, 2003 at 03:12:57 AM EST

However did they cope in 19th century America with the "low information density of paper"? Did the election system work, or not? Answer: it did.

THe "proven history" of paper ballots is a history of one abuse after another. Search on 'election fraud' in American history. Actually, election fraud really got going in America a big way with the advent of the secret ballot in the 1880s (which was near the end of the 19th century, by the way), but that is another issue. Are you really holding up the 1800s as a shining example of pure elections? Some brief time prior to the latter half, I take it? Add "Hayes" to your search above for enlightenment.

its thoughtless gee-whizz revolutionaries like yourself that cause all the fuck-ups in this world.

Yeah yeah, from paper ballots (look up the origin of the word 'ballot', hint: it ain't papyrus) to electricity to airplanes to 'putas, us thoughtless gee whiz types have just made the lives of you stick in the mud and complain about moving forward types worser and worser. But if you'd been paying attention, I've been arguing all over this comment stream against hasty adoption of poorly implemented systems from shysters like Diebold.

replaced by another unproven system that is orders of mangitude more complex and less transparent,

Uhh huh, ok explain in detail how the counts from the precincts make to the secretary of state's office in a statewide election. Don't forget to count the number of security experts you need to trust along the way, the form the vote counts are held in while they move from the precincts to the capital to the hand of the expert charged with certifying the results. Please note that if they move by phone or wire, you have exactly the scary complexity of moving data via electrons that EM is so afraid of above & note that the ballots can be copied, boxes can be lost and when the numbers from the ballots are recorded, they can be altered.

And note that finally, they are entered into a database (whether it is bunch of marks on paper, a calculator, or *shudder* one of them computer thangs, it is a place where a whole big amount of data is kept and added up), the elections officer reads off the totals but doesn't count all the papers herself or memorize and add up all reports from the precincts in her head (what a case of trusting an 'expert' that would be!).

So think again why they switched to paper ballots from colored balls back there in Salem (you did look it up, right?): the demands of democracy had outstripped the voting technology. Now paper ballots are fine when you have two canditates for each seat and maybe a couple issues. Especially if you vote a straight party ticket (along with all your dead relatives & pets).

But if you want to have numerous candidates as well as referenda, the paper soon gets unwieldy: the font gets smaller, the number of items per page gets larger, and increasing numbers of citizens can't make out what they are voting for. Also the expense of printing, voting, trucking, and tabulating all that paper puts limits the number of elections & drives the cost per election up. You have to employ increasing numbers of people to handle the mass, which adds up to increasing chances for the unscrupulous to taint the vote.

So the paper issue only seems "orders of magnitude more transparent", really it's a highly complex system, very prone to fraud, with a long and ongoing history of fraud, that serves to preserve the 2 party status quo. With e-voting done right, we can have more issues to vote on more often, more candidates on more ballots, better assistence for the elderly and the blind to vote, ability to track votes from their entry to their count, and assure their authenticity and anonymity (if we really need to keep that new fangled notion that has enabled so much fraud to occur).

But it has to be done properly, a big part of which is opening the code for public scrutiny. Just like writing was in the 16th century when paper ballots were intrduced in this land (no doubt to the gloom and doom cries of colored ball devotees), there is nothing to coding that ordinary citizens can't understand & plenty of folks who do understand it (& who disagree on the issues) who can explain how it works to us directly & point out any chances for abuse.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Maybe you should consider (none / 2) (#209)
by Everett True on Tue Oct 28, 2003 at 08:41:05 AM EST

That a successful and transparent system makes fraud easily detected and dealt with. I can't see why your electronic voting system (currently implemented on win98 machines and such) would necessarily be any more secure, but it would certainly be a hell of a lot more difficult to detect fraud and successfully deal with it. That instances of fraud with paper ballots were simply and swiftly dealt with, and that paper ballots have not at any stage fundamentally corrupted the democracy of the USA, is an argument in their favour, not their disfavour.

Like it or not, your ideas are more complicated, less transparent, and therefore more susceptible to abuse and argument (with those on different sides picking their "expert witnesses" on either side of the argument). It hugely muddied the waters.

As for having multiple ballots and reams of paper, so what? *Everybody* understands it, and it has worked perfectly successfully for more than 200 years. If the computing systems in a fucking nuclear power station can't be secured by the government, I don't see why we should trust the election system to same.

This is the point: paper ballots work. Suck a fucking dick as you flinch from that uncounterable fact. Your alternative, nomatter how "open", is beyond the ken of 99% of the population, and that means it is DISQUALIFIED from use in any sane democracy. I don't give a fuck if they "could" understand it with a little work. The point is, they don't, they won't, and that's fucking that. Take your antidemocratic yearnings to Soviet Russia!

Today is gonna be the day that they're gonna throw it back to you!
[ Parent ]

Grow up (none / 2) (#211)
by michaelp on Tue Oct 28, 2003 at 11:05:32 AM EST

That instances of fraud with paper ballots were simply and swiftly dealt with, and that paper ballots have not at any stage fundamentally corrupted the democracy of the USA, is an argument in their favour, not their disfavour.

Paper ballots have been frequently & systematically compromised since they were introduced, learn your history & their limitations have drastically altered the ideal of democracy: look up "voting a straight ticket" to learn how.

Like it or not, your ideas are more complicated, less transparent, and therefore more susceptible to abuse and argument (with those on different sides picking their "expert witnesses" on either side of the argument). It hugely muddied the waters

Are you talking about the butterfly (paper) ballot again?

If the computing systems in a fucking nuclear power station can't be secured by the government, I don't see why we should trust the election system to same.

