Hacker News Reader

Comment by Procrastes | Hacker News Reader

Hi Hacker_9. In Texas, where I vote, the machines don't even print paper. We just have to trust that the machine correctly recorded our vote. The reasons for the machines are, 1. cost cutting over manual counts, 2. profiteering by well connected companies, 3. And potentially making sure only the "right" votes count, although that part isn't proven.

In California we were able to vote at home with a mail-in, paper ballot and I much prefer that.

View on HN

If we have to use machines why wouldn't we just source two machines from rival vendors. One machine takes the vote from the voter through whatever method, records the vote and prints a receipt that clearly states that voters choices. Machine two reads the receipt in front of the voter and keeps a second tally of votes. The two tallies are then compared at the end of the night.

Too logical ;)

this with some crypto-sauce would be mixnet voting :)

For starters, cost. Now you have to buy two machines for every single machine you buy. It also means taking up more room. Now, instead of fitting 10 machines in a single space, you can only fit 5, creating even longer lines. This also increase the chances of failures. A single machine failing means both machines in the pair fail. So, you've gone from 10 machines to 5 pairs of machines for twice the cost. You guy from a single machine putting you down to 90% original capacity (assuming 10 machines) where as a single machine brings you down to 80% capacity (40% of original capacity).

Not to mention the additional level of confusion and potential for errors. All of this so you can buy two machines from two different vendors who might, in the end, still have ties via investors.

All good points, maybe there could be one testing machine per area or zipcode instead of one per room. People wouldn't know which voting booth would be tested. If the test pass potential false positive are still there, but you at least a quantifiable bit of error catching and restricted scoping. Cost still up for sure, but maybe people wouldn't mind paying a bit more instead of 100% for trust in the system. Would probably even motivate some to vote.

I don't think we can prevent fraud, the best we can do is source voting machines according to a spec from multiple vendors, and source software from multiple vendors as well, and do independent inspections of both by multiple 3rd parties. It would be a significant undertaking to compromise everything.

With a paper vote, you place trust in those who collect, store and, count, and you can always recount if needed. With electronic voting you place that trust in those that produce the hardware and software.

Trust it is. But the system has to try to reduce it to a minimum.

Some suggest each party should have a set of machines, so that the interest tension would cancel errors a lot.

Make the number of machines inversely proportional to the number of votes in the last election. Suddenly the next president is from the green party.

Where I voted (Massachusetts) you voted with a paper ballot, then watched a poll worker feed it into a machine which recorded it.

Same in VA but the machine in no way indicates that it read it correctly.

>> In Texas, where I vote, the machines don't even print paper. We just have to trust that the machine correctly recorded our vote.

This system just begs for being manipulated.

>> This system just begs for being manipulated.

Paper ballots are not immune from such concerns. The concept of votes being added or removed from a count isn't a new phenomena. So the standard should be whether or not the electronic means are better, rather than them being a perfect counting method.

Using paper ballots you can at least re-count the votes and validate the results of the election. With the electronic system like this you can not validate the results. Does it use an open source firmware? How can you make sure that there is no

IF vote = Clinton AND random <= small error limit not making cheating oblivious THEN vote = Trump;

line in the code?

Yep, you can recount fake paper votes, but how you will distinguish fake votes from real ones? Remember, votes are anonymous, so you cannot track vote back to live being to ask. And even if you will be able to do so, live being can lie you about his vote, to protect himself, or just to not look dumb in eyes of others.

>>> ... but how you will distinguish fake votes from real ones?

Easy. Serial numbers. Like any other anonymous system (paper money, raffle tickets etc) you assign a number to every valid ballot. Should the same number appear twice, or not appear, then you know something fishy is happening. Any extra fake ballots should be discovered, so long as the originals are not removed from the systems. Throw the numbers around randomly and creating undetectable fakes become very difficult.

The order has to be random or you can trace votes to individuals if you keep the order in which they voted.

Most systems seem to have two lines so it's not clear the order people voted in. Excluding hidden cameras which also work with voting machines.

In some voting systems (eg, the UK's), this is a feature, not a bug. Ballot serials are recorded against electoral register entries, and the courts have de-anonymized ballots in a few cases.

This is a very serious bug, even if courts want to exploit it.

Explain "fake paper votes". I'm not sure that this is a real problem with normal auditing -- record all voters that come into a precinct, what time they appear, and compare to the number of received ballots. Comparing signatures of voters to the file signature is common for absentee voting. It's just far harder and riskier to mess around with compromising paper ballots than a mistake (or "mistake") in a line of code.

You cannot force voters to come in and confirm that they sold their votes for exchange of money. They will lie. You must catch fakers, which is tricky when they have chance to win election and punish you, like Stalin did.

> Yep, you can recount fake paper votes, but how you will distinguish fake votes from real ones?

