First, even that isn't necessarily true. You can be buying it from someone who bought it at a lower price.
You're only going to come out with more money by selling it to someone else at a higher price. You want to hold the stock until you think it's at risk, then sell it to someone else. The ideal is someone else holds it while it loses value, and eventually you buy it back at lower prices and continue, pumping money to yourself.
I'm talking about profit sharing. If the company profits then the stock tends to go up and then you can sell it and make money.
The company gives you $0.000000000000000000 of its profits when you do that.
Or the company pays dividends to shareholders.
That actually is profit sharing. Most American traders are going for value (stock revenues) and not dividends.
It doesn't matter if you give the money to the company (for initial shares) or to someone else (in a trade), either way it's an inexpensive way to potentially make some money when the company profits.
The problem is you're mostly taking other people's money, making someone poorer to make yourself richer. Even profits aren't generation of wealth; they're the microeconomics of diverting available income to a particular entity (company), and dividends are the microeconomics of diverting some of that to yourself.
It's quite possible for a company to destroy wealth while profiting, and for your stocks to pay big dividends on that. Hypothetically, if e.g. GM managed to lobby for some kind of tariff that greatly raised the price of foreign cars and also raised the cost of GM cars by a smaller amount, the same consumers would be able to purchase fewer GM cars--and fewer cars in general. People either can't access the newest cars with the best fuel efficiencies and safety features as rapidly or they buy as many cars but fewer other things.
In other words: per individual American (and in the world as a whole), less consumption occurs, there are fewer jobs, and people are poorer. Wealth has been destroyed.
This move cripples Ford and Chrysler, who were lagging in the automarket at the time. Toyota, Mazda, Kia, and Benz can't compete because their cars cost twice as much and GM's only cost 25% more. GM may have lower volume given a particular market, but its market share increases 6 times. GM is now carrying 480% as much profit. Its stock price soars, and you get tons of dividends.
GM becomes more-profitable. You become richer. The world, in total, becomes poorer. Americans especially become poorer. Unemployment soars, but you're fat and happy because you hold GM stock.
That doesn't even get into what happens when you have a State-supported monopoly and prices go up because there's no viable alternative--pulling more profits, paying more dividends.
Ballot sheets have multiple races. You break out the races and give them random IDs; you sort them by ID. Now you can't even assemble a ballot sheet, so you can't say that some unknown person filled out a piece of paper (or a form on a touch screen) that looked like something in particular.
It took me about one pass to figure that one out. First pass I considered whole ballots secure; I iterate quickly but there is still some reflection involved.
Is something that can't be fixed by any voting system.
It is impossible to perfectly monitor someone and prevent them from compromising your ability to control their vote. I've been able to solve this problem with Internet voting, but I don't want to deploy government Internet voting at this time due to other integrity concerns (integrity is reduced nearly to that of a paper ballot, and a theoretical malicious state can demonstrate integrity while freely manipulating the vote).
We can half-solve it by allowing an individual to opt-in for a vote-in-person only restriction, which drops your vote if you send an absentee ballot but cannot be verified even by the individual voter (i.e. you can tell them to remove the restriction, but not check if the restriction is set in the first place). We can almost solve this by doing the same but with a valid voter PIN (optional), such that a ballot mailed in without a PIN or with an incorrect PIN is silently ignored; this requires the voter to have access to additional ballots, however (Internet voting is infinite ballots). Either is almost equivalent to Internet voting with certain additional protections.
I'm seriously considering mobile voting rather than mail-in and Internet voting for public elections. You can put public observers and an EVM on a bus and drive around to bring in-person voting to people instead of bringing people to in-person voting. That reduces the number of ballots without strong election integrity guarantees.
I don't intend to make profits from the machines or the code; it's all consulting and non-government services (e.g. Internet voting--which is designed to meet Government election needs but targeted to private parliamentary groups, because why would I ever move to a lower security model?).
This is also why my EVM proposal requires ballot sheets to break up into races, and races to be randomly numbered, and races to be stored sorted in order of ballot ID: you can't identify the order in which the ballots were cast or reconstruct a ballot sheet from individual votes.
You know the full individual ballots for governor, mayor, delegates, senators, congress, President, etc; but you can't put together a single sheet to say some unknown person cast these particular votes in each of these races.
You also can't observe that someone was the third person to enter the booth and recover that person's specific votes from the EVM. It doesn't work that way.
This is why I've considered heavily how to centralize voting in the polling place, such as by having one EVM with 8 displays so everyone votes on one display, or by copying the votes (via removable media) all to one EVM (they're not networked; this is the price you pay for foregoing computer networks) and only producing a single batch of ballots for the polling place. I want a large pile of ballots to conflate who may have cast which vote.
I'm even considering integrity for mobile polling places (yep!) and whether or not I can maintain integrity by using the mobile polling place to collect ballots from separate polling places and then tally them together. The problem is you weaken the handling chain: I don't want to pare down the final public observers. There's no way I can fit all public observers at 8 or 12 or 15 polling places on a bus to follow all the votes around as they tally.
Americans are afraid of lots of things that never actually happen; however, voter coercion actually happens. A lot. Even our mail-in ballots carry issues with spouses and employers forcing people to vote, or to give them a signed blank to fill out for them (yes, employers have done this--when they get caught, everybody involved suddenly learns why this is an extremely bad idea).
This means that there is no way to have verification that what is in the database is what people actually voted for (refer to the Robin Williams movie "Man of the Year" for a nefarious, yet likely scenario).
You can make the probability approximate zero; you can't make it zero, ever, via any method. Even two judges counting the votes can occasionally make a duplicate mistake. That's almost-impossible; computers can do even better than that, but the marginal difference is meaningless: an election can be one or five votes away, with the likelihood of error being one-trillionth of a vote (i.e. the machine will make a bona fide recording error once every 20 years--remember they're used every other year for a few days).
You need integrity to show that the machine is, in fact, doing what the authority says it does. Your risks in an election are a malicious actor controlling the vote counting, not a faulty machine.
That's achievable; high-profile election administrators have suggested that the methods required to achieve it are "unfair to voting machine manufacturers, who have to make money". Tell those people they're stupid and tell your State legislature to pass laws overruling their broken opinion.