Since paper ballots have been far more frequently hacked than nuclear station computing systems, your logic above favors my point rather than yours.

This is the point: paper ballots work.

Yeah, they work to perpetrate the two party system, woo hoo. But the real point is: computer voting is coming. Better figure out so we get it right.

Suck a fucking dick as you flinch from that uncounterable fact.

Indeed.

"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Hey (none / 3) (#213)
by Everett True on Tue Oct 28, 2003 at 11:41:11 AM EST

Maybe you should understand why having a physical record (eg ballots) is a good idea. Imagine the following conversations:

Rep: Hi, is this your retina scan/public PGP/uberl33t voting code?

Voter: How the fuck should I know?

OR:

Rep:Is this your signature on this ballot?

Voter: Why, yes, it is.

Your points manifestly fail. You see, voting should have a physical record. It should be easily and transparently understood by all. You have completely failed to address either of these points.

If you think that computer voting will get rid of the two party system, then I am doubly against computer voting, because I don't want a political views and systems shoved down my throat that pretend to be mere voting systems.

I don't think computer voting, with its lack of a physical record and hence vulnerability to ""SELECT COUNT(*) FROM ballots WHERE candidate = 'hitler';" attacks are possible without recourse to a physical record.

See, I've no problem with modern technology. With OCD Machines being used to count ballots, and computers making the task easier in general. What I object to is this desire to fundamentally change the lynchpins of democracy. A sane democracy uses a transparent and easily understood voting system with a physical record. Your alternative suffers because it is *technologically inferior* and much more complex and less transparent. Paper ballots, which have held the democratic system together for over 200 years without any major problems, and - in the event of electoral difficulties - they have a clear and simple physical record.

Understand this: Democracy doesn't work if people don't understand and trust the voting system. If they don't, its a FAILURE. You've failed to grasp the simplest points and are trying to push an inferior solution, one that's untried and untested.

"Computer voting systems" are not coming, at elast not here in Britain, where we had a right good laugh at the difficulty of the americans with their absurd mechanical machines. I've no doubt that, should they also introduce computer systems, we'll have ever more laughs as they bring expert witnesses to the court for each side and scandals abound and there's no physical record of voting and voter's confidence in the elctoral system collapses. And we'll still be using a simple paper ballot, a stubby pencil, wrought iron boxes to hold the reciepts, and tossing a fucking coin in the event of a draw rather than going to the courts. And our democracy will be much the healthier and more transparent and trusted because of it.

Today is gonna be the day that they're gonna throw it back to you!
[ Parent ]

So you trust computers in the hands of your (none / 1) (#214)
by michaelp on Tue Oct 28, 2003 at 01:08:48 PM EST

bureacrats for counting votes, but not in the hands of your voters:

With OCD Machines being used to count ballots, and computers making the task easier in general.

Great for you.

Paper ballots, which have held the democratic system together for over 200 years without any major problems

You really shoulda done that search. Paper ballots were introduced in the US in the 1880s, and immediately were abused to limit the choices of voters because the technology allows for limited number of spaces (not to mentioned ballot box stuffing). To further compound your ignorance of history, the history of both America and Britain is full of "major problems" related to voting, you just don't know anything about them.

I don't think computer voting, with its lack of a physical record

Hohum. Why don't you try reading my other posts on this thread before going off all shock and awe? I've not said anywhere that I object to a physical record of computer voting. What I object to is the voting options being limited by what can be crammed into a paper ballot, the attendent disenfranchisement of folks inexpert at analyzing reams of paper & the rampant fraud that has been a part of paper ballot voting since it's inception. I've also objected quite strenously on this very thread to the use of closed source systems in public computer voting.

and hence vulnerability to ""SELECT COUNT(*) FROM ballots WHERE candidate = 'hitler';" attacks are possible without recourse to a physical record.

This can be done in your bureaucrats databases if they are set up poorly. This can't be done in a properly secured database & if the code was open it wouldn't be their expert against ours it would be a horde of programmers against a few company shills (like you are seeing in the Diebold case when the code was finally revealed in all it's amateurish glory).

And we'll still be using a simple paper ballot, a stubby pencil, wrought iron boxes to hold the reciepts, and tossing a fucking coin in the event of a draw rather than going to the courts. And our democracy will be much the healthier and more transparent and trusted because of it.

O yash, everything is just so peachy in Albion:

The Council elections were fraught with allegations that residents had been threatened and intimidated by people wanting to have their postal votes.

Not that it makes the world papers, like most of the world your much more interested in what goes on in Florida than in your own backyard.

But pretty soon some American company will have an internet based voting system that will help your little island deal with this rampant fraud in your primitive voting system.

Till then, cheerio and enjoy the "labour" monopoly.



"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

What? (none / 0) (#220)
by Everett True on Tue Oct 28, 2003 at 07:13:30 PM EST

You council elections example is disingenuous. The problems there were caused by the nature of postal voting and administrative cock-ups. Nothing whatsoever to do with any inadequacies of the paper ballot. Unless you plan to abolish postal voting and limit the franchise of people in the unfortunate position of being unable to make it to the polls to those with internet access, or something equally absurd? What about the 70% of people who don't have the internet, or don't know how to use a computer, or can't afford one?

Your whole argument would be reasonable if everybody had a computer, everybody could easily download sourcecode and understand how it works, and so on and so forth. You are arguing for decreased transparancy, not increased. "A group of computer programmers" (LO-fucking-L)are in adequate abundance such that everyone is a computer programmer and able to understand the electoral system in which they are participating. You seek to exclude people and disenfranchise them by confusing and complicating what must, by definition, be easily understood by all, in principle.

