Is pro-VVAT in the same category as anti-MMR?

Sparks

Originally posted by williamgrogan
I'm glad to see your attitude to the ICS has changed since their first press statement (20-02-04) on the matter when the anti-eVoting (sorry, anti-everyone-except-the-Aussie-Open-Source-pro-paper one) forum slagged it off.

That'd be the press release where the ICS said that a full review of the system was needed and they were glad to see the government call for one? Okay, proof that people were criticising that stance, if you please.

As it so happens I have yet to read the latest ICS press release of the 4th? (I couldn't find it on their web site) if someone could point me to it.

In politics, the Evoting thread lists it here.

Martin Cullen is the Minister. Presumably he is an expert Minister. Remember you can always hire in experts!

Presumably? Not a terribly sceptical attitude. More fawning I would have thought. And incorrect - he was not selected on his merits, remember, nor elected - he was chosen by bertie - for political reasons, by a man with no more qualifications than the leaving cert.
That does not convey qualifications to the recepient.

Maybe none of you are old enough to remember when the ICS regarded PC's as toys. When I first joined I had to fight to get PC's recognised as computers.

Actually I recall when PCs were toys. They've not been usable as serious tools until relatively recently, and even today they're not usable for really serious tasks, nor are they the most common computers in the real world.

williamgrogan

Actually I recall when PCs were toys. They've not been usable as serious tools until relatively recently, and even today they're not usable for really serious tasks, nor are they the most common computers in the real world.

Actually I remember when Main Frames had less power than PC’s. The ICS did not recognise PC’s as computers not because they were underpowered or any other technical reason but because of professional snobbery. If it didn’t have a cooling system it wasn’t a computer. This was a common attitude among computer professionals, I know because I worked on Main Frames and then PC’s. Furthermore this nonsense was also repeated by the now defunct Mini Computer people.

Overall this sentence of yours is the daftest I have seen in a long time.

Back to the main point.

To “prove” that all the commercially available eVoting systems are so unreliable as to need a paper parallel system I have been given “anecdotal stories” as to their failure during elections. I have pointed out that any of these stories I can get the full background on have proved to have perfectly plausible reasons, reasons that were not explained when they were used as “evidence”. Other than that I have read some “horror stories” where much to everyone’s astonishment computers have broken down! Good grief! Hard to believe a machine broke down, well that’s rare. Bring back paper, computers can break down. However even where the machines have broken down, in many, if not most cases, they had not recorded any votes or the votes were retrievable. As I’m sure all you computer experts know, a PC that you take a hammer to can still have the data retrieved from its hard disk.

You have put forward Ms Mercuri as an expert, yet she does not advocate Open Source. Why pick one of her suggestions and not the other?

I have not poo-pooed just one case; the “upset” where two Democrats were dumped was accepted as valid by their campaign manager, the selection of some memos out of context, the “Rob-Georgia” myth, the fact that many of the elections referred to were Micky Mouse elections where security could well be lax, etc.

I am disappointed by the ICS’s latest statement which seems to prove that (as one could see with their statement of the 20th of February) all this has caught them by surprise and we are looking at a knee jerk reaction.

One good thing is that the Government has not done what I thought they might do and that is to simply waste another 20 million on printers to keep you lot from making too much noise.

When I started this thread Wakefield was not “officially” discredited. He now is as is his theory. You lot had a head start on the propaganda and have obviously fooled the ICS. I think (and hope) that by the end of the matter the evidence will swing away from VVAT.

The paper records must be retained and used in a number of randomly selected constituencies at each election to audit the accuracy of the electronically prepared result

I have addressed this. Such a procedure is ONLY another test. If the computer systems are tested then another test will have no extra validity. I have already pointed out that the only way VVAT has any logic is that ALL votes are ALWAYS counted which means the eVoting is a total waste of time and money.

Sparks

Originally posted by williamgrogan
Overall this sentence of yours is the daftest I have seen in a long time.

I'm just not going to bother responding to that.

To “prove” that all the commercially available eVoting systems are so unreliable as to need a paper parallel system I have been given “anecdotal stories” as to their failure during elections.

And you've missed an important point again. I say again, because it's already been mentioned several times. VVAT allows for verification of the result via an independent source of data - which protects against both error and malicious interference.

Secondly, I've already stated that at least one commercially available eVoting system is acceptable and given references.

Thirdly, noone in this country can prove that the proposed system is reliable. No-one. Because noone has legal access to the hardware designs and software source code.

Which means that you're trying to prove that the system is reliable based on the salesman's pitch, and I'm trying to show that it may not be because of past instances where similar machines were shown to be insecure and prone to software errors.

Bring back paper, computers can break down.

That's a rather flippant attitude. I think you'll find that some tasks require that the computer not break down, or at least that the system be failure-tolerant, like fly-by-wire systems. We don't need the system to be failure-tolerant, but it must be fail-safe, or a hardware fault could cause vote data to be lost, corrupting the final result. And we don't have any way to prove that the current system is fail-safe, other than the salesman's pitch.
That's unacceptable.

As I’m sure all you computer experts know, a PC that you take a hammer to can still have the data retrieved from its hard disk.

And as I'm sure your decades of experience as a highly qualified computer expert tells you, all it takes is one hardware fault on that disk and the data is now comprimised or lost. So do the Nedap systems use RAID arrays for data backup?

You have put forward Ms Mercuri as an expert, yet she does not advocate Open Source. Why pick one of her suggestions and not the other?

From wired:

"Open source provides some layer of protection, but that doesn't mean you can remove other security mechanisms, such as the ability to do an independent recount with the system," Mercuri said.

Doctor (let's get it correct, shall we?) Mercuri isn't opposed to open source - she's just said, quite correctly, that it's not the one sole thing needed to ensure security in an eVoting system.

I have not poo-pooed just one case

And I haven't said you had, I said you'd poo-pooed all but one.
And calling them "myths" and "conspiracy theories" and other derogatory terms is just showing that you have no valid argument to counter the points they make.

I am disappointed by the ICS’s latest statement which seems to prove that (as one could see with their statement of the 20th of February) all this has caught them by surprise and we are looking at a knee jerk reaction.

You mean you're disappointed that they didn't support your incorrect view on the matter.

One good thing is that the Government has not done what I thought they might do and that is to simply waste another 20 million on printers to keep you lot from making too much noise.