Maryland spent $65 million on electronic voting machines in 2002, and replaced them for the 2016 election with $28 million of optical scanning machines. Ballots, oddly enough, cost millions of dollars each election (the ballots end up costing you more than the machines). Maryland estimated reprinting Baltimore County primary election ballots (1/6 of Maryland population) at $3 million when there was an error; that means Maryland spends $30 million in 12 years on machines and over $400 million on paper ballots.
I'm working on starting my own elections consulting company, building EVMs and using a strong elections integrity model. A prototype EVM would cost $150; I expect them to cost $120 in mass production, and a braille interface would cost $1,000. That's $2 million of EVMs (10 year lifetime) and another $2 million of braille interfaces with one per polling place.
I don't intend to get into the business of selling EVMs or software; I want support from gifts (donations, but it's a private B-corp), private services (notably Internet voting for parliamentary groups), and public elections consulting (carrying the marginal costs of the business). That means I can keep slim margins on the EVMs.
As for paper ballots? Pointless. Your EVMs need to be verifiably-untampered--every last one of them--at open. You have to demonstrate a verifiably-untampered wipe and install on the EVM at poll open, and demonstrate a verifiably-untampered vote count (or something that can prove a given ballot set is identical to the one collected) at close. That gives you perfect election integrity.
If your machines are tampered, you lose that integrity--paper ballots or not. If you let the ballots go out of sight before getting counts, you lose that integrity--which means all recounts are suspect of tampering. We accept these risks for a subset of ballots (mail-in ballots) because the alternative is disenfranchisement; we do not have to accept it for on-site polling, and we can even get perfect election integrity at a mobile polling center (a van with EVMs inside).
If you're that hard-up for paper ballots, we can encode the ballot data into large QR codes and display them one page at a time on a large screen before copying them from the EVM (yes, you have to carry the votes from the EVM to the Internet-connected machine that sends them to the Board of Elections--you don't plug EVMs into networks and you don't put wireless radios in them). Take photographs. You all have cameras in your phones. Cryptographic signatures built right into the output will prevent tampering, and we spend another $0.6 million on big LCD TVs instead of $500 million on paper ballots.
Paper ballots are useless security theater; that doesn't mean I can't implement requirements. Doesn't mean I won't get creative with ways to save half a billion in taxpayer money, either.
Put the ballots on the Internet to be validated by the public as part of the voting process. You have to demonstrate integrity, not audit when questioned.
You can validate that secret ballots cast are untampered. Non-secret ballots don't actually solve that problem, as people with buyer's remorse can lie about how they voted and accuse the election authority of tampering.
It's Peter Gutmann's principles of best practices: nobody knows wtf they're talking about, but somebody says it's best practice. Paper ballots are considered secure for reasons.
We can get stronger guarantees of integrity out of EVMs than paper, if somebody actually works out an election security model as a starting point. Actually implementing it all costs money.
127 is not "21 states". 127 is how many were caught, for Kansas.
Yes. Kansas has 3 million people; the US has 300 million. 127 x 100. Think about it for a minute.
You assume that the rate is correct for Kansas, and then assume the same rate applies to states with a lot more electoral votes.
The rate would not change due to a factor of population change; other external factors would have to change the rate. Electoral votes aren't cast by registered voters, either, but by electoral delegates.
One of Trump's hand-picked candidates for Governor suggested this rate is probably correct, or close enough, and claimed this is "lots of evidence." You're claiming this isn't a lot of evidence for the same reasons I've stated: it's a tiny proportion.
Some pretty obvious ones. How does this study catch the most common way someone can double vote (a favorite of the Daley machine), which is to show up at a polling place claiming to be a dead guy? Or someone they know won't be voting for some other reason?
Near as I can tell, the rate for that is around the same as returned by the other study. Around 500 in California alone, fewer in other states, pretty much the same rate in localities that lean Republican and Democrat and among parties, and heavier by rate in population-dense areas. It's just barely tens of thousands, being charitable.
Then the problems start.
Judicial review finds large swaths of these being clerical errors (mistakes by polling personnel, such as by misreading "deceased" on the line above a particular voter's info on the roll), data matching errors (death certificate is erroneously matched to a voter registration), and what amounts to typographical errors (someone drew a stray line or filled the wrong bubble on some form, intending to notate something else). Half a percent come down to insufficient information, and you usually find zero dead voters in a state in a given year, or a single-digit count.
So you get pretty close to zero.
How does it catch the Oregon "double votes" that are as simple as pulling someone's discarded ballot out of the trash and sending it in? It doesn't.
Why are ballots in the wastebasket? Electronic voting solves this.
You missed the other side: this study will also see two different people and count them as one person, claiming a double vote. You seem kind of blind in that direction.
No, that implies someone is doing a serious study and might find a correct number, and "the gubernatorial candidate" isn't that kind of study.
"The gubernatorial candidate" was citing studies done by the Heritage Foundation and the Government Accountability Institute, both multi-million-dollar Conservative think tanks with huge biases toward inflating the amount of voter fraud and otherwise acting as mouthpieces for people like Paul Ryan. They poured millions of dollars into this research.
These are extremely-biased sources which want to support your own arguments, and they have enormous resources. It's like saying that Monsanto's research arm found minimal environmental damage and carcinogenicity for Round-Up, but they're not credible and couldn't possibly have done a "serious study" because they would have found even less if they did. They have a manifest conflict of interest to find zero; and GAI and Heritage have millions of dollars and a manifest conflict of interest to find lots of voter fraud and invent additional when they can't find any and they still can't find any.
The people in Chicago, for example, have a lot of experience in doing this kind of thing, so I'd say that you'd find 127 in just a one block area, probably. Maybe not. Certainly you'll find many times that 127 in a city with a population
How is it you can't spoof key strokes, screen shots, and bits of Web cam footage? Get a little done in the first half hour of the day, screenshot (or screen record) several dozen files being edited, and have the clock spliced into the screenshot automatically throughout the rest of the day. Auto-hide your task bar so there's never a clock. Use video recorded on different days throughout the day, based on time of day.
It's a 200-line bash script.