As for the history of the paper ballot, I think it has an excellent one. Of course there has been fraud, especially in the different climate of the 19th century. But never has this fraud, in Britain at least, been particularly major, and these days it is utterly unheard of. You can search for all the irrelevant local councils and obscure 18th century examples you want, but they very facts that the voters trust the current system and we're still in a functioning and healthy democracy in which to discuss this shows that paper ballots have indeed been successful.

Your alternative is untried. Your alternative decreases transparency and trust of the electoral system. You alternative is unneeded, and in the end a crock of shit.

Today is gonna be the day that they're gonna throw it back to you!
[ Parent ]

See above (none / 0) (#231)
by michaelp on Wed Oct 29, 2003 at 01:57:35 PM EST

we've been using boxes as mysterious to the average person as computers & as hackable if unauthorized (expert)persons gain access, since 1892.

Paper breaks down under the numbers in large cities, which is why the Myers Automatic Booth was introduced. People just can't count all the ballots fast enough & esp. given the often very large number of candidates & issues.

The point of the British elections is that low tech solutions are no protection against fraud. According to the British press, the postal ballots are causing widespread consternation, even if you aren't aware of it.

Rigging doubts over postal ballots

"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

What you need is a simple example. (none / 0) (#221)
by la princesa on Tue Oct 28, 2003 at 09:25:37 PM EST

Essentially, using electronic voting exclusively would be the same as using paper ballots for American elections written in Swahili.  Sure, those paper ballots are transparent, but only a tiny subset of the population can understand them.  Even if electronic voting used open source software, only a small subset of people can understand a given computer language enough to know what it's doing to arrange the voting.  Add to that the fact that many Americans don't trust computers and would be suspicious of computer voting, and you'd end up with a scenario where only techno-literate types trusted the voting system and voting plummeted from 50% of the population to probably something like 10-20%.  

At best, if you are to use electronic voting, it must print out the ballot and require a written signature from the voter.  There are ways to do this that can preserve the anonymity of voters, though I suspect that anonymous or not, it may be the only way to allow electronic voting without denying most of the population access to transparent voting.

___
<qpt> Disprove people? <qpt> What happens when you disprove them? Do they disappear in a flash of logic?
[ Parent ]

You're mixing up the design with the technology (none / 0) (#228)
by michaelp on Tue Oct 28, 2003 at 11:46:38 PM EST

paper ballots can be incredibly complex and disenfrancise large numbers of voters. The butterfly ballot design is an example of a commmon problem in many multi-canditate races, it might as well have been in swahili to many of the voters.

Touch screen interfaces can be simple, graphical, and remind folks if they double punch or miss a row.

Both need security provided by experts after the vote is taken, and both need vetting before the vote to be sure there is no bias.

Add to that the fact that many Americans don't trust computers and would be suspicious of computer voting, and you'd end up with a scenario where only techno-literate types trusted the voting system and voting plummeted from 50% of the population to probably something like 10-20%.

Rank speculation. We've used boxes as black to the average person as a computer since 1892 and there has been no corresponding linking of voters uncertainty with how the technology works with voter apathy. More likely your apathy is caused by disconnect between the vote and actual change that the dominance of the paper ballot has perpetrated since it's inception.

Frankly, while the particular attacks on Diebold's implementation of e-voting are right on target, generalizing them to an overall war on e-voting is a silly distraction from the real problem: why with all the voting on all the various machines and papers we do, we have so little actual democracy.


"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]

Going back over this thread... (none / 1) (#219)
by Estanislao Martínez on Tue Oct 28, 2003 at 06:53:58 PM EST

Going back over this thread the following thing stands out: you've failed to directly address any single point I've made.
  1. I have questioned the need for computer voting. You have not addressed this at all; the closest you've come is to say that it could offer faster, more accurate counts. But this is still not a demonstration of need. Do we really need faster counts in elections? Hell, we can afford to wait far more than we already do in most cases. Do we need more accuracy? In some cases, yes; for contested precints we need maximally accurate counts. But, the fact that such districts are contested immediately means that we also need maximally transparent counts for them; and nothing can be more transparent and accurate than a bunch of people slowly counting and recounting the physical ballots, monitored by the representatives of the interested parties.
  2. You've utterly failed to address the argument about the desirability of transparency in an election. Related to this: maximizing the opportunities for participation and monitoring in the production of the results. If you believe maximizing these should not be an overriding desideratum for electionary processes, you should come out and say why you think so. But such an argument will have little to do with databases, crypto, open sores voting software, etc., and all to do with democratic values.
  3. You insist on showing various instances of problems with paper ballots as an argument for computer systems. This is disingenious because:
    • Nobody has argued that paper elections are not subject to abuse. The argument, rather, is that it is easier to detect abuse in a paper election. Hell, most bad paper elections, e.g. in the third world, are obviously so to observers; there is reason to distrust an election where the organizers e.g. deny access to the opposition.
    • You are implicitly comparing bad instances of paper elections, which have actually happened, to an ideal computer based election that has never been done, using hardware, software and procedures that do not exist. But of fucking course a computer election that is not abused is always better than a paper one that is abused. The point is about potential for abuse. Hell, not even that: the point is about the detectability of abuse-- transparency, again.
  4. You want to conduct this conversation at the level of technologies when I've carefully insisted on doing so at the level of social practices. An election process (note how I say "process" and not "technology", dammit) is going to be at the center of conflict by various parties in a society. Its aim is not primarily to count votes; its aim is to settle a conflict between various interests. Technically accurate counting is subsidiary to that end. It is crucial to be able to show that the losers lost to anybody; any possibility for doubting it, even nasty, petty ones, must be eliminated if possible.