And right there you just lost what little credibility you had left.
WG, your points are unsupported. You have presented no credible, logical or rational argument to support your assertions, and have found no counter-arguments other than ad hominem attacks.

I see no further point in wasting my time arguing with someone who gives every appearance of not being qualified enough to understand the issue at hand.

williamgrogan

Mercuri also said….

…. open source alone is not a panacea for the fundamental security problem with computers (this sentence of hers alone indicates an overall opposition to eVoting and in fact computerisation in general.)

A crappy open-source system that can be modified readily is no better than a closed-source system. In fact it could be worse, she said.

When you have open-source software, people can modify it and change it however they want.

Mercuri said open-source systems give people a false sense of security … (well if that’s the case maybe they should not be used, we don’t want a false sense of security do we?)

What's more, election officials have to make sure the system is implemented correctly, and have to find some way to ensure that all of the code they review is actually the code that runs on the system.

Mercuri said that open source provides no additional protection from people who are intent on putting uncertified software in machines used in an election.

But even with extra precautions, … there still is no way to catch all types of sophisticated Trojan horses that might be slipped into code.

So in summary where is the advantage of Open Source? Is it simply that there are lots of people out there who for no pay will professionally test the software?

The Evidence

The common Internet “story” of the 134 “missing votes” is even printed as an entire paragraph in a recent New Scientist WITHOUT the very important fact that there was no democratic candidate. Whoever wrote the NS story is guilty of at least laziness and unprofessional journalism in so far as they picked up a story from the web and didn’t check the background before printing it. The entire editorial is obviously lifted from the web. I said a few weeks ago that NS was “going soft” and this is another example.

In your replies you have never admitted that your “stories” of the unreliability of eVoting systems are no better than those put forward by the parents of children with Autism who blame MMR. When these stories are examined too they hold no water.

Why have you not come back and contradicted me when I point out that I have not come across ONE example of an actual fraud occurring. Not ONE!

The Logic

Those in favour of VVAT say that no amount of testing proves the software is correct and then go on to say that a random sample of electronic votes be compared to a manual count of the printed ballots. This is a test and proves nothing more than normal testing of the software. Furthermore the ballot can be printed and displayed to the voter AND still not represent what was stored on the hard disk so that also proves nothing. You keep ignoring these points as well. VVAT proves nothing at all. Its just another test.

You dismiss all of the commercially available systems and plump for the open source one because you are pro-Open Source. No other reason.

Is it derogatory to point out that people are using as examples of failures events that have a perfectly simple explanation that indicate no failure? Surely the apology should come from those trying to put a spin on these stories? Are you even prepared to admit that many of the stories are without foundation?

PS

You are throwing the irrelevancies as much as I am. I will avoid these.


As I understand it Super Tuesday was primarily done via eVoting. 10,000,000 people voted using Diebold & Sequoia machines. Ten times the number of people who will vote in Ireland's next general election. I haven’t heard of any serious mishaps and certainly no fraud, missing votes or otherwise.

http://www.internetnews.com/bus-news/article.php/3321391

http://www.cnn.com/2004/TECH/03/02/e.voting.test.ap/

http://news.com.com/2100-1028-5168670.html

All the above mainstream media report overall satisfaction with the systems and only a few glitches.

http://yro.slashdot.org/yro/04/03/02/2315211.shtml

However, above is from “News for Nerds” which says that “they were more trouble then expected” and then links to another web site which doesn’t say this at all.

http://www.philly.com/mld/inquirer/news/nation/8099840.htm

ecksor

…. open source alone is not a panacea for the fundamental security problem with computers (this sentence of hers alone indicates an overall opposition to eVoting and in fact computerisation in general.)

Another imaginative leap of logic. More like it shows an understanding of the security problems inherent in computing systems.

Have you read The Inevitability of Failure: the Flawed Assumption of Security in Modern Computing Environments ?

williamgrogan

We know computers are not perfect, but nothing is. To say that computers are fundamentally insecure ignores the fact that nothing is. It’s only stating the bleedin’ obvious. If a computer expert was to make this comment, what would we think of her?

“I won’t use computers to do my companies book-keeping as they are fundamentally insecure.”

ecksor

Originally posted by williamgrogan
We know computers are not perfect, but nothing is. To say that computers are fundamentally insecure ignores the fact that nothing is. It’s only stating the bleedin’ obvious. If a computer expert was to make this comment, what would we think of her?

So, you do admit that modern computers are fundamentally insecure and that this is actually obvious. Good good.

“I won’t use computers to do my companies book-keeping as they are fundamentally insecure.”

You're still making that inventive leap of logic. You're either doing this as an underhanded debate tactic or because you actually believe that one implies the other.

If the latter then it has been explained on many occasions why the anonymity requirements of a voting system mean that audit mechanisms used by credit card or other consumer oriented financial systems are no longer trustworthy. You even indicated that you understood this, so it doesn't make sense for you to continue equating the security of one with the other or, in this case, the reliability of something that can be easily rechecked.

Also, you said that you'd support a VVAT mechanism that didn't involve paper, but you haven't actually pointed out a non-paper based VVAT solution that stands up. (I did, but of course they are relatively new and will need to stand up to scrutiny). How is this not contradictory or inconsistent?

williamgrogan

eVoting systems are tested and are audited. You are simply arguing for more testing via paper. I have said that I have no problem with other "normal" auditing mechanisms. However there is a limit on the expense that should be incurred.

What is often missing in all the arguments against anti-eVoting, pro-VVAT, anti-MMR, anti-Fluoridation, GMO’s, spacecraft that use Nuclear Power and NP itself is balance. Balance, balance, balance. There is a risk but it’s acceptable.

There is no prefect security. To say that a computer should not be used for voting or doing accounts because they are insecure is as ridiculous as saying that pen and paper should not be used as they are also insecure. For that matter we could not fly or drive cars or get out of bed in the morning. In fact this is another parallel with the anti-MMR campaign as many anti-MMR advocates make statements such as, “no matter how tiny the risk is we should not expose our children to it.”

Anyone who is a skeptic knows that a fundamental error made by those afraid of technology is the inability to assess risk.

There are a great number of people and a large amount of money to be made out of “security”, i.e. security in the sense of policemen, security guards, insurance, and computer security. In fact newspapers, magazines and booksellers also have an interest in talking up the threats to our security. There has even been a suspicion that the anti-Virus people may be behind the creation of viruses. Anyone involved in IT security must keep pushing the agenda that more and more security is needed. However, in reality people take security to the level they deem fit.