Coding time and KLOC are similar concepts. You won't hold freelance contracts very long if you produce at extremely high costs--such as by having normal hourly rates but charging for 6x as many hours.
Shrug. I imagine this isn't captured by every state; and in any case, I can implement EVMs to pare down the data to just Voter ID and not attach to ballots, which would capture the usual public data only (that you voted on a particular date). I quite like that approach anyway.
When it comes to EVMs, I implement various configurable options by implementing them in separate assemblies, and then removing those assemblies performing non-necessary functions from a deployment image (i.e. the programming code to do a particular thing simply does not exist in the EVM). Can't use a function that doesn't exist. In this case, the class to store certain voter information would simply store less, and the option to capture more-detailed voter identifying data wouldn't appear because reflection would not find a class exposing such a thing.
Amusing to see Donald Trump's voting history. I'm pretty sure these are by registration at the time, and not by ballot votes.
The assertion was that you get a piece of "wealth creation" if you own stock. In America, the securities market is very much about value: it's based on the stock price going up and down. European markets are more about dividends, which take a portion of profits, although that's also not really "wealth" (the difference isn't subtle, just confusing).
You don't get to "invest in companies" by buying stock on the secondary securities market. You're picking up a little pink slip that you think will be worth more or less later. You're somewhat gambling on a horse race, and you have the capacity to get more information and win money more often than not if you put work into it.
tl;dr: buying stock doesn't get you a share of wealth generation. Buying stock lets you take somebody else's money--not the company's revenues, not some kind of profits, but money out of the pocket of another securities trader who is losing. You don't produce anything in the process, so there is not additional wealth generated on the side.
The amount of money made in the stock market equals the amount of money lost. The ball gets bigger as more people buy in and inflation brings more dollars to the table.
I'm starting to move away from assuming I know what precise defect of thinking someone is exercising. I've recently noticed you can play that one out as a debate tactic to make someone argue with themselves and look like a complete tool.
The stock market makes it possible for those small investors to invest in companies, without the company getting to decide whether they get to buy stock in it.
Ah. No.
The IPO allows businesses to issue stock and get money. After that, it's secondary securities: you're not giving the business any money when you buy its stock; you're giving another person money.
The secondary security market is an information game, like poker. Most people see poker as gambling because the cards are random; they get destroyed by skilled poker players. Bluffing and reading are big parts of poker: you want to manipulate your opponent's information while deriving as much true information as you can. You want a vague read of how strong an opponent's hand is while preventing them from identifying your hand's relative strength. If you look too confident with a good hand, we know when you have a weak hand; if you bluff too hard on a weak hand, we know when you have a strong hand. A bluff has to look like a calculated risk you might take on a full house; so does a raise for an obscenely-strong hand.
I know who trades on what books. The second-level trading data thus can tell me if the big investment brokers are bailing on a stock or if it's just retail traders. I know if people are buying in on E-Trade while Goldman Sachs and Freddie Mac are dumping their shares to these fools.
Goldman and Freddie know something you don't.
I'm dumping my shares, too. I'm walking away with your money. I don't know what they know, but I sure as hell know something you don't know.
In Europe, investors focus more on stocks with dividends. That's actual profit sharing. It's also not wealth generation, just wealth distribution. Wealth generation involves structural change--increases in productivity--which are distributed through the economy, not through profits. Still, it's closer to what you're asking for than the usual American secondary securities trading game.
As to wealth generation, that's structural change, as noted. Basically, if a business starts making some good 10% cheaper and sells it for a 10% lower price, it makes the same marginal profit while consumers buy 10% more. If that's commodity good, then most consumers are buying it. Some 10% of the business's workers might get laid off--even if that's 50,000 workers, it's 0.03% of the labor force, leaving 99.97% of consumers able to buy 10% more. That's 9.997% more purchasable goods and services, and slightly less than that per person: overall, you're looking at being able to produce, purchase, and consume more than 1.099x as much per person.
I'm sure you can understand how having more stuff per person is more wealth. This is also why social insurances are important and fair: maybe we can all be 8% richer instead of 9% richer, and those people who lost their incomes because of a factory moving or a new technology can be supported as we rebuild their local economy so they can get jobs and self-support once more (which will quickly make us all even richer).
So imagine if the business just kept that money as profit.
That money isn't being spent to purchase more, so the jobs lost aren't replaceable: those people can't go out and sell something to anyone because 100% of the money people were already spending on things is already being spent on the same things. You just have fewer jobs to produce those things.
In the real world, some of that profit gets invested back into research and expansion. Some goes to dividends, where it might get banked into the financial securities market or spent to consume. There's a level of loss there. In the real world, such improvements in efficiency also lower barriers to entry and increase price competition, so prices go down and nobody really just keeps a fat profit margin; these losses are acceptable and don't harm actual people (although the broken architecture of our government's social insurances does).
If you just want to blindly invest in the stock market, you could just give me your money now. I'm actually not playing these days: it's too much work if you don't want to lose the shirt off your back.
The solution is a provisional ballot. The voter swears he is registered in location X, and the poll workers at location Y write that data down along with his ballot and it goes to the central election office to be verified.
We count those separately. You know the marginal risk. If there isn't any chance of changing the results, we don't even count them.
That has never happened in any polling place where I've voted.
Then your election procedures suck and you need to take 1,000 of your friends to your State Board of Elections to complain. Loudly.
Of course election offices keep track of who voted. They do NOT keep track of how anyone votes. You don't know "for whom you voted", unless you are a participant in vote fraud.
I paid my State Board of Elections $75 for a voter file. The voter file includes the past ten years of elections. It includes Voter ID, name, registered address, mailing address, party affiliation, election date, district, type of race, and party for whom you voted.
In a Primary election, it will say you were a Democrat, voted Democrat for Congressional 5, voted Democrat for Legislative 14 Delegate, and did not vote for Legislative 14 Senate. Those races are all on one ballot (Primary election), which means they have to actually check what you actually voted on your ballot to know that you did vote in two races but did not vote in one.
In the General, it will tell me the same thing. That means if you're in Maryland Congressional 7 and it says you voted Democrat for Representative in Congress and Republican for President, I know you voted for Elijah Cummings and Donald Trump. I know where you live, too.