You already have a log of who showed up to vote. Compare the number of shows S with the number of ballots V which must be <= S due to poorly marked/unreadable ballots. Simple.

Edit: and this all statistically correlates with exit polls. It's very, very hard to fake all three of these in order to rig an election.

So yes, you have log, you have votes. How to distinguish real votes from fake ones? State can start election again from scratch, when it will find a problem, but people will vote very differently at this next election, because whole story will affect their minds and votes. Brave fakers can raise stakes and win.

It's not necessary for a subsequent vote to have the same outcome as the first. The issue is whether the election is secure and people have trust in the system.

So first goal of fakers is to ruin that. It even written as one of main goals in Russian trolls goal books. Find a copy of such book and read it: you will learn a lot.

Once there's no longer trust in the system, you're right, secure voting is no longer a problem.

Which is why it's important to fix this earlier rather than later.

If it's your goal to cast more FUD on the issue, that's your choice. I can't even ask to you be honest about it because that wouldn't be congruent with spreading FUD.

Just trust your leaders and FUD will not work.

Indeed, here is a recent example:

https://en.wikipedia.org/wiki/Australian_Senate_special_elec...

As is frequently attributed to Stalin:

What counts is not who votes, but who counts the votes.

"You know, comrades," says Stalin, "that I think in regard to this: I consider it completely unimportant who in the party will vote, or how; but what is extraordinarily important is this — who will count the votes, and how."

Original in Russian:

Каменев. стараясь снизойти до уровня Сталина, говорит: "А вот по вопросу, как завоевать большинство в партии". - "Знаете, товарищи, - говорит Сталин, - что я думаю по этому поводу: я считаю, что совершенно неважно, кто и как будет в партии голосовать; но вот что чрезвычайно важно, это - кто и как будет считать голоса". Даже Каменев, который уже должен знать Сталина, выразительно откашливается.

На следующий день Сталин вызывает к себе в кабинет Назаретяна и долго с ним совещается. Назаретян выходит из кабинета довольно кислый. Но он человек послушный. В тот же день постановлением Оргбюро он назначен заведующим партийным отделом "Правды" и приступает к работе.

В "Правду" поступают отчеты о собраниях партийных организаций и результаты голосований, в особенности по Москве. Работа Назаретяна очень проста. На собрании такой-то ячейки за ЦК голосовало, скажем, 300 человек, против - 600; Назаретян переправляет: за ЦК - 600, против - 300. Так это и печатается в "Правде". И так по всем организациям. Конечно, ячейка, прочтя в "Правде" ложный отчет о результатах ее голосования, протестует, звонит в "Правду", добивается отдела партийной жизни. Назаретян вежливо отвечает, обещает немедленно проверить. По проверке оказывается, "что вы совершенно правы, произошла досадная ошибка, перепутали в типографии; знаете, они очень перегружены; редакция "Правды" приносит вам свои извинения; будет напечатано исправление". Каждая ячейка полагает, что это единичная ошибка, происшедшая только с ней, и не догадывается, что это происходит по большинству ячеек. Между тем постепенно создается общая картина, что ЦК начинает выигрывать по всей линии. Провинция становится осторожнее и начинает идти за Москвой, то есть за ЦК.

(Stalin faked election by printing reverted votes in «Pravda»).

Paper ballots at least leave a paper-trail.

How do you distinguish between a real ballot and a fake ballot?

How do you know if someone decided to throw away some of them?

(While also ensuring voters remain anonymous and allowing voters to verify their votes.)

Scrutineers : https://en.wikipedia.org/wiki/Scrutineer

You have a number of differently affiliated persons watching the proceedings, and having at the end a rough idea of the number of votes that were returned per polling station. Then (with some coordination that should be trivial for the smallest of political parties) those results can be independently reckoned and compared back to the official totals. Any irregularities should be quite obvious. Recounts are probably the Achilles heel of paper ballots, as you need a way to verify that they were not tampered with in the meantime

Scrutineers also have the advantage of being understood by anyone with a pulse. "Votes go in here, mutually opposing interests watch them like a hawk until they get to the counting center, then the count is watched by those mutually opposing interests. You can be a scrutineer yourself if you're concerned. The record can also be re-tallied if there's a concern".

Compare to voting machines: "Just trust us. You need to have deep domain knowledge in several fields before you can even start to evaluate our trustworthiness (software, hardware, security, etc)... so just trust us. No, you can't examine the machines."

Yes but that is not the point. You can also add votes to electronic voting by just allowing someone to cast more votes at it. But the difference is that with paper voting, you as a single actor can only influence one voting booth, while as a hacker, you can manipulated thousands of voting machines. So the influence of a single actor could be much bigger. Furthermore, while you can manipulate a paper ballot, you can only do so so long as an investigator is not looking. If he keeps the booth in view the whole time, you cannot add anything or change votes or whatever. A computer can not be surveyed that way because I do not see what happens inside as long as its not open source code and hardware which is signed etc.

