Anyone know what the theoretical speed limit of copper cable is? 10Gbs seems faster than copper can go to me.
Depends on its length, thickness, surrounding dilectric, shieling/balance/discontinuities, and the speed of the carrier/modulation. (For any given design of wire it's mainly the length.)
Copper, not being a superconductor, has resistance. The resistance combines with the stray capacatance between the conductors to form a distributed RC low-pass filter/delay line, which attenuates and delays higher frequencies more than lower frequencies - progressively more as the wire gets longer.
It gets even worse for REALLY high frequencies, because they create eddy currents in the copper that impede the penetration of current into the conductor, restricting the current to the outer part of the conductor (the "skin effect") and thus raising the effective resistance and exaggerating the frequency-selective attenuation.
This selective attenuation and delay weakens the signal - more at high frequencies than at low. As the wire gets longer the signal gets weaker and competing noise pickup gets stronger, reducing the signal-to-noise ratio and thus the amount of signal that can be carried.
But the selective attenuation and delay also distorts the waveform, creating "intersymbol interference" (stored charge from previous bits affecting the latest bit). This can be compensated for.
Current technology using SERDESes (fast serial bit streams), with some compensation for the selective attenuation (both preemphasis at the transmitter and compensation at the receiver), can get 3 Gbps through about a yard of printed circuit, or several yards of wire. More advanced devices (using tricks like four-level encoding to get two bits per modulation perios and feedback from the receiver to the transmitter by a return path) can go faster and a bit farther. (A transciever using all four pair of a Cat-5e, as of last year, could get gigabit ethernet across 30 meters.)
Frequency-domain techniques (like ADSL) can do still better. And coding schemes have been developed that get within 50% (turbo codes) or even 90%+ of the Shannon limit bit rate.
But what IS the shannon limit bit rate: It depends on a LOT of things. The biggest are:
- Length of the wire.
- Thickness of the wire.
- Quality of the dilectric around the wire.
- Interference coupled into the wire (i.e. how many other wires are in that bundle, what signals they're carrying, {for twisted pair} how tight the twists are and how they vary from conductor to conductor), how hot the wire is, etc.
You should be able to get gigabit rates to a box on your block with copper pair, with a small router there and fiber to the rest of the net. (This is "fiber to the curb".) For 10G or beyond you'll probably need CO-AX (ala cable TV) or fiber from the curb box as well - otherwise the curb boxes would need to be so close together that they get too costly - and you might as well have strung fiber from the one-per-neighborhood boxes.
(Maybe they'll push it a little farther. But I wouldn't hold my breath. Remeber that, in the US at least, you've typically got Cat-3 to the "curb" box which serves no more than 100 homes. If you're going to spring the bux dig it up and string 5e or 6 you might as well string some fiber. Later that can easily be upgraded to Tbits and beyond by transciever changes at the ends.)
So, in the spirit of trying to find flaws in order to preserve democracy for all, who is in charge of loading the open-source software onto the machines?
It doesn't MATTER if the software is corrupted. (Except that it delays that's precinct's election until it's fixed.) The software only exists to collect the voter's vote and print the filled-in ballot - which he can read - for him. If it doesn't do it correctly, he gets to try again until it DOES do it right - or fall back on another machine or a manual ballot - meanwhile raising a stink to the elction officials, press, and if necessary, courts.
The machine printed ballot that he read and checked is the official ballot. It may be the one that's machine counted. But even if the voting machine computes the tally as a convenience, the hardcopy is one that's counted if there is ANY question of the election outcome.
I worked for five years in politics, and never encountered any major official who was involved in such a thing.
Then you must not have worked in California. Or had anything to do with the US Congress' dabbling in voting rules.
Absentee ballots without excuses and perpetual absentee ballots. (Several thousand at one address, too, and I'm not talking about a nursing home or general delivery at a post office.)
Motor-voter. (A recipie for fraud, even when NOT combined with perpetual absentee ballots.)
Illegal alien voters. (And: rules against checking ID at polling places, helpful people teaching migrant workers and child-only welfare families (i.e. mommy's not a citizen) how to register and vote, "get out the vote" vans taking people from precinct to precinct - where the riders ALL go in at each precinct).
Floating ballot box tops as a hazard to navigation.
I could go on.
Yes, most of the poll workers are honest and hard-working. But it doesn't take many bad apples to spoil the barrel, since one fraudster can generate thousands of votes - and swing a close elections with millions of voters.
Once or twice, a local party official, it's true, has cheated-- and they're looked down upon and attacked, especially by the ones they 'help'.