    Thus the "conflicting experts" scenario that I brought up, and you've ignored. Let me restate it, with minor modifications. Let's suppose we have an e-voting system right out of your dreams (did I already mention that no such thing exists?). There's a race with two candidates: Mary, the incumbent, and Terry, who the press favors, and the polls show him winning the race. Opinion, however, is VERY POLARISED between their supporters. Mary wins cleanly, but by a very narrow margin. Terry contests the result in the courts; he hires experts to testify in his favor that the voting mechanism is subject to abuse. In fact, it isn't, since it comes right out of your dreams. So Terry's experts are flat out lying; but in the courtroom, since the judges aren't experts, it comes down to the word of one group of experts vs. another (who, as it happens, split perfectly on whether they support Mary or Terry). The judges can't judge. But whatever decision they make, a large part of the population won't believe it anyway.

    This is, again, transparency. The fewer people that need to defer to experts, the better. It is up to you to tell us why we should accept a situation where the only way most people can ascertain whether a result is fair involves relying on a third party that can be done away with.

In short, you should actually address the points that are being made.

--em
[ Parent ]

The points I addressed (none / 0) (#232)
by michaelp on Wed Oct 29, 2003 at 03:25:47 PM EST

first, were the FUD in this:

if it's computer-based, it therefore must inevitably be better, since as any seller of technological solutions knows, the only thing better than complicated technology is more complicated technology. Right?

I have questioned the need for computer voting.

We need speed and accuracy in large precincts with large numbers of issues and candidates. This is why we've been using 'black box' voting in such places since 1892. This point was addressed here:the old machines are worn out and expensive to fix and/or dirty and hard to keep clean, etc.

You've utterly failed to address the argument about the desirability of transparency in an election.

No, you've just abused the term, using it to mean 'everything but computers' as if the decision is transparency or computers.

But a real dichotomy is transparency vs. efficiency, and the decision to reduce one to increase the other has been made many times before, which was the point of introducing the mechanical voting machines in my original response. *And*, we now have accessibility added into the equation by sec. 508 of the rehabilitation act (under which paper ballots may well be considered a violation of federal law): paper ballots can't easily be made accessible, computer voting interfaces can.

Further, e-voting can be made as transparent as paper voting, even to the point of printing out a paper receipt for the user to check & can be made dramatically moreso by posting the tallies in real time.

Also adressed here:

http://www.kuro5hin.org/comments/2003/10/21/2367/2543?pid=162#182
http://www.kuro5hin.org/comments/2003/10/21/2367/2543?pid=204#207 & here http://www.kuro5hin.org/comments/2003/10/21/2367/2543/208#208

You insist on showing various instances of problems with paper ballots as an argument for computer systems. This is disingenious because: Nobody has argued that paper elections are not subject to abuse. The argument, rather, is that it is easier to detect abuse in a paper election.

And I've argued that this is incorrect (obviously in such an argument, one needs to use examples), it is easier to detect abuse in a properly designed e-voting system: because you can keep a record of every time a vote is accessed, and any changes therin. You can do this at all in mechanical voting machine, we've seen the problems with it in punch cards, and it requires similar examination by experts in forensics to tell if paper ballots have been accessed and or changed. The examples of numerous abuses in paper ballots is to show that such detections need to rely on witness accounts and as such are likely only a small subset of the total. E-voting systems can be (now) designed to send alerts on any access attempts, along with a record of what was attempted and in many cases by whom, a dramatic advantage over paper.

Addressed here: http://www.kuro5hin.org/comments/2003/10/21/2367/2543?pid=213#214
http://www.kuro5hin.org/comments/2003/10/21/2367/2543/207#207
http://www.kuro5hin.org/comments/2003/10/21/2367/2543?pid=197#202

You want to conduct this conversation at the level of technologies when I've carefully insisted on doing so at the level of social practices.

Ahh, no, EM, you introduced technologies in your original post: since as any seller of technological solutions knows, the only thing better than complicated technology is more complicated technology. Right?

Then you raised it again here: This simply because any computer system will be orders of magnitude more complicated than a mechanical or paper-based system. Even if the source code and the hardware of the machines were publicly available, the skill set required to evaluate them is just too prohibitive.

and yet again here:

The operation of mechanical machines does not involve invisible subatomic particles (i.e. electrons) moving around through devices which modulate the flow in hard-to-understand ways (e.g. resistors, transistors, ICs) in order to create a general-purpose computing device which, then, must be "programmed" into acting like a particular abstract algorithm.

A mechanical machine, on the other hand, involves visible physical objects moving and exerting forces upon other such objects, at a human-sized scale. Mechanical machines, even if they can be quite complex, still are easier to understand than computers.

Which are all speculations apparently drawn from your own personal experience. I doubt the number of folks who fully understand how a computer works is actually less than the number of folks who fully understand how a mechanical voting machine works, esp. since the voting machines are being replaced because no one makes them & few can fix them any more! As we see in both the example of the butterfly ballot and the punch card machines, there are a small number of experts who understand how ballot design can skew results or how punch card mechanics & maintainance can skew results.

In contrast, as seen from the Diebold controversy, there are thousands of folks who know a computer well enough to see the problems with the Diebold technology. In contrast, the 'source' for the punch card readers has long been publically available, yet only a few experts could point out what a hanging chad might mean or explain why they hang.