How do you suppose the State is able to supply me with this information going back as far as 1992?
Yes, I targeted voters as "Base" if they were Democrat and voted Democrat in the past 4 General elections, and "Swing" if they voted Democrat in the past 2 or 3 and Republican in no more than 1 race. I even counted who voted in every election even if they voted Republican half the time. I wanted to know who was ideologically biased, who was influenceable, who was a reliable voter, and who likely abstained because they didn't like the incumbent. I targeted people who voted for Vaughn once in a while--Democrats who voted for a Republican because they just didn't like Cummings.
You believe a lot of strange things.
Your "Kansas gubernatorial candidate" does not have voting data for 21 states. Your extrapolation is flawed and likely biased, ignoring population differences. You claim you know it is "nonpartisan", except you don't.
Kansas has 3 million people. The United States has 300 million. Kobach said his State had 127 non-citizens attempting to register (with many successful) and some fewer actually voting in the past several years--using as his authority the fact that he works in the office that tracks and prosecutes all this shit, and has been involved in some high-profile ongoing research into this very thing.
127 x 100 = 12,700. 15,000 is 25% more than 12,000, but I forgot these are small numbers--15,000 is only an 18% plump. My bad.
Kobach also did claim 8,500 incidents of double-voting by legally-registered citizens in a 21 state area. Kansas is 1/100 of the population of 50 states. A 21-state area is going to be closer to 40% of the population of 50 states, although coastal states do have higher population density. You can call it 20,000 (no population adjustment) or you can call it 40,000 (assuming these states are half as populous as the remaining states on average); it's still a drop in the bucket.
Your one-state gubernatorial candidate got his hands on some questionable numbers, and that proves an "impressively-low fraud rate". If HE could find THAT MANY attempts at fraud, then the real number is probably much,
If you want to involve a computer, then at least have it print the ballot out so the voter can put it into the ballot box.
Which is then OCR scanned by a computer (that's how paper ballots are handled here) or hand-counted by election judges (which still produces errors in recounts--and an election judge can manipulate the error rate).
You can have a computer produce outputs which only reproduce from the exact same ballot set. If you can prove the EVM is untampered, you can prove the ballot set released is the same ballot set cast.
People place large amounts of faith in a paper trail that doesn't really provide auditability or verification simply because it's familiar and they can't imagine tampering with it invisibly. We're basically dealing with Peter Gutmann's Law of Best Practices: everybody says things are "best practice" because someone with no credentials whatsoever decided this was "best practice" and said so, in turn because everybody was doing it that way at the time. Encryption, password policy, and paper ballots.
It's a difficult problem because the authority is the attacker. People worry a lot about Russian hackers breaking into EVMs and changing votes, or about Diebold putting election-manipulation code into their software; they think the Board of Elections needs assurance that their voting hardware does what they want it to do. That's not right. The voters need assurance that the votes they cast are the votes the Board counts--and that the Board isn't tampering with the votes itself.
When you frame it in that sense, you realize the problem is extremely narrow: you have to prove the ballot box (EVM) is untampered when the polls open, and provide proof that the ballots taken from the ballot box are the same ballots counted. That means, again, that the EVM must provide information which can demonstrably prove a set of ballots published by the State is the same set of ballots cast on the EVM--before you put the ballot set at any risk of tampering. You have perfect, unhackable security for a ten-hour period while the EVM is in view of the public--and even that's questionable, just like the assumption that an election judge hasn't identified registered voters who don't vote and spent the day quietly slipping extra paper ballots into the ballot box while nobody's looking.
People put antivirus on EVMs and bring them into the polling center preloaded with software and data and everything. Ludicrous. How can the public verify that? Your ballot box is already tampered, and your election is compromised.
One would assume the State collects record of your individual votes. They distribute information about the party for which you voted, which means they either store a reduced amount of information or they store detailed ballot information and provide political candidates with filtered information.
Amusingly, they know if you did not vote in a particular primary--that is: you got the Democratic ballot sheet because you are a registered Democrat, and you only voted in the Congressional and Delegate race but not for any State senator. You ask the Board of Elections about a particular race and they give you a particular pile of information about said race. Congressional District 7 may say a specific voter voted Democrat (in the Democratic Primary, yes...) while Legislative District 40 State Senate may say the same voter did not vote in the same election. Both races are on the same ballot sheet.
That means they have to know what you actually put on your ballot sheet, not simply which Primary ballot sheet you received.
You'd be surprised at what's out there. Plenty of services will sell you voting history and voter research--some know everything you've purchased and considered purchasing, and will recommend things to say, issues to target, and what kind of money to ask for (politicians run on everyone else's money, and even a well-funded PAC can only donate $5,000) based on everything about your life.
The campaign services market is a very dark place. You can learn more about people than they know about themselves by trying to get their vote--and that's before you talk to them.
The tear-off serial numbers are indeed for ballot tracking. They're used to count ballots to ensure the same number are cast as are given out.
I have a voter file from my State Board of Elections that tells me the date of an election, the Voter ID of the voter, their name, their registered voting address, their mailing address, their political party affiliation at the time of the election, and for what party they voted. In a Primary, those match; in a General, however, you can be a registered Democrat and vote for a Republican--who is exactly one specific candidate because there is only one candidate from each party. I actually categorized registered Democrats by whether they were base, reliable, or swing voters.
That means, yes, if you voted for Donald Trump in 2016, the State has a record that says you, personally, voted for Donald Trump in 2016.
The party for which they voted. In the General election, that's a candidate; in the primary, well...in my district, that's 91% likely to be a particular candidate.
In one column, the voter is noted as registered Democrat/Republican/Independent. In another, the voter is noted for having voted a particular way. I have entries that say on a particular date in a primary or general election they were a registered Democrat and voted for a Republican (again: in the General, there is only one Republican).
Slashdot Mirror
First, even that isn't necessarily true. You can be buying it from someone who bought it at a lower price.
You're only going to come out with more money by selling it to someone else at a higher price. You want to hold the stock until you think it's at risk, then sell it to someone else. The ideal is someone else holds it while it loses value, and eventually you buy it back at lower prices and continue, pumping money to yourself.