Because they cheated? Or because they got caught, making the candiate and party look bad?
Arizona Albertson's [after being the last holdout] recently added [loyalty cards... but with] a check box at the bottom that says "I will not share info with you, but give me a card anyway"
Good for them!
When Safeway and Lucky's were rolling 'em out in my area I signed up - but using a handle rather than my truename and leaving the address info blank. (I did give them my correct approximate income, since it couldn't be easily tied to me.) This lets 'em data mine my preferences without collecting a database to haunt me personally later.
The person doing the signing up at Lucky's said that this was just fine. (Indeed, said person's own loyalty card was in the name of a deceased movie star.) I understand that cartoon characters, inventors, and other historical figures hold a rather large number of cards at these stors. B-)
And I was happy to sign up with Scolari's in Nevada (though still without tracable identity info), since they give a small cut of your purchace price to the charity of your choice from a long list - which includes Nevada gun training groups. B-)
Downsides include not being able to use checks and refraining from using credit or debit cards. But my bank has ATMs near all the places I shop so I don't need 'em. Also: If I ever lose my keyring It'll probably end up at one of three stores that have no address or contact info and an incorrect name.
The person doing the signing up said that this was just fine. (Indeed, said person's own loyalty card was in the name of a deceased movie star.)
One interesting effect was that, maybe a year or two later, they decided to do a "friendly cashier" bit. The point-of-sale terminal displays my "name" and the clerk says "thank you Mr ". Cue the theme from the Twilight Zone. B-)
Big Brother was so named because, like most dictators (popular/propaganda movies to the contrary) his carefully-cultivated public image is exceptionally friendly and helpful.
You do not get protection for the ideas in the work, which can be freely used by anyone in their own programs, so long as they don't copy your code verbatim.
That's the basic idea. There are slight exceptions in two directions:
1) Fair use includes using VERY SMALL amounts verbatim. If you've got a three-line hack that's really cute buried in there, it might be fair use to clone it - especially if it's being cloned into something other than a program to do the same job. (How much is fair use depends mainly on how much the cloning impacts the value of your work. Thus much more could be used in an article about it than in an unrelated computer program, and still less in a competing product.)
2) "Derived works" don't have to be exact copies. If somebody changes all the variable names, or translated it from C++ to Java (or English to German), it's still an infringement.
(Note that GPL stuff very definitely uses copyright as its base. Do you want to have to register every little release to have a valid GPL on it?)
YES! And SCO too!
One of the problems with GPL - and most of the other open licenses: The new law makes works "born copyrighted", even absent any labeling of WHO owns the copyright. This often makes it impossible to determine what person, or set of people collectively, have the authority to grant a license under any other terms.
It also makes it impossible to determine when the author dies, and thus when the copyright expires and the work passes into the public domain. (Given life-extension technology improvements, some of us might actually still be around then. But even if not, our great**n grandchildren shouldn't be penalized.)
Registration would require declaration of who is claiming the copyright, and what previous copyrights he's acknowledging. That gives you a starting point on figuring out who to ask (or offer some bux to).
And having a record in the Library of Congress on who the actual authors are would also make things like the SCO suit a tad easier to settle. B-)
As for registration of every release, what's the big deal? Burn a CD of the CVS repository, another of the distribution files, maybe another of a mirror of the project website, generate a hardcopy cover letter, and drop 'em in the mail, return-reciept requested. (Should be an open-source tool to generate the cover letter and filled-out versions of any forms within a month after they're defined, eh? Assuming the LoC doesn't just provide an inbound FTP site).
Absent a prohibitive fee, what's so hard about it?
- Transistors. - Information theory. - Graph Theory. (Especially as related to signal interconnectivity and switching.)
I could go on for pages. (One copy of the Bell Labs Journal collected back issues took up several shelves in the University library when I was a freshman - in 1965 - and much of that related to or enabled some aspect of comptuers.
Management, like politics, is most often short term focussed. Pure research drives what society will use 20-50 years down the track and it can be profitable for companies, as IBM is showing. It just requires long-term management.
Precicely.
Basic research pays off - in time to make a bundle for investors.
But once you've got a bunch of stuff from your OLD basic research in your portfoloio, a bunch of strip-mining Harvard types can make your balance sheet look REALLY GOOD for a few years (long enough for THEM to cash out their bubbled-up stock) by neglecting to plow any of it back into the NEXT projects. Then the guys who inherit the mess get the blame when the house of cards folds up.