Addressed here: http://www.kuro5hin.org/comments/2003/10/21/2367/2543/207#207
http://www.kuro5hin.org/comments/2003/10/21/2367/2543/132#132

The point is, any voting 'process' depends on some technolgy to effect it, and you have been consistently arguing against a particular technology, not a process.

Thus the "conflicting experts" scenario that I brought up, and you've ignored.

I've addressed this red herring several times. The punch card argument is one example, the butterfly ballot is another. Any voting system is going to bring out "conflicting experts". At least with e-voting these folks will have to point to & explain actual code (which necessarily follows logical structures making it much easier to demonstrate what happened than fingerprints, voice patterns, handwriting recognition, and certainly witness testimony) to discuss abuse, rather than speculate about the effects of putting places to mark the x in logical order or how hanging a chad can be.

"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]
Fine, be deaf and blind... (none / 0) (#133)
by Eater on Fri Oct 24, 2003 at 06:43:22 PM EST

Because that is exactly what you are if you think counting several million votes by hand is better than doing it electronically, or that it is easier to alter the results if they are in WELL-DESIGNED electronic form than in paper-form.
Sure, the human way is more transparent, but a well-designed, well documented computarized way can be just as transparent.

Eater.

[ Parent ]
Its a trust thing (none / 2) (#151)
by gcmillwood on Sat Oct 25, 2003 at 11:32:09 AM EST

A computerised way can be transparent, yes.  But when the computer system counts the votes in 10 seconds flat and tells me that candidate X got 32 votes and candidate Y got 1053 votes, I need to trust the makers of the system that it got it right.

When doing things the old fashioned way (i.e. paper and pencil, and 100 or so people counting the little slips of paper) there are two major things which increase my trust in the results:
1) It isn't just one person counting the results.  There are people checking the results that the first person produces.
2) Even if one person starts 'faking' the results, the effect of this is going to be (relatively) minor, as they are only 1% of the people doing the counting.  (Of course this is still not a good thing, and we would all hope this would be noticed and dealt with).

Computerised systems are very nice, but to many members of the public they are just 'black boxes' which produce a result at the end of the night.  Joe Average doesn't know how they work, so why should he trust the results it gives?

(Unfortunately Joe Average probably doesn't vote or care about the results, but that is another discussion really).

[ Parent ]

You mean to tell me (none / 1) (#157)
by Eater on Sat Oct 25, 2003 at 04:31:40 PM EST

That your Joe Average doesn't know how computarized voting results are counted up, but DOES know how it is done conventionally? Now granted, he might understand if you explained it to him, but then again, he could understand if you explained how the computerized system worked too, it would just take an extra hour or so of explaining.
I think the main point that you're trying to make is that with computers it's just as easy to inflict massive damage to results as it is to inflict minor damage (while this problem doesn't exist with conventional voting), and I think this is a valid point about the weakness of modern computerized voting, but much like electric cars, computerized voting is today in its infancy, and modern cryptography technology shows that it is possible to provide computer systems with reliable security, while it is highly unlikely that a conventional voting system will ever be able to exceed its modern-day efficiency and speed (not to mention accuracy - after all, people can forget to carry the 1, but computers will either forget it every time - in which case it will be detected - or not at all). So yeah, it could use a little work, and the current implimintation is certainly not what should have been done, but that doesn't mean that computerized voting is inherently inferior to conventional voting.

Eater.

[ Parent ]
How complicated is it? (none / 1) (#167)
by gcmillwood on Sun Oct 26, 2003 at 12:36:52 PM EST

It is easy to understand how conventional vote counting works.  Each person gets given a single piece of paper to vote with, they pick their chosen cadidate and put it in a sealed box.  So you end up with a big box full of bits of paper with someones name on, and the election officials count up the bits of paper with the same name on.  Joe understands all of this quite easily, with maybe a minute of explanation (if he is particularly dense).

With a computer on the other hand, Joe has to understand many things which are considerably more complex.  Just a few examples:
How can each person only vote once and not multiple times?
How does the computer remember each persons vote without forgetting a previous persons vote?
How can Joe be sure that the result given by the computer wasn't predetermined before the election?

Now I'm not saying that these things are not an issue with conventional voting, but there are simple, highly visible actions to answer these questions that Joe can see, understand, and therefore trust.  With a computer he can't see it doing these things and therefore has no reason to trust the system.

You are of course correct that computerised voting is still very young, but throwing in modern crytography, digital signatures and so on isn't going to help.  These are just yet more obstacles to Joe's understanding and acceptance.

Accuracy - yes, you are right.  Computers are of course more likely to be accurate (assuming the code is correct).  But accuracy isn't worth a damn when the security is such that any techy with access can falsify the results.  That is what I mean when I say it is a trust thing - unless the voters trust the system, they cannot trust the results the system gives.

[ Parent ]

Those same questions (none / 0) (#171)
by Eater on Sun Oct 26, 2003 at 02:00:33 PM EST

Those same questions you state can be applied to conventional voting (except maybe the forgetting question). With a rudimentary understanding of how computers work, Joe can easily grasp the basic concepts in computarized voting, and this rudimentary knowledge is going to become more and more common as computers become more popular (the emerging generation already has far superior understanding of computers on average than the previous one - if this trend continues, the majority of the population will see computarized voting as just as transparent as conventional voting). And conventional voting is not that simple either - there is the whole bit about how modern-day votes are counted (popular vs. electoral), and all that junk which most people don't really understand - you can't explain all that in one minute. So even with conventional voting, if someone is stupid and doesn't care about how votes are counted, they won't know either way, but with computerized or conventional voting, if someone knows how things around them work (paper, ink, and counting up things, or how basic computer software works), if someone really wants to know, they'll be able to find out.