The evidence is there that eVoting is trusted and is secure enough when taken in the context of the real world with its predominately honest politicians, public officials, physical keys, passwords, policemen etc. It is paranoid to insist on parallel paper printing, storage and counting to prove the computerised systems work.

Many of those debating these points with me give away their opinions on other matters that do lead me to believe that they see business and politics as essentially corrupt, that the world is full of those planning to steal our votes, that there are conspiracies out there where “ensuring the votes from Ohio go to Bush” means stealing those votes and not actually what is happening which is healthy democracy and the active support of individuals for political of parties. To support the political parties that most endorse your opinions is democratic and not a sign of evil intent.

The 10,000,000 ordinary people who trusted eVoting last Tuesday in America are proof that you are worrying unduly.

No one seems to want to respond to my charge that you are using exaggerated, anecdotal evidence to push your claims therefore we must take that as agreed!

williamgrogan

You have a new champion in the anti-eVoting lobby, the right wing Irish Times writer, Breda O’Brien. Her article today is headed, “Voting by paper ballot is not a badge of backwardness”.

In the age of the computer, the Government appears to find it embarrassing to vote using paper and pen and to count votes manually. Yet what kind of vision of progress is it to replace something which works well, which is replete with ritual and tradition, and enjoys a high degree of public trust, with something which people do not trust at all

Some other opinions of Ms O’Brien; she is anti-Abortion and very pro The Catholic Church

The Church made a promise last Monday to do everything necessary to serve the truth. It now needs a breathing space so that it can put mechanisms into place to keep that promise,

or

Ninety-seven per cent of sexual abuse is not perpetrated by clergy

she also thinks more interest should be shown in those priests wrongly accused of abuse

she is also anti-Stem Cell Research

ecksor

Originally posted by williamgrogan
eVoting systems are tested and are audited.

How is the proposed Irish system audited?

There is a risk but it’s acceptable.

Anyone who is a skeptic knows that a fundamental error made by those afraid of technology is the inability to assess risk.

Fine then, give us the risk analysis.

Be warned, I mean a proper risk analysis, not any sort of fuzzy waffle. I'd be very very impressed if you could give us such a risk analysis, because nobody else has been able to do this and that is a fundamental problem with the proposed system.

No one seems to want to respond to my charge that you are using exaggerated, anecdotal evidence to push your claims therefore we must take that as agreed!

What anecdotal evidence have I used? Also, if I'm supposed to respond to your charges of anecdotal evidence, then why are you using cases of 'ordinary americans voting' as a case to back up your argument?

Btw, I've never heard of Breda O'Brien, but why do her Catholic beliefs or opinions on abortion impact upon her opinions of e-voting. You may form whatever personal opinions of her you may wish based upon those beliefs (and admittedly, so would I), but you can't logically associate one belief with another in the way that you seek to.

ecksor

Originally posted by williamgrogan
There is no prefect security. To say that a computer should not be used for voting or doing accounts because they are insecure is as ridiculous as saying that pen and paper should not be used as they are also insecure.

By the way, as has been pointed out to you many times, nobody has suggested here that computers shoudn't be used for voting or for doing accounts. And even if anybody had suggested one of those things, it wouldn't automatically suggest the other. Stop putting words in people's mouths and try arguing the actual points being made.

williamgrogan

Anecdotal evidence

I again was using the plural you. The anecdotal evidence is the evidence that is bandied about that seems to indicate widespread abuse, lax security and machine errors but that when examined does not hold up. It’s quite clear what I mean. I gave the example from New Scientist where an entire paragraph was devoted to the story about the “missing votes” in Florida but did not mention that there was no democratic candidate. I have been asked on this thread to link to web sites that are full of these “incomplete” stories that are effectively lies when used to indicate problems that do not exist.

Do yee not admit that just like the anti-MMR campaign, the anti-eVoting and its fellow traveller the pro-VVAT campaign are using stories with spin?

Is VVAT anti-eVoting – yes.

As far as I can tell from Rebecca Mercuri’s statements that she is effectively opposed to eVoting because it is not fundamentally secure and that Open Source does not improve the security enough. I think, and you can disagree, that pro-VVAT is effectively anti-eVoting. This is because of added complexity and extra problems, the extra cost, the complete parallel production, storage and counting of paper votes and the ensuing disagreements about which was correct. Remember, a counted vote is unlikely to ever agree exactly with an electronic one, particularly with our complex voting mechanism.

Estimating the Risks

What’s wrong with “Fuzzy Logic”? Every day we make decisions on risks associated with our actions that are totally fuzzy. The difference between people is that some cannot estimate the risks and act on them in a real world scenario and others can. People will smoke in front of their children, drive fast with them in the car, leave them unattended in the car, feed them sh1t and make them fat and then not vaccinate them because the risks are too high.

The risks of any voting system are difficult to calculate accurately. Using the “undervote” as MIT did is unsatisfactory.

I think a sort of Drake Equation might do. The sum of the risks is related to obvious factors that vary from election to election, country to country and from time to time. The elements and their “weight” can be debated. The risks say off someone trying to subvert an election would presumably be greater in say Iraq or Russia than Ireland.

I suggest as a formula that estimates the Significant Risk of Corruption of an election as follows...

V Validation in the voting mechanism, e.g. no validation with paper
O Opportunity – access to the voting system
T Technology, e.g. ballot stuffing via a photocopier is easier than writing a program
E Election (is a school board election safer than a general election?)
C Country, e.g. Would Iraq gets a higher probability than Ireland?
A Audit mechanism
M Multiplicity, e.g. can just one event corrupt event affect the entire outcome
P Polls, up to date opinion polls act as a deterrent to gross corruption
W Does someone Want to corrupt the election?
J risk of Jail
S Security level, e.g. school board less secure than general election
R Rule of law and the public adherence to the law
G Testing of system, the more useful testing the higher the accuracy

The factors can be exact numbers or a broad estimate from say 1 to .000001. Some elements increase the risk others reduce it.