As I understand it:
Bell Labs was a case in point of how basic research pays off. During the consolidation of the Bell System their size was pumped up partly as a regulatory scam: Once it had the mandated monopoly, the system could get their rates adjusted to make something like 6% profit on anything they spent on the telephone system - including research and development. So Bell Labs' real job was to spend as much money a possible on research on anything vuagely related to telephony. It looked like a no-lose: Spend a dollar, charge the telephoning public $1.06.
But it "failed" totally: From the first year of this hack onward through the disvestiture and beyond, patent licensing and the like brought in more than Bell Labs spent. Awwwwww... B-)
Xerox PARC, on the other hand, appears to have been an accounting error. (No by the same team that also mis-accounted lease revenue from their mainframe business, thought it was losing money, folded it, sued IBM for antitrust, and got laughed out of court when IBM's accounting team got hold of their books and found that it had actually been wildly profitable.)
Seems that early Xerox machines used electromechanical logic to control the device. Lots of relays and moving parts, all expensive. One of Xerox PARC's first projects was to build a control panel using one of those new-fangled microcomputer chips. This saved them a LOT of money on each machine. And PARC kept getting credited with part of this savings. So even though corporate never managed to productize most of the stuff they came up with, they looked profitable on the books. (They DID actually sell some rights to windowing systems and mice to Apple, Though. B-) )
[quoting the Washington Post reporting on some poll or other]:
Ominously, the poll showed some increased support in Muslim countries for suicide bombings and other forms of violence; 82 percent of Jordanians, 40 percent of Moroccans, 41 percent of Pakistanis and 15 percent of Turks said such violence could be justified.
Yeah, right.
And I bet if you worded it correctly you could get a similar fraction of the US population to say that there might just be SOME scenario where it would be OK to nuke Russia or China.
Sorry, AC. I don't believe a word of the news reports on polling results. Even when the polls are in English. The characterizations the reporters make of them don't usually have much to do with the actual questions - even when the newsies didn't commission a poll to push the respondents into supporting the editors' position on the subject.
THIS one was no doubt in Aarbic, not English. Before I'd trust it to say ANYTHING about how Muslims feel about blasting or burning innocent civilians to death by the thousands I'd want the original wording - along with translations and commentary on the meaning of it by at least one prominent member of each of the major madabs in each area where the poll was taken who is also a native speaker of the local dialect.
There's ONE pollster I'd trust to run such a poll, get results that have any meaning, and report them honestly: Zogby. But even there I'd only give weight to HIS own interpretation of his results, not some reporter's.
Meanwhile, my handy Sufi says I'm right and they're wrong. So there.:-b
This is the "personal computer" all over again!
on
Hack This, Please
·
· Score: 2
Just think about it: A personal computer is not a box that exists to DO something. It's a box that exists purely to be customized to do whatever the user wants.
The difference is that if you sue the Federal Government (as opposed to a State or private organization) they get to decide if you are allowed to sue them or not. That may seem unfair on the face of it, but considering the number of people that make a living from deep-pockets lawsuits it's not hard to understand (there aren't any deeper pockets than the U.S. Government.)
That's a useful effect. But it's not the reason for the rule.
The reason that's in there is so that when the fed tries to exercise its constitutional authority on some OTHER big pockets (i.e. a state), it doesn't get tied up in a bunch of legal crud.
The fed, however, DOES let such suits proceed much of the time.
And after the CIA ripped off a company's database, installed spyware, and marketed it cheap to the world's banks (wrecking the market for the original authors), and the fed refused to be sued, resulting in the database's company folding, Congress got pissed and passed some laws to prevent it happening again. (That's why you see that bit about federal agencies in ELUAs - to set up a suit they can't wiggle out of if an agency tries to pull it again.)
I don't know if this would keep the agencies from ducking the suits, but it might. (IANAL, and haven't even read the darn gobbledegook directly.) Let's see what shows up on groklaw.
First of all, the federal government would probably stomp a hole in MS's skull for anti-trust concerns [....]
Of course they DID win antitrust suits against Microsoft - TWICE. And both times they slapped Microsoft's wrist, told it to play nice, and sent it back out to the playground.
Maybe if Microsoft tells the craziest guy in his gang to punch the principal in the snoot things will be different?
(Cross your fingers. "Third time's the charm.")
Now what's the penalty for being a "three time loser" in antitrust? "Three strikes and you're out?"
We still do - it's why we were able to know that the Iranians were lying about not creating weapons-grade material.
Hmmm.
Now that you mention it, I also recall that it's how we first detected the Chernobyl incident, even before the CIA infrared satellite noticed the hot spot from the missing roof.