Eater.

[ Parent ]
I both agree and disagree (none / 0) (#179)
by gcmillwood on Sun Oct 26, 2003 at 07:23:32 PM EST

As both of us have already pointed out, the same questions can be raised about both ways of voting.  But you seem to have missed the fact that these are easily and (more importantly) visably solved when using pencil and paper:
To stop someone voting multiple times they only get one ballot to put into the box.
The result isn't predetermined because dozens of people are involved with counting the votes and could/would raise the alarm is a different result was given to the one they counted.  (Note I'm discounting the possibility of vast conspiracies, ballot box stuffing, etc. though these too have easy, visible solutions).

Maybe I just know some less intelligent Joes than you, but some of my friends and relatives would have trouble understanding that the vote database cannot have been changed from when the last vote is cast to when the results are calculated, or that the database was truly empty before voting began.  Coming back to my original point this is because they have no choice but to trust the computer when it says that these things are so.  And given most peoples experience of computers, they really have no real reason to believe what it says!  It may well be a psychological thing more than anything else.  There is no real reason the computer is any less trustworthy than the election officials.

It is very true that this will become less of a problem in the future.  I code and maintain programs for a living, my mother uses the things, and my grandmother has never even touched a computer to my knowledge.  But that means a good few years will have to pass until enough of the population understands what is going on for computerised voting to become the default, rather than the exception.

The popular vs. electoral votes is really a red herring.  I don't truly understand this now (and frankly, living in the UK, I don't mind this at all).  The thing is, if this is a problem it will be a problem however Joe Average votes, be it by pen and pencil, punchcard, computerised touch screen or telepathically.  Joe just wants to know that his vote will get counted, and that no-one else gets an unfair advantage in the immediate election (the election the guys he is electing are going to have later on probably doesn't even enter his head, as you implied).

[ Parent ]

Yeah, there is a point (none / 1) (#199)
by Eater on Mon Oct 27, 2003 at 08:31:23 PM EST

Of course, you have a point in that many people inherently distrust computers today, but my point was that this doesn't make them irreperably unsuited for transparent voting. The purpose of the remark regarding the electoral system was to point out that the REALLY stupid Joes won't understand how the election works anyway, regardless of how the votes are counted, which means that those that do understand will be much more likely to also be able to understand why a certain computerized voting system is not easy to fool. The actual bare facts of how a computerized voting system works (assuming a person can get past the initial intimidating part of "oh my god I have to learn about computers") are only slightly more complex than the systems used to prevent a person from voting twice, and both may well be ill-understood by voters who don't really care, but I believe that a majority of voters, if they really want to know how a computerized voting system works, will be able to understand it with little difficulty - the only real barrier is the psychological perception that computers cannot be trusted.

Eater.

[ Parent ]
gah (none / 1) (#196)
by Estanislao Martínez on Mon Oct 27, 2003 at 04:55:33 PM EST

With a rudimentary understanding of how computers work

You've disqualified yourself from this discussion already by making computer expertise a requisite for understanding election. (And I'd bet that the list of things that you have in mind only count as "rudimentary" in the autistic world-view of a computer g**k.)

--em
[ Parent ]

You're awfully quick to insult (none / 0) (#198)
by Eater on Mon Oct 27, 2003 at 08:23:11 PM EST

In order to understand conventional voting, you need a rudimentary grasp of mathematics, organization, and government. You also need to know how the convoluted American electoral system works. And I don't appreciate your sad attempt at an insult either - unlike the majority of people here, I do not take well to such complete lack of manners.

Eater.

[ Parent ]
You actually don't need to know those things. (none / 0) (#203)
by la princesa on Mon Oct 27, 2003 at 09:54:47 PM EST

Transparent paper, or rock, or suchforth voting is and will ever remain far more comprehensible to a wider selection of humanity than any computer-mediated voting system one could devise.  The point is transparency, access and accuracy for the largest amount of voters.  Computer voting can't assure the first and second things to more than a tiny fraction of the population, though it might well best paper voting in the third thing.  However, it wouldn't do so to such an extreme degree that people ought to entirely replace paper voting.  Computers are not access-for-all objects.  Computer program source code is not immediately comprehensible by the bulk of the population, nor should it be just to ensure transparency in voting.  Ultimately in a democracy, transparency should come first, and maximum levels of access.  It boggles the mind that you'd work to deny much of the population both access and transparency out of a mistaken notion that computers always do it better.  

___
<qpt> Disprove people? <qpt> What happens when you disprove them? Do they disappear in a flash of logic?
[ Parent ]
No (none / 0) (#208)
by michaelp on Tue Oct 28, 2003 at 05:13:06 AM EST

computer voting has paper beat hands down for accessibility too.

IOW, Jaws don't read paper. Not to mention the design bias of multi-candidate paper ballots. Fact is, the paper ballot helps maintain the two party status quo by making multi-candidate ballots unusable. If designed 'right' it may even skew an election without any need for counterfit ballots or ballot boxes dumped in the river.

far more comprehensible to a wider selection of humanity

Well, a "wider selection of humanity" doesn't even have a slightly democratic govt.

Computer voting is certainly doable for the US, and if done right will likely will lessen the rampant fraud inherant in the paper ballots, not to mention making it possible for actual democracy to replace the representative corruption (the real reason for all this FUD about e-voting?) we have now.



"Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed."