>> ...with paper voting, you as a single actor can only influence one voting booth, while as a hacker, you can manipulated thousands of voting machines.

That depends where the person is. The people managing the process, those doing the counting, have plenty of opportunity for large manipulations. There are safeguards, but the possibility remains and must be accounted for.

One fake vote can be enough if winner takes all.

Sure, but that is a problem of "winner takes all" and not any voting method.

As I know it, you have 2-3 persons that count together one urn. That's not one person doing it. And you can do recounts. And adding large numbers of ballot papers is not easy because its marked how many there are in a ballot. (What is more effective is declaring a number of ballots invalid if they are not the right candidate but still, we are speaking about in extrem cases 100 votes, not possibly millions as would be possible in hacking an electronic voting machine)

> In California we were able to vote at home with a mail-in, paper ballot and I much prefer that.

I guess you trust the mail service and the people on the receiving end to properly record your vote.

I prefer the day of (also in CA), where you get to put it into the counting machine yourself—at least then I know it was counted at my polling place.

When you do mail-in voting, your ballot has a tab you pull off with a number on it. You can check on a government website whether or not your ballot was counted using this number (though it won't show how you voted).

"At least then I know that I put it in a machine physically, which gave me psychological comfort, but no actual verification that my vote was counted."

^-- Fixed that for you.

But seriously, at some point, unless everyone sticks around to watch everyone else's votes being counted, there has to be some level of trust with the system. The only thing we can do better is to make vote counting machines' code open sourced and have the code signed with a trusted Public Key Infrastructure of some sort.

In my country some people stick around in the voting place to watch how the votes are counted. As long as somebody from your preferred political party stays, you don't have to stick around.

The paper system is very open source already. So open that even non-developers can understand it.

In the U.K. the ballot staff are volunteers and cheap/free venues are used (schools, church halls, etc.)

(I'm not defending the method you describe from Texas where there's no physical proof to audit against)

I was listening to something specifically talking about California counting mail-in ballots. They said California took longer than most states because it's big, it has liberal laws about eligibility (counting provisional ballots) and citizens are pretty sloppy about filling in mail-in ballots. They described coffee and spaghetti stains obscuring the choice. They will fill out "clean" ballots in pairs with their best guess. I'm sure that's a small number of ballots.

When I've voted in person they deliberately have me feed my ballot into a machine to confirm it was valid.

Mail-in ballots are the devil, they allow vote-buying. It happens with regularity that elections are annulled in ethnic enclaves in the UK over that. Elections don't have to be convenient, first of all they have to be safe.

In America we prefer the Clinton "walking around money" type of vote buying?

> In America we prefer the Clinton "walking around money" type of vote buying?

Is it just me or does this sentence not make sense?

It doesn't, it sounds like something that a Markov chain would say. It sounds vaguely critical of Clinton, but doesn't quite parse as English and is devoid of actual information or analysis. I fear that Markov chains aren't just a way to replicate things that people might say, but might actually model human cognitive processes in some way. :(

It kind of makes sense to me, I think the poster is trying to say Clinton buys votes but in a way we prefer/tolerate.

That way being "walking around money" in the sense of mingling with wealthy people at expensive events funded by other wealthy people in an attempt gain favor (and votes) from the people who can afford those events.

I think it's being purposely dramatic, but the English (kind of) adds up.

That's not walking around money.

Walking amongst money?

It is legal and common in the US [1]. It's typically done by the democrats because in general more people voting is good for democrats and bad for republicans.

[1] https://en.wikipedia.org/wiki/Street_money

Which elections were annulled?

There was a fairly well publicised incident in Tower Hamlets (part of London) where a local mayoral election was declared void for reasons including "Voting fraud: ballots were double-cast or cast from false addresses" [1]

[1] http://www.bbc.co.uk/news/uk-england-london-32428648

Was this guy https://en.wikipedia.org/wiki/Lutfur_Rahman_(politician)#Fal...

The one I was thinking about was probably this one from Bradford: https://www.theguardian.com/politics/2010/sep/06/men-jailed-...

The whole Luftur Rahman BS is (I think) more to do with the fact that he's a totally dodgy c&nt (as are quite possibly some of his close associates - juicy 'gossip' in Private Eye), not his ethnic or religious background. The long line of DodgyC&nts™ at TH has yet to be broken, even going back to "The-Good-Old-Days".

Haha, keeping with tradition then. The conservative choice: a Dodgy C&nt. :)

I find it curious that you don't think this also happens in middle-England Tory stronghold constituencies, or certain middle-class London Labour ones for that matter.

Reply on news.ycombinator.com