Step 2) Armageddon doesn't come. No end of the world. The faithful will be left with a corrected view of reality.
That would be about step 4). Step 2) is the whole Islamic population going on a Jihad. Step 3) is the destruction of most of humanity in the resulting world war.
That's close enough to armageddon for me even without any second-comings and heavenly hosts.
The atheist is not an idol worshiper which is what the unfaithful were when Mohammed received the message. Only a modern interpretation would have it that way. I doubt early Islam could have even conceived of someone who doesn't believe in anything divine at all.
Sorry, but that's wrong. There were plenty of agnositcs and athiests at the time. "Show me" predates Missouri by a bunch.
They ought to be monitoring the air right now, as they did in the cold war, for signs of radioactive material released during the CONSTRUCTION of bombs.
(Hmmm... Maybe they ARE, and saw some, which is why this is showing up now.)
This is actually done with PREVENTION in mind. Given an existing legitamite threat, this is well-spent money.
Interestingly, that was also the intent behind the doctrine of Mutual Assured Destruction (MAD) and the arms race.
Interestingly, it appears to have worked. Death rates from wars had been high and generally rising for most of recorded history. They took a nosedive right after WW II, and have been miniscule by comparison ever since.
(Yes I know a lot of people have been killed in wars since then. But you young whipper-snappers have NO idea what things like WW II were like. Or WW I, or the US Civil War, etc.)
Downside is that to make it work a President has to appear sane enough to be elected but just crazy enough that he might push the button. (The best example of that, IMHO, was probably Reagan - no doubt due to his acting experience. B-) But maybe Kennedy had him beat. He had to reestablish enough credibility to pull it off 18 months after backing out of the Bay of Pigs invasion.)
Bad idea - it's part of their armageddon scenario.
on
U.S. Prepares to Get Nuked
·
· Score: 4, Insightful
Drop a nuke on Mecca first... And make it clear their God is dead or never existed like everyone else's.
Bad idea. The destruction of Mecca by the Infidels is part of their armageddon scenario. Which continues, by the way, with the second coming of Jesus (whom they refer to as the prophet Issa, a particluarly holy man, whom they believe went bodily to heaven and will be back shortly before the end).
Playing into that scenario would essentially require the bulk of the Islamic world (most of which consider terrorism to be heresy) to go on a holy war against the bombers and their allies.
Given that (if I recall correctly) there's over a Billion of 'em last count, and they DO beileve that dying in a war to defend the faith is a ticket to paradise, this would be very very bad.
By the way, It's not "their" God. It's "our" God. Assuming you and I are both Christian and/or Jewish. (Of course that might be problematic, given your statment about the non-existence and/or death of God.)
"Allah" is just Arabic for "God" - specifically the Arabic pronounciation of the word that Hebrew pronounces "Yahweh", which became "Jehova" in English translations. It's the word that is used by Arabic-speaking Muslums, Jews, and Christians alike when referring to God.
You know, if you really believe there IS no God, or that God is dead, then you're playing into another part of the scenario. Their version of armageddon is the war between the UNfaithful and the "people of the book" - members of EVERY divinely-inspired religion, along with everybody who converts to any of 'em along the way (with Jesus back to give the last word on it all).
Drop that bomb and you're exactly what they've been waiting for.
Nip it in the bud.
You're about 1,500 years too late.
But maybe we can nip YOUR idea in the bud. Before you set off WW III in the form of the sixth Crusade.
If in fact they did only reselect 10 ballots to test, then it is a *lucky* chance. It's a small population indeed that 10 individuals can provide an accurate representation of.
Again you're misrepresenting the sampling process.
Unlike polling, the sampling is NOT to determine the probable outcome. The sampling in this case is to determine, beyond a reasonable doubt, that the machines are not miscounting the ballots.
To do this you only have to sample enough votes to detect ONE error - then follow up if you find one.
What's the acceptable error rate? Most people might throw out a number like one in a million where voting is concerned. I can't say I'm inclined to disagree.
Since you're talking millions of votes I suspect you're talking about a national election.
So what's the sample size you need to achieve, say, a 99% confidence that there is less than one error per million votes in a national election?
I'll give you a hint: It's a LOT less than 999,999 per million.
But, no, one in a million is NOT necessarily the acceptable error rate.
The acceptable error rate is any error rate that is low enough that it does not swing the election.
Thus close elections get closer scrutiny.
And since you don't know if the election is close until after the election, you need a paper trail to be able to INCREASE your level of scrutiny after a close one without busting the budget staring at every pebble in the landslides.