I'm talking about profit sharing. If the company profits then the stock tends to go up and then you can sell it and make money.
The company gives you $0.000000000000000000 of its profits when you do that.
Or the company pays dividends to shareholders.
That actually is profit sharing. Most American traders are going for value (stock revenues) and not dividends.
It doesn't matter if you give the money to the company (for initial shares) or to someone else (in a trade), either way it's an inexpensive way to potentially make some money when the company profits.
The problem is you're mostly taking other people's money, making someone poorer to make yourself richer. Even profits aren't generation of wealth; they're the microeconomics of diverting available income to a particular entity (company), and dividends are the microeconomics of diverting some of that to yourself.
It's quite possible for a company to destroy wealth while profiting, and for your stocks to pay big dividends on that. Hypothetically, if e.g. GM managed to lobby for some kind of tariff that greatly raised the price of foreign cars and also raised the cost of GM cars by a smaller amount, the same consumers would be able to purchase fewer GM cars--and fewer cars in general. People either can't access the newest cars with the best fuel efficiencies and safety features as rapidly or they buy as many cars but fewer other things.
In other words: per individual American (and in the world as a whole), less consumption occurs, there are fewer jobs, and people are poorer. Wealth has been destroyed.
This move cripples Ford and Chrysler, who were lagging in the automarket at the time. Toyota, Mazda, Kia, and Benz can't compete because their cars cost twice as much and GM's only cost 25% more. GM may have lower volume given a particular market, but its market share increases 6 times. GM is now carrying 480% as much profit. Its stock price soars, and you get tons of dividends.
GM becomes more-profitable. You become richer. The world, in total, becomes poorer. Americans especially become poorer. Unemployment soars, but you're fat and happy because you hold GM stock.
That doesn't even get into what happens when you have a State-supported monopoly and prices go up because there's no viable alternative--pulling more profits, paying more dividends.
Ballot sheets have multiple races. You break out the races and give them random IDs; you sort them by ID. Now you can't even assemble a ballot sheet, so you can't say that some unknown person filled out a piece of paper (or a form on a touch screen) that looked like something in particular.
It took me about one pass to figure that one out. First pass I considered whole ballots secure; I iterate quickly but there is still some reflection involved.
Is something that can't be fixed by any voting system.
It is impossible to perfectly monitor someone and prevent them from compromising your ability to control their vote. I've been able to solve this problem with Internet voting, but I don't want to deploy government Internet voting at this time due to other integrity concerns (integrity is reduced nearly to that of a paper ballot, and a theoretical malicious state can demonstrate integrity while freely manipulating the vote).
We can half-solve it by allowing an individual to opt-in for a vote-in-person only restriction, which drops your vote if you send an absentee ballot but cannot be verified even by the individual voter (i.e. you can tell them to remove the restriction, but not check if the restriction is set in the first place). We can almost solve this by doing the same but with a valid voter PIN (optional), such that a ballot mailed in without a PIN or with an incorrect PIN is silently ignored; this requires the voter to have access to additional ballots, however (Internet voting is infinite ballots). Either is almost equivalent to Internet voting with certain additional protections.
I'm seriously considering mobile voting rather than mail-in and Internet voting for public elections. You can put public observers and an EVM on a bus and drive around to bring in-person voting to people instead of bringing people to in-person voting. That reduces the number of ballots without strong election integrity guarantees.
I don't intend to make profits from the machines or the code; it's all consulting and non-government services (e.g. Internet voting--which is designed to meet Government election needs but targeted to private parliamentary groups, because why would I ever move to a lower security model?).
This is also why my EVM proposal requires ballot sheets to break up into races, and races to be randomly numbered, and races to be stored sorted in order of ballot ID: you can't identify the order in which the ballots were cast or reconstruct a ballot sheet from individual votes.
You know the full individual ballots for governor, mayor, delegates, senators, congress, President, etc; but you can't put together a single sheet to say some unknown person cast these particular votes in each of these races.
You also can't observe that someone was the third person to enter the booth and recover that person's specific votes from the EVM. It doesn't work that way.
This is why I've considered heavily how to centralize voting in the polling place, such as by having one EVM with 8 displays so everyone votes on one display, or by copying the votes (via removable media) all to one EVM (they're not networked; this is the price you pay for foregoing computer networks) and only producing a single batch of ballots for the polling place. I want a large pile of ballots to conflate who may have cast which vote.
I'm even considering integrity for mobile polling places (yep!) and whether or not I can maintain integrity by using the mobile polling place to collect ballots from separate polling places and then tally them together. The problem is you weaken the handling chain: I don't want to pare down the final public observers. There's no way I can fit all public observers at 8 or 12 or 15 polling places on a bus to follow all the votes around as they tally.
Americans are afraid of lots of things that never actually happen; however, voter coercion actually happens. A lot. Even our mail-in ballots carry issues with spouses and employers forcing people to vote, or to give them a signed blank to fill out for them (yes, employers have done this--when they get caught, everybody involved suddenly learns why this is an extremely bad idea).
This means that there is no way to have verification that what is in the database is what people actually voted for (refer to the Robin Williams movie "Man of the Year" for a nefarious, yet likely scenario).
You can make the probability approximate zero; you can't make it zero, ever, via any method. Even two judges counting the votes can occasionally make a duplicate mistake. That's almost-impossible; computers can do even better than that, but the marginal difference is meaningless: an election can be one or five votes away, with the likelihood of error being one-trillionth of a vote (i.e. the machine will make a bona fide recording error once every 20 years--remember they're used every other year for a few days).
You need integrity to show that the machine is, in fact, doing what the authority says it does. Your risks in an election are a malicious actor controlling the vote counting, not a faulty machine.
That's achievable; high-profile election administrators have suggested that the methods required to achieve it are "unfair to voting machine manufacturers, who have to make money". Tell those people they're stupid and tell your State legislature to pass laws overruling their broken opinion.