V*O*T*E*C*A*M*P*W*J*S*R*G

Recount myth

People have been arguing that there is no recount facility with the electronic votes. There never was with the paper votes. When we speak of a recount of the paper vote we speak of the need to re-count the actual physical votes that we have because of the inaccuracy of the counting mechanism (the humans). It is not a validation of the votes themselves but just of the accuracy of the count. No such recount is necessary with a computer as it can count with 100% accuracy. When people demand the need for a recount with electronics they are mixing up two things, accuracy of the count and validity of the votes that are being counted. There is no parallel mechanism with paper and we had no problem. Why demand it with electronics? Is it the man in front of the car with the flag syndrome?

williamgrogan

Rebecca Mercuri's Heath Robinson contraption.

This is a joke isn't it?

ecksor

I didn't stay "fuzzy logic" :rolleyes:

I thought you were going to back up your statement that people who are afraid of technology are unable to assess risk by leading us through an analysis that showed that the risk of failure of the electronic system was negligible. I don't see that in your post, but there's a few interesting ideas there and the risk analysis of voting systems is very interesting in its own right since I haven't been able to find reference to anyone who's pinned down a model for it.

Ok, you seem to have made a stab at assessing the probability of something, but I see no mention of impact of undesirable outcomes. It also seems to be covering only malicious activities instead of problems that aren't due to malice.

When I was thinking about this problem I figured that something like "What percentage of voters are likely to be misrepresented in an election?" as the quantity that people would like to minimise. Whether through vote tampering or ballot stuffing effectively 'cheapening' the vote or people being disinclined to vote due to something about the system or people not being able to go to the polls or a counting error (either human or software) etc etc etc. Some of the factors you mention will affect the probability of an undesirable outcome by looking at the threat profile and what is to be gained as well as opportunity and controls in effect. This is good. However, the likely impact of each potential threat has to be considered also. For example, the imperfect human counting machines might have a high probability of error, but the actual error is low in real terms as it effects the overall will of the electorate. On the other hand, imperfect human counters are everywhere in the manual system and their error is bound to be present at all count centres, so the checks and balances to counteract those has to be examined. By contrast, computer programs are deterministic and if correctly constructed the chances of them miscounting are extremely low and practically negligible. However, if they are incorrectly constructed then they may inaccurately count the votes in many or all constituencies. Sabotage during the software build process may not be likely but the impact definitely affects all counts. The improved user-friendliness of the voting machines reduces the chance that a voter will accidentally spoil the vote. Etc etc etc, arguments to be made on both sides, these are merely for illustration.

Any thoughts on how to develop that model? I'm sure you are as keen to get to the bottom of this as I am and since we're both on staunchly opposite sides of the argument we should be able to keep each other honest in terms of developing an objective model.

Sparks

More news on US eVoting system problems cropped up on /. today, from orange county. Seems that more votes than voters were recorded. I'd have to say myself that that counted as a somewhat serious flaw, no?

Full story here.

williamgrogan

I can’t believe Sparks actually read this report and then tried to use it as evidence of a problem with the machines. It is entirely a problem with the humans. I can’t even see that the problems would not have occurred with any voting mechanism.

Some quotes from the report………

But the problems, which county officials have blamed on insufficient training for poll workers

Election officials acknowledged that poll workers provided some voters incorrect access codes that caused them to vote in the wrong legislative districts

"This is a procedures problem more than anything else. It's not a problem with a new kind of voting system," said Brady, the UC Berkeley voting systems expert

Rodermund said that despite the problems, he is satisfied with the performance of Orange County's new electronic voting system.

He said that with 22,000 combinations of local, state and federal races for each of the parties, the election was one of the most complicated in the state.

All these elections seem to have been primaries and not actual congressional elections. Presumably those holding the elections where party workers and not state officials. I did say that I expect more human related problems with more minor elections.

Sparks conclusion is ….. Seems that more votes than voters were recorded This is obviously not the case if you read the article. This was an incorrect assumption on your part. It clearly says in the article that the reason this appears to be the case is that people voted in the wrong district. This means their district had less votes and a neighbouring district more votes.

Are you just trying to wind me up?

All this report does is prove how poor journalism can be and how spin can create a false impression in the mind of the non skeptic reader. There isn’t a single example or even suspicion in this report of a machine or software failure.

Sparks

Originally posted by williamgrogan
I can’t believe Sparks actually read this report and then tried to use it as evidence of a problem with the machines.

You know, for someone who posts on the skeptics forum, you seem to have skepticism remarkably confused with belligerence. The story did not say the problem was in the machines, but it most certainly was with the machines, and in the evoting system, unless you're saying that an evoting system will never require people to monitor ir.
The machines were not user-friendly; this lead to a large number of errors by the people supposedly trained in their use; and this lead to more votes being tallied than voters; and this indicates a serious problem.
It's quite rational and logical. Qualities I was lead to believe that skeptics deign to be mandatory when questioning something...

Sparks conclusion is ….. Seems that more votes than voters were recorded This is obviously not the case if you read the article.

Actually, if you read the article, it specifically says that the error was flagged by there being more votes recorded than voters in some districts.

It also points out that there's no VVAT and because of "voter security measures", there's now no way to tell for sure how many votes were mistakenly cast and thus no way to point to the result and say "this is correct". The best estimates say 7,000 people had votes erroneously recorded.

So we know there's a serious problem, but can't tell exactly how serious, or fix it, or prevent it. Wonderful.

Are you just trying to wind me up?

Asked and answered. I note however, that you've stated what you're up to with the totally illogical assertions you've made, and you've not yet listed what qualifications allow you to say that the proposed implementation the government is pushing is fault-free and trustworthy.

There isn’t a single example or even suspicion in this report of a machine or software failure.

So a widespread systemic failure caused by poor training and poor UI design doesn't make the grade as "serious according to W.G."?
:rolleyes:

williamgrogan

Practical Problems with the Rebecca Mercuri VVAT Mechanism

Anyone who trains people to use printers, uses them themselves or in particular maintains them will know that there are many problems associated with using printers.

I will assume that the printer is a tractor/continuous feed dot matrix type as this seems the only practical solution, (and also seems to be the one used in the Mercuri drawing).

The first problem is that to successfully use printers you do need to know the little tricks and knacks that ensure they work smoothly. Officials who use these machines for a day on average every 18 months will not know the problems or solutions and will not have the dexterity or experience.