Oh come on - noble Nixon resigned for the good of his office while Clinton was selfish in fighting impeachment?
I certainly wouldn't call the old Ratf***er "noble". And I celebrated when he left office and again when he finally made it apparent in an interview that he had been involved in more than the coverup.
But he already HAD a track record of sacrificing his presidency to preserve the integrity of the office.
He "lost" the election vs. Kennedy by the electoral votes of Illinois. And he "lost" Illinois by less than one vote per precinct - in Democratic Mayor Richard Daley's Chicago, one of the most corrupt city-eras in US history, where the graveyard voters and other phantoms often outnumber the live ones.
Like Gore, he COULD have demanded a recount and an investigation. And in his case (UNlike Gore) he certainly would have had the votes to win. And (also unlike Gore) he also would have turned over the log and exposed the rot and thriving vermin. Instead (still unlike Gore) he chose NOT to contest it.
I have no doubt that Nixon was a very not-nice person on a number of fronts. But it's pretty clear that, like Lincoln, he put some things ahead of his personal gain and position, with a big one being his perception of the interests of the USA.
Nixon was going to be thrown out for covering up a crime. [...]
And as for Clinton - following a multiyear, $70M witch hunt, they finally caught him lying about a blowjob. Should he have resigned for that?
Absolutely. Because of the WAY he lied about it.
He lied about demanding a blowjob from an employee on the witness stand in a trial. This resulted in denying the victim her right to due process and recovery for crimes against her.
This is PERJURY. This is OBSTRUCTION OF JUSTICE. Both are FELONIES. As a lawyer he was sworn to uphold the law. (And he was later disbarred for this crime.) As a governor, and later as the president, he was the chief law enforcement officer of the land. (For this you want a multiple felon, whose felonies consist specifically of subverting the law?)
Then he repeated the lie, and others like it, before congress, while his merry men slammed the reputations of that victim, and several others. Perjury again, while in the presidential office.
His lies in office deliberately smeared the reputations of his victims. By suckering womens' rights organizations into the smear campaign he marginalized them. (You'll notice that NOBODY takes them seriously any more, now that they've taken the side of the office sexual harasser-in-chief against his victims.) And he effectively gutted the workplace sexual-harassment laws by creating the "Clinton Defense": (If it's OK for the president to do it, it must be OK for every other boss in the country.)
You bet your boots he should have resigned. And the congresscritters should have convicted him after they impeached him, when he didn't.
Much as I disliked his actions as president, I could have forgiven Clinton for "lying about a blowjob" if he'd done it solely to protect the reputation of a voluntary partner. But Clinton "lied about a blowjob" to protect himself, and to SMEAR and further harass the women on whom he'd tried to force his lusts.
So let's just drop the "lied about a blowjob" line, shall we?
It's almost as if the US govt wants having about the lowest voter turnout in the western world. Get rid of the machines & replace them with simple hand counted 'tick the box' paper ballots & I bet the turnout increases at least 10%,
or do anything else to increase the perceived accuracy
then change the vote to saturday & I bet turnout increases at least another 10%.
IMHO the real cause of the low voter turnout is apathy and resignation. When the choice is voting for the puppet on the left and the puppet on the right, there's really little point in bothering, because both puppets are being controlled by the same people.
Whether it's tweedledum/tweedledumber choices, a perception that the election is rigged, or a perception that the district is so lopsided that the outcome is predtermined, it amounts to a feeling that the vote doesn't matter - so why bother. And all three of those are bad.
But there's a fourth reason for not voting - or at least not voting on a particular candidate or issue: If the voter doesn't really care which way it comes out. THEN he shouldn't vote. (To make a random choice, or worse - a systematic choice along with other voters - corrupts the process.)
This is because the purpose of an election is NOT to count "all the people's opinions". It is to head off civil war by figuring out in advance which side would win - effectively, to hold the civil war without getting a lot of people hurt or killed.
As such, people should not be voting unless the matter is important enough to them that they would actually put some effort into DOing something about it.
The election process WAS a reasonably good model: Time off from other pursuits to register and vote - about as much for the rank-and-file electorate as going down to a recruiting office. More for the people running the election - the noncoms of the army. Still more for the party functionaries (commissioned officers) and candidates (the general and general staff). Elegibility to vote going specifically to groups that have historically proven their ability and will to organize mass violence.
Trying to get people to vote just to get a high participation level, or anything that reduces the perceived accuracy of the count, creates the risk that the losing side might think they can reverse it by force. This tends to destabilize the society.