Maryland spent $65 million on electronic voting machines in 2002, and replaced them for the 2016 election with $28 million of optical scanning machines. Ballots, oddly enough, cost millions of dollars each election (the ballots end up costing you more than the machines). Maryland estimated reprinting Baltimore County primary election ballots (1/6 of Maryland population) at $3 million when there was an error; that means Maryland spends $30 million in 12 years on machines and over $400 million on paper ballots.
I'm working on starting my own elections consulting company, building EVMs and using a strong elections integrity model. A prototype EVM would cost $150; I expect them to cost $120 in mass production, and a braille interface would cost $1,000. That's $2 million of EVMs (10 year lifetime) and another $2 million of braille interfaces with one per polling place.
I don't intend to get into the business of selling EVMs or software; I want support from gifts (donations, but it's a private B-corp), private services (notably Internet voting for parliamentary groups), and public elections consulting (carrying the marginal costs of the business). That means I can keep slim margins on the EVMs.
As for paper ballots? Pointless. Your EVMs need to be verifiably-untampered--every last one of them--at open. You have to demonstrate a verifiably-untampered wipe and install on the EVM at poll open, and demonstrate a verifiably-untampered vote count (or something that can prove a given ballot set is identical to the one collected) at close. That gives you perfect election integrity.
If your machines are tampered, you lose that integrity--paper ballots or not. If you let the ballots go out of sight before getting counts, you lose that integrity--which means all recounts are suspect of tampering. We accept these risks for a subset of ballots (mail-in ballots) because the alternative is disenfranchisement; we do not have to accept it for on-site polling, and we can even get perfect election integrity at a mobile polling center (a van with EVMs inside).
If you're that hard-up for paper ballots, we can encode the ballot data into large QR codes and display them one page at a time on a large screen before copying them from the EVM (yes, you have to carry the votes from the EVM to the Internet-connected machine that sends them to the Board of Elections--you don't plug EVMs into networks and you don't put wireless radios in them). Take photographs. You all have cameras in your phones. Cryptographic signatures built right into the output will prevent tampering, and we spend another $0.6 million on big LCD TVs instead of $500 million on paper ballots.
Paper ballots are useless security theater; that doesn't mean I can't implement requirements. Doesn't mean I won't get creative with ways to save half a billion in taxpayer money, either.
Put the ballots on the Internet to be validated by the public as part of the voting process. You have to demonstrate integrity, not audit when questioned.
You can validate that secret ballots cast are untampered. Non-secret ballots don't actually solve that problem, as people with buyer's remorse can lie about how they voted and accuse the election authority of tampering.
Peter Gutmann is particularly aware of this in the security industry, although IanG brought it up.
It's Peter Gutmann's principles of best practices: nobody knows wtf they're talking about, but somebody says it's best practice. Paper ballots are considered secure for reasons.
People think silly things, like that the voter keeping his own paper ballot somehow makes the system auditable, or that paper ballots recounted days or weeks later can't be tampered.
We can get stronger guarantees of integrity out of EVMs than paper, if somebody actually works out an election security model as a starting point. Actually implementing it all costs money.
127 is not "21 states". 127 is how many were caught, for Kansas.
Yes. Kansas has 3 million people; the US has 300 million. 127 x 100. Think about it for a minute.
You assume that the rate is correct for Kansas, and then assume the same rate applies to states with a lot more electoral votes.
The rate would not change due to a factor of population change; other external factors would have to change the rate. Electoral votes aren't cast by registered voters, either, but by electoral delegates.
One of Trump's hand-picked candidates for Governor suggested this rate is probably correct, or close enough, and claimed this is "lots of evidence." You're claiming this isn't a lot of evidence for the same reasons I've stated: it's a tiny proportion.
Some pretty obvious ones. How does this study catch the most common way someone can double vote (a favorite of the Daley machine), which is to show up at a polling place claiming to be a dead guy? Or someone they know won't be voting for some other reason?
Near as I can tell, the rate for that is around the same as returned by the other study. Around 500 in California alone, fewer in other states, pretty much the same rate in localities that lean Republican and Democrat and among parties, and heavier by rate in population-dense areas. It's just barely tens of thousands, being charitable.
Then the problems start.
Judicial review finds large swaths of these being clerical errors (mistakes by polling personnel, such as by misreading "deceased" on the line above a particular voter's info on the roll), data matching errors (death certificate is erroneously matched to a voter registration), and what amounts to typographical errors (someone drew a stray line or filled the wrong bubble on some form, intending to notate something else). Half a percent come down to insufficient information, and you usually find zero dead voters in a state in a given year, or a single-digit count.
So you get pretty close to zero.
How does it catch the Oregon "double votes" that are as simple as pulling someone's discarded ballot out of the trash and sending it in? It doesn't.
Why are ballots in the wastebasket? Electronic voting solves this.
You missed the other side: this study will also see two different people and count them as one person, claiming a double vote. You seem kind of blind in that direction.
No, that implies someone is doing a serious study and might find a correct number, and "the gubernatorial candidate" isn't that kind of study.
"The gubernatorial candidate" was citing studies done by the Heritage Foundation and the Government Accountability Institute, both multi-million-dollar Conservative think tanks with huge biases toward inflating the amount of voter fraud and otherwise acting as mouthpieces for people like Paul Ryan. They poured millions of dollars into this research.
These are extremely-biased sources which want to support your own arguments, and they have enormous resources. It's like saying that Monsanto's research arm found minimal environmental damage and carcinogenicity for Round-Up, but they're not credible and couldn't possibly have done a "serious study" because they would have found even less if they did. They have a manifest conflict of interest to find zero; and GAI and Heritage have millions of dollars and a manifest conflict of interest to find lots of voter fraud and invent additional when they can't find any and they still can't find any.
The people in Chicago, for example, have a lot of experience in doing this kind of thing, so I'd say that you'd find 127 in just a one block area, probably. Maybe not. Certainly you'll find many times that 127 in a city with a population
But this is in the case of when you do the work and the client _doesn't_ pay.
You get a lawyer.
How is it you can't spoof key strokes, screen shots, and bits of Web cam footage? Get a little done in the first half hour of the day, screenshot (or screen record) several dozen files being edited, and have the clock spliced into the screenshot automatically throughout the rest of the day. Auto-hide your task bar so there's never a clock. Use video recorded on different days throughout the day, based on time of day.