[ Parent ]
Literacy (none / 0) (#222)
by Eater on Tue Oct 28, 2003 at 09:35:23 PM EST

During the 18th and 19th centuries, the requirement to be literate in order to vote was often a major obstacle in the way of would-be voters - not only was voting not transparent to them because it used this "new" technology, they couldn't even vote! And computerized voting doesn't even prevent you from voting. You do not need to be able to read code to understand the basic principles of how a computer program works. You do, however, have to take someone's word for it, and assume they are not lying when they tell you that a certain algorithm prevents changes to the database after votes are registered. But then again, every time you go to vote, even if you understand how voting is done, you have to trust the government not to just toss all the votes and say Bush won by a landslide with 90% of the votes (with the Democrat getting a mere 80% of course).

Eater.

[ Parent ]
You are wrong. (none / 1) (#223)
by Estanislao Martínez on Tue Oct 28, 2003 at 10:47:49 PM EST

But then again, every time you go to vote, even if you understand how voting is done, you have to trust the government not to just toss all the votes and say Bush won by a landslide with 90% of the votes (with the Democrat getting a mere 80% of course).

The electionary process has set up in such a way that participants can monitor the whole thing. If this is not so, e.g. if the government does not allow the opposition to monitor the polls on the floor, it's not a valid election.

--em
[ Parent ]

Appearently not (none / 0) (#233)
by Eater on Thu Oct 30, 2003 at 12:11:00 AM EST

Since, after all, there is still controversy over the last election, direct evidence contradicts what you're saying.

Eater.

[ Parent ]
Don't be so afraid (none / 1) (#212)
by sellison on Tue Oct 28, 2003 at 11:21:10 AM EST

Diebold is run by good, moral Christian people, the sort you'd want to be in charge of voting.

They might make some mistakes while they learn this new technology, but they will be honest mistakes and they will get it right.

Meanwhile, the socialist technoweenies who hacked into their private files should be prosecuted to the full extent of the law!


"No, I don't know that atheists should be considered as citizens, nor should they be considered as patriots. This is one nation under God."- George H.W. Bush
[ Parent ]

I've worked as an elector and this is scary shit! (2.88 / 9) (#130)
by Theranthrope on Fri Oct 24, 2003 at 06:02:59 PM EST

I've worked as an elector twice for the Nevada department of Elections. I'll tell working for them is a trip. EVERYTHING must be done in an exact, certain, documented way. Even more so than other government agentcies. There is a feeling that all the steps we have to go through are there for a reason: someone, somewhere has screwed something up in a past election. We work from that painful, bloody, expierence to bring a fair and accurate election.

If there were 1138 eligable voters who voted from my precinct XXXX. There better be 1138 signatures on the sign on sheets, 1138 vote "tickets", and 1138 vote counts on the voting machine(s) otherwise I can't go home for the night until the mistake is found and fixed (and I don't get overtime).

The scary thing is: dispite all the long, annoying, tedious work that we do on our end, dibold can ruin (or steal) an election simply due to their incompitance (or malice (Occam's razor, yadda yadda))!
"Turmeric applied as a suppository will increase intelligence." -- HidingMyName

That's Hanlon's Razor [nt] (none / 1) (#242)
by riceowlguy on Tue Nov 04, 2003 at 11:51:58 PM EST


"That meant spending the night in the living room with Frank watching over me like some kind of Lovecraftian soul-stealing nightmare creature-Azag-Frank[ Parent ]

Ok - let's solve the problem. (none / 1) (#138)
by lukme on Fri Oct 24, 2003 at 08:42:46 PM EST

Lets just build an open source voting system!!

It should be cheap for municipalities to build and administrate.

Any voting systems needs to be:

1) easy to use
2) easily verifiable for
a) the voter
b) the vote counter (think florida)
3) perferrably machine and human readable
The advantages of having a stand alone computer system to write the vote card is to standardize the printing. Furthermore, the printing on the card could be made in a 6-point font, where as the interface in the computer could be adjustable depending on the user. A magnifying glass could be provided in the voting booth to read such small font, hence this would be voter readable.

The ideal font would be selected such that it would be easily readable by OCR - hence machine readable.

The cards would also need a ID number so that the total count would be known, and it would be easily verified that you had the complete set of cards. Finally, the cards could be the size of business cards which would make storage more convient, and buying blank business cards could be done in bulk.

Would there be any problems with this syste?


-----------------------------------
It's awfully hard to fly with eagles when you're a turkey.
Desired Properties (none / 1) (#146)
by NateTG on Sat Oct 25, 2003 at 12:05:33 AM EST

Here are the things that a voting system should have:
  1. Guarantee no lost votes
  2. Guarantee no extra votes
  3. Guarantee no changed votes
  4. Guarantee that votes cannot be coerced.
You'll also want

1. Code independance.
That's right, the voting system has to be reliable in a way that doesn not rely on the code being correct.  Even if I wrote the code (heck especially if I wrote the code) I want it to be set up in a way that will work even the code fucks up in the middle of the process.
2. Commodity Hardware.
You should not need to rely on special technology.  Ideally any PC and Printer should work.  If you can make it cheap, you're more likely to get it into place.

If you want to put electronic equipment into the booth, it should probably also be tempest hardened.

It's a really hard thing to do well.

[ Parent ]

uhhh, been done already (none / 1) (#163)
by Alhazred on Sat Oct 25, 2003 at 10:11:15 PM EST

There is a GNU (FSF) project to do just this. I believe they have some sort of code. I haven't really followed the project closely, but at the very least they've worked out what needs to be done and built a prototype system.