(Also: It doesn't model assassination or asymmetric warfare {terrorism}. So those are still with us from time to time.)
On the flip side, this wouldn't jive with Diebold, because the CEO has already promised the next election to His Favorite Candidate. And, before the Rabid Right flames me for being pissed at the CEO of Diebold, remember -- I'm mad because it's a conflict of interest, not because he's a Republican. I'd be screaming murder if the Democrats, who I don't like much either, tried to pull the same crap.
I agree with you completely except for one thing: I think it makes a difference that he's a Republican.
My perception of the members of the parties is that the Republicans tend to be rule-bound and the Democrats to be "whatever works". So the Republicans are likely to react to a perception of error in the voting process by trying to fix it, while the Democrats have little incentive to do so in the districts they control (since it's still electing them, so who cares?)
But the perception that the system might be rigged by the Republicans gives a BIG incentive to the Democrats to work toward fixing it. (Or at least to make it possible to check whether the Republicans have rigged it. B-) ) Republicans, on the other hand, are likely to ALSO react by trying to fix it, both because of the rule-bound thing and because it makes them look bad and they care about that.
So with a Republican activist in control of the major e-voting machine manufacturer I expect BOTH parties to be trying to get it fixed ASAP. If it were a Democratic activist I'd expect the Democrats to let it ride.
(And maybe even the Republicans would let it ride, rather than raise a stink - much as Nixon resigned rather than further muddy the Presidential office, but Clinton fought regardless of the consequences.)
My objective is to get the elections as accurate as possible. So this makes the head of Diebold being a Republican a GOOD thing, IMHO. B-)
why couldn't the database on the back end be configured to flag any ballots that seemed irregular for inspection? for instance, if the counting machine recorded ballot #41768 as being entirely blank, this could be flagged [...]
Better yet - why is the mark read as "yes/no" rather than "yes/no/maybe" by the optics?
Back in 1966(!) I had a job that included operating an IBM optical mark reader. It did exactly that, grading each potential mark as black/grey/white. If there was a single mark on the paper that it considered grey, it could be programmed to kick the sheet out into a separate hopper for correction and reentry.
Of course for an election the stack in the separate hopper would be set aside for manual examination (to see if the "grey" marks were a light vote, an erased change, or a paper flaw) and manual tabulation.
Here we are 39 years later and the technology has gone BACKWARD on its way to incorporation into what is arguably the most important tabulation job in the country.
"Seeing statements about at least some of these errors being caught almost purely by chance is very disconcerting."
Seeing statistical sampling characterized as "found the errors purely by chance" in order to create FUD is more than disconcerting. It's appalling. And it's a prime example of "how to lie with statistics".
Yes, it's technically accurate, since chance was carefully DESIGNED INto the procedure. But the characterization uses a different meaning of "chance" to imply that the discovery of the errors was a lucky accident.
This is using a pun to tell a lie. In fact the procedure did EXACTLY WHAT IT WAS SUPPOSED TO DO - discover that there was a problem, and drive further investigation to characterize the problem and correct it, both in this and in future elections.
This event:
- Shows one reason a paper trail is needed. (It doesn't directly address deliberate software or database tampering.)
- Provides a counter-argument to claims that optically-scanned paper ballots are an acceptable substitute for machine-printed paper trail ballots. (An optically-marked ballot may look just fine but scan incorrectly.)
Touch-screen machines that print a voter-readable paper trail currently appear to be the most reliable solution for error-resistant and cheat-resistant elections.
His response was [...] a system that "anonymizes" data by an encryption technique called hashing. Because the data are scrambled, [...] secret watch lists can be distributed to private entities, all without fear--because they can't be read
Although this is a step in the right direction [...]
Actually it does NOTHING to prevent misuse of the lists as blacklists - or even vigilante hit lists.
As a blacklist: Put in the name/identifying information, turn the crank, out comes a yes/no. Dump on the subject if it's yes. You've just extracted the entries for everybody you interact with who's on the list, just in time to dump on 'em.
As a hit list: Put in the names/info of everybody in town (or everybody who appears in ANY public database). Turn the crank. Out comes a list list yes/nos. Discard the names that generated "no"s and you have your hit list. If you use an extensive enough database of people you end up extracting ALL from the "encrypted" list (plus any false-positives generated by the hash function).
So the hashing function does NOTHING to secure the list (except maybe pollute it with extra false-positives.)
Slashdot Mirror
Anyone know what the theoretical speed limit of copper cable is? 10Gbs seems faster than copper can go to me.