It's a 200-line bash script.
Coding time and KLOC are similar concepts. You won't hold freelance contracts very long if you produce at extremely high costs--such as by having normal hourly rates but charging for 6x as many hours.
Shrug. I imagine this isn't captured by every state; and in any case, I can implement EVMs to pare down the data to just Voter ID and not attach to ballots, which would capture the usual public data only (that you voted on a particular date). I quite like that approach anyway.
When it comes to EVMs, I implement various configurable options by implementing them in separate assemblies, and then removing those assemblies performing non-necessary functions from a deployment image (i.e. the programming code to do a particular thing simply does not exist in the EVM). Can't use a function that doesn't exist. In this case, the class to store certain voter information would simply store less, and the option to capture more-detailed voter identifying data wouldn't appear because reflection would not find a class exposing such a thing.
Amusing to see Donald Trump's voting history. I'm pretty sure these are by registration at the time, and not by ballot votes.
The assertion was that you get a piece of "wealth creation" if you own stock. In America, the securities market is very much about value: it's based on the stock price going up and down. European markets are more about dividends, which take a portion of profits, although that's also not really "wealth" (the difference isn't subtle, just confusing).
You don't get to "invest in companies" by buying stock on the secondary securities market. You're picking up a little pink slip that you think will be worth more or less later. You're somewhat gambling on a horse race, and you have the capacity to get more information and win money more often than not if you put work into it.
tl;dr: buying stock doesn't get you a share of wealth generation. Buying stock lets you take somebody else's money--not the company's revenues, not some kind of profits, but money out of the pocket of another securities trader who is losing. You don't produce anything in the process, so there is not additional wealth generated on the side.
The amount of money made in the stock market equals the amount of money lost. The ball gets bigger as more people buy in and inflation brings more dollars to the table.
I'm starting to move away from assuming I know what precise defect of thinking someone is exercising. I've recently noticed you can play that one out as a debate tactic to make someone argue with themselves and look like a complete tool.
The stock market makes it possible for those small investors to invest in companies, without the company getting to decide whether they get to buy stock in it.
Ah. No.
The IPO allows businesses to issue stock and get money. After that, it's secondary securities: you're not giving the business any money when you buy its stock; you're giving another person money.
The secondary security market is an information game, like poker. Most people see poker as gambling because the cards are random; they get destroyed by skilled poker players. Bluffing and reading are big parts of poker: you want to manipulate your opponent's information while deriving as much true information as you can. You want a vague read of how strong an opponent's hand is while preventing them from identifying your hand's relative strength. If you look too confident with a good hand, we know when you have a weak hand; if you bluff too hard on a weak hand, we know when you have a strong hand. A bluff has to look like a calculated risk you might take on a full house; so does a raise for an obscenely-strong hand.
I know who trades on what books. The second-level trading data thus can tell me if the big investment brokers are bailing on a stock or if it's just retail traders. I know if people are buying in on E-Trade while Goldman Sachs and Freddie Mac are dumping their shares to these fools.
Goldman and Freddie know something you don't.
I'm dumping my shares, too. I'm walking away with your money. I don't know what they know, but I sure as hell know something you don't know.
In Europe, investors focus more on stocks with dividends. That's actual profit sharing. It's also not wealth generation, just wealth distribution. Wealth generation involves structural change--increases in productivity--which are distributed through the economy, not through profits. Still, it's closer to what you're asking for than the usual American secondary securities trading game.
As to wealth generation, that's structural change, as noted. Basically, if a business starts making some good 10% cheaper and sells it for a 10% lower price, it makes the same marginal profit while consumers buy 10% more. If that's commodity good, then most consumers are buying it. Some 10% of the business's workers might get laid off--even if that's 50,000 workers, it's 0.03% of the labor force, leaving 99.97% of consumers able to buy 10% more. That's 9.997% more purchasable goods and services, and slightly less than that per person: overall, you're looking at being able to produce, purchase, and consume more than 1.099x as much per person.
I'm sure you can understand how having more stuff per person is more wealth. This is also why social insurances are important and fair: maybe we can all be 8% richer instead of 9% richer, and those people who lost their incomes because of a factory moving or a new technology can be supported as we rebuild their local economy so they can get jobs and self-support once more (which will quickly make us all even richer).
So imagine if the business just kept that money as profit.
That money isn't being spent to purchase more, so the jobs lost aren't replaceable: those people can't go out and sell something to anyone because 100% of the money people were already spending on things is already being spent on the same things. You just have fewer jobs to produce those things.
In the real world, some of that profit gets invested back into research and expansion. Some goes to dividends, where it might get banked into the financial securities market or spent to consume. There's a level of loss there. In the real world, such improvements in efficiency also lower barriers to entry and increase price competition, so prices go down and nobody really just keeps a fat profit margin; these losses are acceptable and don't harm actual people (although the broken architecture of our government's social insurances does).
If you just want to blindly invest in the stock market, you could just give me your money now. I'm actually not playing these days: it's too much work if you don't want to lose the shirt off your back.
The solution is a provisional ballot. The voter swears he is registered in location X, and the poll workers at location Y write that data down along with his ballot and it goes to the central election office to be verified.
We count those separately. You know the marginal risk. If there isn't any chance of changing the results, we don't even count them.
That has never happened in any polling place where I've voted.
Then your election procedures suck and you need to take 1,000 of your friends to your State Board of Elections to complain. Loudly.
Of course election offices keep track of who voted. They do NOT keep track of how anyone votes. You don't know "for whom you voted", unless you are a participant in vote fraud.
I paid my State Board of Elections $75 for a voter file. The voter file includes the past ten years of elections. It includes Voter ID, name, registered address, mailing address, party affiliation, election date, district, type of race, and party for whom you voted.
In a Primary election, it will say you were a Democrat, voted Democrat for Congressional 5, voted Democrat for Legislative 14 Delegate, and did not vote for Legislative 14 Senate. Those races are all on one ballot (Primary election), which means they have to actually check what you actually voted on your ballot to know that you did vote in two races but did not vote in one.