The ink cartridge will dry up after a few months of storage. I would imagine part of the instructions on using the printer would be to replace the cartridge with a new one for each election. Changing an ink cartridge is not easy, even for experienced users. It has to be put in at a particular angle and into an exact place on the printer. The ribbon needs to exactly go between two pieces of metal/plastic that are less than 1mm in width. This probably eliminates any election official with poor eyesight or who is nervous about anything mechanical. (Women would be debarred on this basis from becoming election officials. )

The tractor feed sprocket wheels need to be exactly the width of the paper sprocket holes. If they are misaligned then the paper will eventually roll off the sprockets. This will happen after between a few sheets and maybe a few dozen. The paper will also come off if it is not fed in exactly straight.

There is a very small lever that moves the print head back and forward that is needed to allow for different paper depth/number of copies. If this lever is too far back no print will appear on the paper or worse will be very faint. If it’s too far forward the paper will get jammed. This setting is quite specific. It’s a balance between print quality and the risk of jamming. I know someone who has a printer and computer system for 15 year and at their last “month end” in early March wasted twice as many pages as needed printing her statements because of printer problems.

The printer cable has pins that when incorrectly forced into the female end can get bent and even broken. An inexperienced user may not even be aware there is a correct orientation of the cable head and try and force the male connections into the female and break one or more pins. The position of this port may be in an awkward position so those with a bad back would also be eliminated as election officials. A screwdriver is also needed to screw in the two screws that hold the cable in place. A really bizarre set of clips that exist as far as I know only on the end of printer cables clips the other end to the DRE. Maybe we could use USB but then there is the danger of breaking the silly little plastic thing that holds the cable in place. (I doubt that these DRE’s have or could have USB ports.)

The paper will have to be lined up horizontally and vertically correctly or the ballot will be printed incorrectly on every ballot from then on. Top of Form will have to be set and new software installed in the DRE machines to print test pages.

Paper will run out, get jammed etc. and have to be replaced in the middle of the voting. Under pressure the inexperienced officials may make matters worse.

Continuous stationery does not happily fold or roll into a box. It also gets jammed at this point. This is often exacerbated due to dampness and static electricity. Women with mechanical skills and with good backs will not be allowed to wear nylon stockings (nor cross dressing male officials.) Paper does get damp when stored.

All the above problems are magnified in each count centre because every DRE needs its own printer and “Mercuri shed”.

If the paper printing mechanism fails this will lead to the failure and the taking out of service of the entire unit. I would estimate the failure rate of the units to be at least 10 times greater with a printer than otherwise.

The Physical Nature of the Mercuri Cabinet

I would estimate that the time and effort needed to erect the Mercuri “shed” as 10 times greater than the DRE itself. The storage space needed would also be several times greater. The expense of delivering, storing and erecting the shed would also be several times greater than the cost of the DRE machines.

The drawing of the proposed system seems to show that the voter must walk outside the curtain and around the back (or is the side) to see the printout of their vote. This cannot be as others outside could then see how you voted. This whole contraption would have to be enclosed in a curtain that goes around the front the side and the back. I estimate the curtain length of 21ft (6+3+3+3+6). The floor space now becomes 54 sq ft. (Excuse the imperial measurement but I think it’s easier to envisage.)

The cost of building the small shed that houses the printer and the glass window would be of the order of €5,000 (€20,000 if the Board of Works does it.)

Sparks

Okay, two serious questions now, for the general audience and especially whomever is acting as moderator:

1) Is there no requirement in this forum for standards of logic and supporting of arguments and standards of debate?

2) At what point did we get nearly 5200 votes in the poll? Last time I looked, we had less than a hundred....

williamgrogan

The machines were not user-friendly

Where does it say this in the article?

The errors were caused by the humans giving the voter the incorrect “ballot” to vote on. In other words a voter from District “A” was given a ballot to vote for someone in District “B”. This meant in the case of every error district “B” had 1 vote too many.

Actually, if you read the article, it specifically says that the error was flagged by there being more votes recorded than voters in some districts.

It clearly does not only say this. It does not say that overall there were more votes cast than registered voters but that because voters were given ballots to vote in the wrong district then there were too many votes in some districts and obviously too few in others.

The problem was caused by the humans and were not related, as confirmed by the “Voting Expert”, to the electronic machines.

The election was described as the “most complicated” and the errors were entirely caused by “poor training” of the officials, who were identified as “amateurs”.

We are debating whether or not VVAT is similar to anti-MMR and this article is absolutely typical of anti-MMR articles written by anti-MMR people.

There was no evidence in that report of any description that eVoting was the problem. On the contrary, it was the humans and the eVoting reduces the opportunity for human intervention and is therefore better.

Sparks

Originally posted by williamgrogan
Where does it say this in the article?

The voter would take the ticket to a second worker, who was supposed to scroll through a computer screen and use the voter's precinct and political party to select an access code that would identify the appropriate ballot. Several workers who handled this stage of the process — including some who said they didn't know more than one precinct had been assigned to their polling place — gave voters codes for the wrong precincts, causing the wrong ballots to appear on their screens.

In other words, problems with the poll workers working with the machines were a root cause of the problem. When I said there was a UI problem, I was referring to the poll workers' UI, not the voting one.

It clearly does not only say this. It does not say that overall there were more votes cast than registered voters but that because voters were given ballots to vote in the wrong district then there were too many votes in some districts and obviously too few in others.

Indeed - but there were different people running in each district, so now we know that some people got more votes than were legally available.
Again, this is a serious problem.

The problem was caused by the humans and were not related, as confirmed by the “Voting Expert”, to the electronic machines.

Except that unless you happen to have a totally human-free system, there will always be problems like this in eVoting systems. Which is a reason for wanting to have an independent record of voting data.

The election was described as the “most complicated” and the errors were entirely caused by “poor training” of the officials, who were identified as “amateurs”.

First off, of course they're amateurs, they're volunteers by law. Secondly, are you saying that you know for a fact that similar training-related problems won't happen in the June elections here? If so, how are you able to state that you know it for a fact?

We are debating whether or not VVAT is similar to anti-MMR and this article is absolutely typical of anti-MMR articles written by anti-MMR people.

Anti-MMR has been shown by professional scientists both individually and in collective professional bodies to be unfounded scientifically.
VVAT proponents, on the other hand, have been backed by professional engineers and computer scientists, both individually and in collective professional bodies.
Ergo, the two are not the same.

But then, I pointed this out a dozen pages ago or more, and you didn't accept it then, because apparently you know better, though you show no desire to tell us why...