Depends on its length, thickness, surrounding dilectric, shieling/balance/discontinuities, and the speed of the carrier/modulation. (For any given design of wire it's mainly the length.)
Copper, not being a superconductor, has resistance. The resistance combines with the stray capacatance between the conductors to form a distributed RC low-pass filter/delay line, which attenuates and delays higher frequencies more than lower frequencies - progressively more as the wire gets longer.
It gets even worse for REALLY high frequencies, because they create eddy currents in the copper that impede the penetration of current into the conductor, restricting the current to the outer part of the conductor (the "skin effect") and thus raising the effective resistance and exaggerating the frequency-selective attenuation.
This selective attenuation and delay weakens the signal - more at high frequencies than at low. As the wire gets longer the signal gets weaker and competing noise pickup gets stronger, reducing the signal-to-noise ratio and thus the amount of signal that can be carried.
But the selective attenuation and delay also distorts the waveform, creating "intersymbol interference" (stored charge from previous bits affecting the latest bit). This can be compensated for.
Current technology using SERDESes (fast serial bit streams), with some compensation for the selective attenuation (both preemphasis at the transmitter and compensation at the receiver), can get 3 Gbps through about a yard of printed circuit, or several yards of wire. More advanced devices (using tricks like four-level encoding to get two bits per modulation perios and feedback from the receiver to the transmitter by a return path) can go faster and a bit farther. (A transciever using all four pair of a Cat-5e, as of last year, could get gigabit ethernet across 30 meters.)
Frequency-domain techniques (like ADSL) can do still better. And coding schemes have been developed that get within 50% (turbo codes) or even 90%+ of the Shannon limit bit rate.
But what IS the shannon limit bit rate: It depends on a LOT of things. The biggest are:
- Length of the wire.
- Thickness of the wire.
- Quality of the dilectric around the wire.
- Interference coupled into the wire (i.e. how many other wires are in that bundle, what signals they're carrying, {for twisted pair} how tight the twists are and how they vary from conductor to conductor), how hot the wire is, etc.
You should be able to get gigabit rates to a box on your block with copper pair, with a small router there and fiber to the rest of the net. (This is "fiber to the curb".) For 10G or beyond you'll probably need CO-AX (ala cable TV) or fiber from the curb box as well - otherwise the curb boxes would need to be so close together that they get too costly - and you might as well have strung fiber from the one-per-neighborhood boxes.
(Maybe they'll push it a little farther. But I wouldn't hold my breath. Remeber that, in the US at least, you've typically got Cat-3 to the "curb" box which serves no more than 100 homes. If you're going to spring the bux dig it up and string 5e or 6 you might as well string some fiber. Later that can easily be upgraded to Tbits and beyond by transciever changes at the ends.)
So, in the spirit of trying to find flaws in order to preserve democracy for all, who is in charge of loading the open-source software onto the machines?
It doesn't MATTER if the software is corrupted. (Except that it delays that's precinct's election until it's fixed.) The software only exists to collect the voter's vote and print the filled-in ballot - which he can read - for him. If it doesn't do it correctly, he gets to try again until it DOES do it right - or fall back on another machine or a manual ballot - meanwhile raising a stink to the elction officials, press, and if necessary, courts.
The machine printed ballot that he read and checked is the official ballot. It may be the one that's machine counted. But even if the voting machine computes the tally as a convenience, the hardcopy is one that's counted if there is ANY question of the election outcome.
I worked for five years in politics, and never encountered any major official who was involved in such a thing.
Then you must not have worked in California. Or had anything to do with the US Congress' dabbling in voting rules.
Absentee ballots without excuses and perpetual absentee ballots. (Several thousand at one address, too, and I'm not talking about a nursing home or general delivery at a post office.)
Motor-voter. (A recipie for fraud, even when NOT combined with perpetual absentee ballots.)
Illegal alien voters. (And: rules against checking ID at polling places, helpful people teaching migrant workers and child-only welfare families (i.e. mommy's not a citizen) how to register and vote, "get out the vote" vans taking people from precinct to precinct - where the riders ALL go in at each precinct).
Floating ballot box tops as a hazard to navigation.
I could go on.
Yes, most of the poll workers are honest and hard-working. But it doesn't take many bad apples to spoil the barrel, since one fraudster can generate thousands of votes - and swing a close elections with millions of voters.
Once or twice, a local party official, it's true, has cheated-- and they're looked down upon and attacked, especially by the ones they 'help'.
Because they cheated? Or because they got caught, making the candiate and party look bad?