In the General, it will tell me the same thing. That means if you're in Maryland Congressional 7 and it says you voted Democrat for Representative in Congress and Republican for President, I know you voted for Elijah Cummings and Donald Trump. I know where you live, too.
How do you suppose the State is able to supply me with this information going back as far as 1992?
Yes, I targeted voters as "Base" if they were Democrat and voted Democrat in the past 4 General elections, and "Swing" if they voted Democrat in the past 2 or 3 and Republican in no more than 1 race. I even counted who voted in every election even if they voted Republican half the time. I wanted to know who was ideologically biased, who was influenceable, who was a reliable voter, and who likely abstained because they didn't like the incumbent. I targeted people who voted for Vaughn once in a while--Democrats who voted for a Republican because they just didn't like Cummings.
You believe a lot of strange things.
Your "Kansas gubernatorial candidate" does not have voting data for 21 states. Your extrapolation is flawed and likely biased, ignoring population differences. You claim you know it is "nonpartisan", except you don't.
Kansas has 3 million people. The United States has 300 million. Kobach said his State had 127 non-citizens attempting to register (with many successful) and some fewer actually voting in the past several years--using as his authority the fact that he works in the office that tracks and prosecutes all this shit, and has been involved in some high-profile ongoing research into this very thing.
127 x 100 = 12,700. 15,000 is 25% more than 12,000, but I forgot these are small numbers--15,000 is only an 18% plump. My bad.
Kobach also did claim 8,500 incidents of double-voting by legally-registered citizens in a 21 state area. Kansas is 1/100 of the population of 50 states. A 21-state area is going to be closer to 40% of the population of 50 states, although coastal states do have higher population density. You can call it 20,000 (no population adjustment) or you can call it 40,000 (assuming these states are half as populous as the remaining states on average); it's still a drop in the bucket.
Your one-state gubernatorial candidate got his hands on some questionable numbers, and that proves an "impressively-low fraud rate". If HE could find THAT MANY attempts at fraud, then the real number is probably much,
With electronic voting you can't see the ballot box.
The EVM is the ballot box.
How can you tell if the ballot box is tampered with if you can't see it?
Carefully-crafted handling procedures.
If you want to involve a computer, then at least have it print the ballot out so the voter can put it into the ballot box.
Which is then OCR scanned by a computer (that's how paper ballots are handled here) or hand-counted by election judges (which still produces errors in recounts--and an election judge can manipulate the error rate).
You can have a computer produce outputs which only reproduce from the exact same ballot set. If you can prove the EVM is untampered, you can prove the ballot set released is the same ballot set cast.
People place large amounts of faith in a paper trail that doesn't really provide auditability or verification simply because it's familiar and they can't imagine tampering with it invisibly. We're basically dealing with Peter Gutmann's Law of Best Practices: everybody says things are "best practice" because someone with no credentials whatsoever decided this was "best practice" and said so, in turn because everybody was doing it that way at the time. Encryption, password policy, and paper ballots.
It's a difficult problem because the authority is the attacker. People worry a lot about Russian hackers breaking into EVMs and changing votes, or about Diebold putting election-manipulation code into their software; they think the Board of Elections needs assurance that their voting hardware does what they want it to do. That's not right. The voters need assurance that the votes they cast are the votes the Board counts--and that the Board isn't tampering with the votes itself.
When you frame it in that sense, you realize the problem is extremely narrow: you have to prove the ballot box (EVM) is untampered when the polls open, and provide proof that the ballots taken from the ballot box are the same ballots counted. That means, again, that the EVM must provide information which can demonstrably prove a set of ballots published by the State is the same set of ballots cast on the EVM--before you put the ballot set at any risk of tampering. You have perfect, unhackable security for a ten-hour period while the EVM is in view of the public--and even that's questionable, just like the assumption that an election judge hasn't identified registered voters who don't vote and spent the day quietly slipping extra paper ballots into the ballot box while nobody's looking.
People put antivirus on EVMs and bring them into the polling center preloaded with software and data and everything. Ludicrous. How can the public verify that? Your ballot box is already tampered, and your election is compromised.
One would assume the State collects record of your individual votes. They distribute information about the party for which you voted, which means they either store a reduced amount of information or they store detailed ballot information and provide political candidates with filtered information.
Amusingly, they know if you did not vote in a particular primary--that is: you got the Democratic ballot sheet because you are a registered Democrat, and you only voted in the Congressional and Delegate race but not for any State senator. You ask the Board of Elections about a particular race and they give you a particular pile of information about said race. Congressional District 7 may say a specific voter voted Democrat (in the Democratic Primary, yes...) while Legislative District 40 State Senate may say the same voter did not vote in the same election. Both races are on the same ballot sheet.
That means they have to know what you actually put on your ballot sheet, not simply which Primary ballot sheet you received.
You'd be surprised at what's out there. Plenty of services will sell you voting history and voter research--some know everything you've purchased and considered purchasing, and will recommend things to say, issues to target, and what kind of money to ask for (politicians run on everyone else's money, and even a well-funded PAC can only donate $5,000) based on everything about your life.
The campaign services market is a very dark place. You can learn more about people than they know about themselves by trying to get their vote--and that's before you talk to them.
there's less ability for common small investors to benefit from wealth generation.
Explain.
The tear-off serial numbers are indeed for ballot tracking. They're used to count ballots to ensure the same number are cast as are given out.
I have a voter file from my State Board of Elections that tells me the date of an election, the Voter ID of the voter, their name, their registered voting address, their mailing address, their political party affiliation at the time of the election, and for what party they voted. In a Primary, those match; in a General, however, you can be a registered Democrat and vote for a Republican--who is exactly one specific candidate because there is only one candidate from each party. I actually categorized registered Democrats by whether they were base, reliable, or swing voters.
That means, yes, if you voted for Donald Trump in 2016, the State has a record that says you, personally, voted for Donald Trump in 2016.
The party for which they voted. In the General election, that's a candidate; in the primary, well...in my district, that's 91% likely to be a particular candidate.
In one column, the voter is noted as registered Democrat/Republican/Independent. In another, the voter is noted for having voted a particular way. I have entries that say on a particular date in a primary or general election they were a registered Democrat and voted for a Republican (again: in the General, there is only one Republican).