There was no evidence in that report of any description that eVoting was the problem. On the contrary, it was the humans and the eVoting reduces the opportunity for human intervention and is therefore better.

On the contrary - the eVoting system required more training and when that wasn't up to snuff, errors were caused. Ergo, the error lies with the system, if not with a flaw in a particular piece of machinery past the UI.

However, this story in no way proves that malicious subversion or accidental error is impossible.

Finally, a single, easily answered question :
Have you seen the source code and hardware designs for the specific system proposed for use in the June elections here?
Yes or No please.

williamgrogan

In other words, problems with the poll workers working with the machines were a root cause of the problem. When I said there was a UI problem, I was referring to the poll workers' UI, not the voting one.

Red herring then! We are discussing eVoting and not other issues related to automation. This point is no more relevant than the computer system used for voter registration. BTW I cannot see how scrolling down a piece of paper is going to be any easier.

The implication in the article is that the counting or recording of the votes was wrong, it wasn’t, i.e. that the DRE’s were faulty. They weren’t and this is all that’s relevant to eVoting.

I have said that there will be human mistakes, there always are. They are just less with computers which is why they are being implemented in the first place.

There was no extra training for the eVoting. The article doesn’t say this. The training was to decide which ballot to give the voters and that training was inadequate. That had nothing to do with what happened in the polling both.

If your argument re this article is to be accepted then all you are saying is that they should not use the computers to pick which ballot to give to people which is nothing to do with Ireland and our system.

My comment re MMR was that anti-MMR people write slanted articles in newspapers like the UK Independent that put a spin on a story that makes MMR look bad. That article does the same vis a vis eVoting. That article is another example of “evidence” that is twisted to suit an agenda rather than to be used in a clinical analysis of what the correct mechanism should be.

In summary. You asked us to read this article as proof that eVoting caused problems. Nothing in this article even said this.

Have you seen the source code and hardware designs for the specific system proposed for use in the June elections here?

I do not have to answer questions you put with Yes or No.

My clients use my software and have done so for 20 years and none of them has ever seen the source. They don’t have a problem with this. The vast majority of software running in the world is a commercial secret and in this fact partly lies its success and partly its security. I have said that I do not have a fundamental problem with Open Source but our elected government did not choose it. Keep campaigning and maybe they will change their minds for the next time or maybe in another generation the politicians may be you people and may adopt it. If you want to open a thread on this do so. However as I’m banned from debating in the eVoting forum I cannot contribute.

Sparks

Originally posted by williamgrogan
Red herring then! We are discussing eVoting and not other issues related to automation.

How does that work then? You're saying we're hysterical idiots because we express concern at the outright replacement of an established system with a new untested system, and when I point out a news article documenting a serious problem with a very similar system, you claim it's a red herring because it's the human-machine interface that broke and not the machine itself?

The implication in the article is that the counting or recording of the votes was wrong

No, the implication in the article was that the system's design led to a serious error, which it did.

I have said that there will be human mistakes, there always are. They are just less with computers which is why they are being implemented in the first place.

And I say otherwise. Now prove that you're correct, if you can do so.

There was no extra training for the eVoting. The article doesn’t say this.

You're correct, it doesn't say this. It says the training for the eVoting system which was given was poor in quality. Ergo, there was training.

If your argument re this article is to be accepted then all you are saying is that they should not use the computers to pick which ballot to give to people which is nothing to do with Ireland and our system.

No, I am saying that there are more sources for errors in the system than just those that show up inside the machines, and a VVAT would allow for a fallback system to calculate the correct results in such a case.

My comment re MMR was that anti-MMR people write slanted articles in newspapers like the UK Independent that put a spin on a story that makes MMR look bad. That article does the same vis a vis eVoting.

Actually, it doesn't. And the articles I've seen which do, don't go to any great lengths to do so, because it's unnecessary - some eVoting systems just look like naked attempts to introduce a highly corruptable and rather error-prone system.

That article is another example of “evidence” that is twisted to suit an agenda rather than to be used in a clinical analysis of what the correct mechanism should be.

No, that article listed off what happened, including quotes from several people from the involved (the CEO of the company that made the equipment) to independent experts.

In summary. You asked us to read this article as proof that eVoting caused problems. Nothing in this article even said this.

On the contrary, the entire article said this. The problem may have been caused by the human element in the eVoting system, but to say that that's a trivial and "fixable" problem is to deny the realities of human capabilities.

I do not have to answer questions you put with Yes or No.

True enough, you are under no requirement to do so, but your argument is nonsense unless you do.

My clients use my software and have done so for 20 years and none of them has ever seen the source. They don’t have a problem with this.

Are your clients using your software to run their electoral system or to read their email? Don't make such patently silly comments please.

The vast majority of software running in the world is a commercial secret and in this fact partly lies its success and partly its security.

That's just plain laughable.

I have said that I do not have a fundamental problem with Open Source but our elected government did not choose it. Keep campaigning and maybe they will change their minds for the next time or maybe in another generation the politicians may be you people and may adopt it. If you want to open a thread on this do so.

Being elected to run the government does not entitle someone to change the electoral system in such a seriously flawed manner. Full stop.

However as I’m banned from debating in the eVoting forum I cannot contribute.

Why were you banned from the debate?

ecksor

Originally posted by Sparks
Why were you banned from the debate?

Because he kept posting about how the proponents of VVAT were motivated by luddism and not debating the pros and cons of the actual system that is being implemented in Ireland. In fact, he actually posted that he didn't have any interest in the e-voting issue as such, but more the psychology of the people who think differently to him about the risks to voting systems (which doesn't seem to have changed judging by the title of this thread). Furthermore he admitted that he hadn't even looked at the details of the system being discussed, backed up by the marvellous point that he didn't need to to have a discussion about it. Some would say that's the major flaw of modern democracy! "Nothing to say but determined to say it" comes to mind, as does "desparate to get into a fight with someone, anyone, oh god why won't anybody talk to me" or just "troll" but who's to know. At any rate, he was unbanned from there well over a week ago.

As for your previous questions, the voting system is a computer and therefore above reproach no?
If we want to go through the thing of challenging the result it'd be a bit unprecendented but I'm sure I could [STRIKE]alter[/STRIKE] check the audit trail if necessary.