And from what I understand, yes, it is REALLY hard to build something like this. In fact I would say in a sense impossible. I suspect we will have to abandon some of our preconcieved ideas about what elections are in order to build something secure, which probably means its not going to work under current legal/political systems.
That is not dead which may eternal lie And with strange aeons death itself may die.
[ Parent ]

the free e-democracy project (none / 0) (#189)
by FlipFlop on Mon Oct 27, 2003 at 12:19:04 AM EST

I believe you are referring to this.

Choice quote from the site:

We used to develop the GNU.FREE Internet Voting software and we retain a strong interest in electronic voting issues, primarily through advocating why we feel it's an undesirable advance.

AdTI - The think tank that didn't
[ Parent ]

Better than the memos -- try it out yourself (none / 3) (#160)
by radghast on Sat Oct 25, 2003 at 09:13:01 PM EST

Here's a link to the Diebold GEMS software itself, some test data and live election data, and the procedure on how easy it is to hack it.

All of this was apparently left lying around on Diebold's FTP site, according to this site -- and the live data was allegedly copied from an active voting machine during an  Alameda election!  

Too unreal.
"It remains to be seen if the human brain is powerful enough to solve the problems it has created." -- Dr. Richard Wallace

Did someone say Diabolic? (none / 0) (#234)
by trezor on Thu Oct 30, 2003 at 08:03:23 AM EST

This system is not only inheretly insecure.

This system was designed to be tampered with. Is there really anything more to add?

Appart from the obviousness of that the guy certifing this should be fired, and never ever aloved to work for any goverment agency ever again?


--
Richard Dean Anderson porn? - Now spread the news

[ Parent ]
My mirror of the memos (none / 3) (#173)
by Luminescent on Sun Oct 26, 2003 at 02:36:12 PM EST

http://ultron.dyndns.org/diebold-memos.tgz

This is hosted on my personal server, on the Boston University network. I'd rather not get a DMCA notice, but this is exceedingly important.

Feel free to post this link elsewhere.

Update (none / 1) (#174)
by Luminescent on Sun Oct 26, 2003 at 02:52:27 PM EST

Since my server's hard drive as been unstable, recently, I've uploaded the diebold memos to a server run by BU:

http://csa.bu.edu/~chrisn1/diebold-memos.tgz

Keep on mirroring.

-Chris

[ Parent ]

Can't access it (none / 0) (#188)
by FlipFlop on Mon Oct 27, 2003 at 12:08:09 AM EST

Since my server's hard drive as been unstable, recently, I've uploaded the diebold memos to a server run by BU:

http://csa.bu.edu/~chrisn1/diebold-memos.tgz

Keep on mirroring.

Just thought I'd let you know, I can't access that site right now. It should be approximately five minutes after midnight in Boston at the moment.

AdTI - The think tank that didn't
[ Parent ]

Thanks, Fixed Link Here (none / 0) (#227)
by Luminescent on Tue Oct 28, 2003 at 11:43:06 PM EST

http://cs-people.bu.edu/chrisn1/diebold-memos.tgz

That's what I get for not testing the link before posting.

[ Parent ]

So what happens (none / 1) (#226)
by KrispyKringle on Tue Oct 28, 2003 at 11:15:28 PM EST

When you get caught? I've mirrored this, as well, on my website. But I'm curious what happens if I get a DMCA letter. Do I just take it down, write back saying I took it down, and we're cool? Or do I get sued?

And, no, keeping it up and getting sued isn't something I'd enjoy.

Dan

[ Parent ]

Uncertain, but... (none / 1) (#229)
by Luminescent on Tue Oct 28, 2003 at 11:53:54 PM EST

So far anybody who's been cease + desisted has taken it down, either through their own choice, or through their ISP / university's choice.

I also have absolutely no desire to go to court over this. As such, when the cease + desist arrives, I'll take it down. The current strategy being discussed at why-war.com is to make them play whack-a-mole. Each person who is cease + desisted is to find two friends to post the files anew.

Being sued after compliance isn't outside the realm of possibility, but it's unlikely, given the negative publicity it would generate, since Diebold would like nothing better than this issue to die quietly.

Furthermore, each time somebody posts this file and then is forced to take the it down, it add evidence to the argument that the DMCA chills free speech.

Amherst and MIT were just cease + desisted today. The letter can be found here (use abiword):

http://why-war.com/features/mit-cease.doc

Of note, one MIT link is down, the other remains. Perhaps somebody at MIT isn't backing down. It isn't clear yet.

[ Parent ]

Yeah, (none / 0) (#230)
by KrispyKringle on Wed Oct 29, 2003 at 12:14:43 AM EST

I asked `micah' of why-war the folllowing after I decided to mirror: Out of curiosity, if I get a C&D letter, I can just remove the link, and they'll go away, no? I'd rather not get sued.

Her (his?) reply was: To be honest, the answer is "sort of". If you remove the link Diebold will likely leave you alone (they have yet to actually pursue anyone) but removing the link DOES NOT limit your liability. But, we doubt Diebold is willing to sue, instead it looks like they just want to intimidate.

I `chmod 700'-ed my lists.tgz for fear of a suit, but it does seem so far that they won't sue unless you resist. So maybe I'll put it back up (this was the one on my university server; I kept the one on my private server--it's much lower profile).

[ Parent ]
no paper trail = BAD (none / 0) (#243)
by wingnutx on Tue Dec 30, 2003 at 12:43:19 PM EST

I can't support any method of voting that doesn't leave behind some physical evidence of the voter's intent. A scanner card is fine. You can still count them via computer, and you have your hard copy for a manual recount. Touch-screen machines would be fine, as long as they print an official receipt to be collected as a backup.
tanstaafl
Diebold Met with 'Electronic Civil Disobedience' | 243 comments (231 topical, 12 editorial, 1 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!