As for moderation, that depends upon the forum. This wouldn't have lasted 5 posts on the Politics forum and the e-voting mods left it go a bit longer on their forum (although I was the one who eventually carried out the deed).

williamgrogan

Because he kept posting about how the proponents of VVAT were motivated by Luddism and not debating the pros and cons of the actual system that is being implemented in Ireland………..

Several proponents of VVAT have admitted that some of the anti-eVoting lobby are Luddites and some were hysterical. So then you banned me for an invalid reason. I am quite entitled to point out that much of the opposition to the system being implemented in Ireland is Luddism.

I most certainly was debating the pros and cons. I am almost the only one who is arguing against VVAT and I am backing that up with logical arguments, research and evidence. I posted today a detailed reason why printing might fail. No one else has bothered to do this in any media I have read even thought many of the lazy journalists are paid to do exactly this instead of trawling through “anti” web sites for their copy.

I cannot take anyone’s credentials seriously who bans someone from a debate on eVoting who claims that they are seriously worried about its effect on democracy, a central plank of which is free speech. This is total hypocrisy. There are other solutions, one can route named posters to the “delete folder”, a poster could be limited to say one post per day or a post of a certain length.

I do not think that say Homeopathy should be only tackled via the normal scientific argument. People who believe in Homeopathy have a fundamental flaw in their reasoning and it is quite legitimate to tackle this problem. The same can be said about VVAT. It is a perfectly valid position to hold. I may be wrong but then it’s up to you to prove me wrong and me to prove me right.

I do have an interest in eVoting but not a particular interest in which software is used. I have said this. I am confident to let the department decide on that. I am however totally opposed to what I see as a preposterous notion – VVAT.

I do think pro-VVAT is as silly as anti-MMR. There are at least a dozen parallels.

It is now a slightly different argument that Wakefield has been discredited but that happened since I started the thread. I know people do not like being told they are silly, the anti-MMR people don’t like it, neither do the Homeopathy or the Water Diving people. I can’t help that, they are silly.

In the Skeptics forum we must be prepared to listen to any approach because Skepticism goes beyond the normal arguments and assumptions and can examine anything, including why intelligent people believe in weird things.

The Catholic Church banned free speech too. You should use that facility very sparingly, particularly when you disagree with the person you are banning as it smells of bad sportsmanship.

ecksor

Originally posted by williamgrogan
I most certainly was debating the pros and cons. I am almost the only one who is arguing against VVAT and I am backing that up with logical arguments, research and evidence.

We'll have to agree to disagree on that one.

I cannot take anyone’s credentials seriously who bans someone from a debate on eVoting who claims that they are seriously worried about its effect on democracy, a central plank of which is free speech. This is total hypocrisy. There are other solutions, one can route named posters to the “delete folder”, a poster could be limited to say one post per day or a post of a certain length.

I suppose there's a certain sweet irony in someone debating the system while admitting ignorance of the system on a thread that was dealing with the problem of ignorance amongst the electorate about the system.

Boards isn't a democracy (somewhere between a meritocracy andn a benevolent dictatorship is where we'd fall I guess). I haven't actually made any connection between the failings of the system and its effect on democracy, merely its inability to carry out its desired function in a trustworthy way. What that desired function supports is the obvious next step, but I don't think I've shown my hand wrt political beliefs nor do I have any interest to.

I am however totally opposed to what I see as a preposterous notion – VVAT.

But you're sympathetic to VVAT that isn't paper based. So you're actually opposed to paper and not VVAT? Didn't you make some comment a few days ago on this thread that the system is audited but didn't clarify how?

Can you state for the record that you are opposed to an audit mechanism in the voting system or else propose a satisfactory audit mechanism that isn't actually a VVAT.

The Catholic Church banned free speech too. You should use that facility very sparingly, particularly when you disagree with the person you are banning as it smells of bad sportsmanship.

Maybe, but I asked other mods and admins to review the case for me and they all agreed that you deserved a ban. Unmoderated discussion forums are noisy and annoying and it is very difficult to get any sort of reasonable discussion going on them. If you tried sticking to the point and leaving the stream of consciousness noise and unfounded assumptions and wild leaps of logic out of your post you'd probably even get your points across better.

Anyway, there's a forum for this sort of discussion.

Sparks

Originally posted by williamgrogan
I am quite entitled to point out that much of the opposition to the system being implemented in Ireland is Luddism.

You would be if it were true. However, it is not - the bulk of the opposition originated with highly qualified professionals in the computer and IT sector.

I am backing that up with logical arguments, research and evidence.

Really? Care to post those in this thread?

I posted today a detailed reason why printing might fail. No one else has bothered to do this in any media I have read even thought many of the lazy journalists are paid to do exactly this instead of trawling through “anti” web sites for their copy.

We know how printers can fail. It's just that having to reenter a vote on a backup machine is less hassle than an unverifiable electoral system.
Besides which, laser printers are more reliable than the 1960s technology you described. Hell, even my office printer prints enough in a regular day to cope with the requirements of a polling booth, and it's not specially built for reliability.

I cannot take anyone’s credentials seriously who bans someone from a debate on eVoting who claims that they are seriously worried about its effect on democracy, a central plank of which is free speech. This is total hypocrisy.

Except for three points:
1) Democracy and Free Speech are seperate concepts which, though often seen together, are not interrelated;
2) Even in a society with free speech, you can't yell "Fire" in a theatre - and that's the same principle that allows a mod to tell you that you're not allowed troll;
3) You were unbanned and can currently post there and have been told that by a moderator right here.

I may be wrong but then it’s up to you to prove me wrong and me to prove me right.

Yes, however if you stick your fingers in your ears, refuse to follow basic logic, and never back up your arguments, there's no real point arguing with you is there?

I do have an interest in eVoting but not a particular interest in which software is used. I have said this. I am confident to let the department decide on that. I am however totally opposed to what I see as a preposterous notion – VVAT.

So you don't care but you do.
Wonderful.

I do think pro-VVAT is as silly as anti-MMR. There are at least a dozen parallels.

All of which hold as much water as a sieve.

In the Skeptics forum we must be prepared to listen to any approach because Skepticism goes beyond the normal arguments and assumptions and can examine anything, including why intelligent people believe in weird things.

But it cannot do so by abandoning logic. Arguing that UFO sightings are more likely due to rational explainations can't be done by saying that the Leprechauns told you that that was the case.


Now, back on topic, you have a question to answer:

Have you seen the source code and hardware designs for the system proposed for use in the June Elections?

williamgrogan

I’m hardly going to go back and post on a thread where I may end up in the situation I already did, where people replied to my points and I could not then reply to them. This makes one assume that their points successfully rebutted mine.

highly qualified professionals in the computer and IT sector

That’s arguing from authority. Furthermore the majority in Ireland opposed to eVoting are not IT professionals but the same people who are opposed to any change – Luddites. You only have to listen to them and read what they are saying to realise that.

The main proponent, and as far as I can see the inventor, of VVAT is Rebecca Mercuri. Every where in the world eVoting crops up she attacks it. She believes that computers “are fundamentally insecure” and that includes open source systems. Most of the “other IT professionals” and clearly including the I.C.S. are just aping her position without critically considering it. As a skeptic I didn’t fail to.

Laser v Dot Matrix Printers: It doesn’t matter. Lasers are no more reliable than any other printer. Anything that feeds paper through it can jam. Lasers that have a roll or drum of paper as a feed are unusual. The last one I saw cost £250,000. The Laser on your desk might print the odd CV but successfully printing thousands of ballots in one day per 18 months just is not realistic. Furthermore the hopper would normally only hold a few hundred pages. I would say that the failure rate would be in the hundreds. (The reason I suggested Dot Matrix was I thought that the printer in Mercuri’s drawing looked like one.)

I have shown clearly that VVAT is useless, prone to more errors than the solid state DRE’s and too big and costly (21ft of curtain?).

1) A Printed ballot does not prove that the recorded vote was the same
2) Checking a random sample is just another test
3) Printers are more prone to error
4) Validation via paper is pure Luddism.

Have you seen the source code and hardware designs for the system proposed for use in the June Elections

Have you seen the source code and hardware designs for the air traffic control system that was used during your last flight?

PS

Sticking your fingers in your ears to cross out input, is not very logical when reading a computer screen.

Sparks

Originally posted by williamgrogan
I’m hardly going to go back and post on a thread where I may end up in the situation I already did, where people replied to my points and I could not then reply to them. This makes one assume that their points successfully rebutted mine.

So you won't return and argue by supporting your points instead of dogmatically screaming "LUDDITE!" at anyone who disagrees with you?
Please forgive me for not being impressed!

That’s arguing from authority.

Actually, it's not.

Argument from Authority: Arguing from authority is when an individual uses the words of another to support his or her claim—perhaps pointing out that the person being cited has a Ph.D. to make him or her sound more credible.

I'm not citing an individual with a degree who is giving his personal opinion - I'm citing a professional body made up of a large number of trained professionals in the specific technical area we're discussing, who are expressing their professional opinion on a technical matter based on examining the evidence to hand. That's not arguing from authority - that's the debating equivalent of expert witness testimony.

Furthermore the majority in Ireland opposed to eVoting are not IT professionals but the same people who are opposed to any change – Luddites. You only have to listen to them and read what they are saying to realise that.

Sorry to disagree, but from all that I have seen and read both in mainstream media and discussion groups like this one, the majority are not Luddites, they are concerned people who are worried about this specific implementation. And apart from newspaper pundits who are paid by the word to write contraversial articles to sell newspapers, I haven't actually seen anyone talk about this topic whom I could point to and say "this person is a luddite".

The main proponent, and as far as I can see the inventor, of VVAT is Rebecca Mercuri. Every where in the world eVoting crops up she attacks it. She believes that computers “are fundamentally insecure” and that includes open source systems. Most of the “other IT professionals” and clearly including the I.C.S. are just aping her position without critically considering it. As a skeptic I didn’t fail to.

That's an ad hominem attack on both Dr. Mercuri and the ICS.
And since you've not proven your qualifications to question the professional opinion of either of them, I think you should either prove your case with evidence more objective than your own personal feelings on the matter, or withdraw the comment.

Laser v Dot Matrix Printers: It doesn’t matter.

Yes, it does. We're discussing a technical issue here, not a philosophical one. Details matter.

I have shown clearly that VVAT is useless, prone to more errors than the solid state DRE’s and too big and costly (21ft of curtain?).

No, you haven't. You've come up with your own design for a VVAT printer, making design choices that even I can point out flaws in just on the first read-through, and you're claiming that this exceptionally poor design's problems are sufficent to throw the whole concept of VVAT out; and yet you have no qualms about a non-VVAT system, despite the fact that we have documented cases on the record of several different non-VVAT designs failing in non-recoverable (and in some cases, non-detectable) ways.

1) A Printed ballot does not prove that the recorded vote was the same

You pick your candidates on the screen. The data is printed on paper. You check they're both the same and push "confirm".
Now, explain to me how the vote on paper can not be what you voted for?

4) Validation via paper is pure Luddism.

Mistrust of human beings with vested interests in one set of results over another and an ability to undetectably alter the results is not Luddism, it's realism. Please stop calling it anything different. A luddite would be screaming to get rid of eVoting completely. The pro-VVAT group are not arguing against eVoting and stop saying that they are, unless you have proof to the contrary that doesn't consist of selective misrepresentative quotes and your personal feelings on the matter!

Have you seen the source code and hardware designs for the air traffic control system that was used during your last flight?

That's not an answer to my question. I'll ask it again:

Have you seen the source code and hardware designs for the system proposed for use in the June Elections?

williamgrogan

When the ICS made its first statement on eVoting they were attacked by the “anti-eVoting as proposed by the government” lobby. When the made their second they were supported. You can’t have it both ways. Either you are saying that they are right based on being a body of professionals or not.

expert witness testimony

…. And we have seen how poor that has been in the UK vis a vis mothers being incorrectly jailed on the opinion of a highly qualified individual.

As far as I know I never once wrote “Luddite” in upper case.

You pick your candidates on the screen. The data is printed on paper. You check they're both the same and push "confirm". Now, explain to me how the vote on paper can not be what you voted for?

Because if I load a hacked program to steal votes I make the program print what’s on the screen and then record something else. The same can be said of a bug. Therefore the entire “window on printer/looking at screen/comparing it to paper” technique is useless!

In fact that little window and the whole rigmarole of checking the screen against the ballot is a waste of time and money. It does not tell the voter that what he voted is what was recorded.

Have you seen the source code and hardware designs for the system proposed for use in the June Elections?

I’ll answer that if you will answer this. Have you seen the hardware designs of the braking system of your car?