Arizona Albertson's [after being the last holdout] recently added [loyalty cards ... but with] a check box at the bottom that says "I will not share info with you, but give me a card anyway"
Good for them!
When Safeway and Lucky's were rolling 'em out in my area I signed up - but using a handle rather than my truename and leaving the address info blank. (I did give them my correct approximate income, since it couldn't be easily tied to me.) This lets 'em data mine my preferences without collecting a database to haunt me personally later.
The person doing the signing up at Lucky's said that this was just fine. (Indeed, said person's own loyalty card was in the name of a deceased movie star.) I understand that cartoon characters, inventors, and other historical figures hold a rather large number of cards at these stors. B-)
And I was happy to sign up with Scolari's in Nevada (though still without tracable identity info), since they give a small cut of your purchace price to the charity of your choice from a long list - which includes Nevada gun training groups. B-)
Downsides include not being able to use checks and refraining from using credit or debit cards. But my bank has ATMs near all the places I shop so I don't need 'em. Also: If I ever lose my keyring It'll probably end up at one of three stores that have no address or contact info and an incorrect name.
The person doing the signing up said that this was just fine. (Indeed, said person's own loyalty card was in the name of a deceased movie star.)
One interesting effect was that, maybe a year or two later, they decided to do a "friendly cashier" bit. The point-of-sale terminal displays my "name" and the clerk says "thank you Mr ". Cue the theme from the Twilight Zone. B-)
Big Brother was so named because, like most dictators (popular/propaganda movies to the contrary) his carefully-cultivated public image is exceptionally friendly and helpful.
You do not get protection for the ideas in the work, which can be freely used by anyone in their own programs, so long as they don't copy your code verbatim.
That's the basic idea. There are slight exceptions in two directions:
1) Fair use includes using VERY SMALL amounts verbatim. If you've got a three-line hack that's really cute buried in there, it might be fair use to clone it - especially if it's being cloned into something other than a program to do the same job. (How much is fair use depends mainly on how much the cloning impacts the value of your work. Thus much more could be used in an article about it than in an unrelated computer program, and still less in a competing product.)
2) "Derived works" don't have to be exact copies. If somebody changes all the variable names, or translated it from C++ to Java (or English to German), it's still an infringement.
(Note that GPL stuff very definitely uses copyright as its base. Do you want to have to register every little release to have a valid GPL on it?)
YES! And SCO too!
One of the problems with GPL - and most of the other open licenses: The new law makes works "born copyrighted", even absent any labeling of WHO owns the copyright. This often makes it impossible to determine what person, or set of people collectively, have the authority to grant a license under any other terms.
It also makes it impossible to determine when the author dies, and thus when the copyright expires and the work passes into the public domain. (Given life-extension technology improvements, some of us might actually still be around then. But even if not, our great**n grandchildren shouldn't be penalized.)
Registration would require declaration of who is claiming the copyright, and what previous copyrights he's acknowledging. That gives you a starting point on figuring out who to ask (or offer some bux to).
And having a record in the Library of Congress on who the actual authors are would also make things like the SCO suit a tad easier to settle. B-)
As for registration of every release, what's the big deal? Burn a CD of the CVS repository, another of the distribution files, maybe another of a mirror of the project website, generate a hardcopy cover letter, and drop 'em in the mail, return-reciept requested. (Should be an open-source tool to generate the cover letter and filled-out versions of any forms within a month after they're defined, eh? Assuming the LoC doesn't just provide an inbound FTP site).
Absent a prohibitive fee, what's so hard about it?
We chose the MTSEC security from Bizland Consultants inc instead, and it saved us over 700%
I see.
Not only did your maintainence budget go to zero but Bizland Consultants paid YOU six times your former budget.
Where do I sign up for THAT deal? B-)
= = = =
On second thought, forget it. TANSTAFFL, so they must be getting something from you that's worth even more.
- Transistors.
- Information theory.
- Graph Theory. (Especially as related to signal interconnectivity and switching.)
I could go on for pages. (One copy of the Bell Labs Journal collected back issues took up several shelves in the University library when I was a freshman - in 1965 - and much of that related to or enabled some aspect of comptuers.
Management, like politics, is most often short term focussed. Pure research drives what society will use 20-50 years down the track and it can be profitable for companies, as IBM is showing. It just requires long-term management.
Precicely.
Basic research pays off - in time to make a bundle for investors.
But once you've got a bunch of stuff from your OLD basic research in your portfoloio, a bunch of strip-mining Harvard types can make your balance sheet look REALLY GOOD for a few years (long enough for THEM to cash out their bubbled-up stock) by neglecting to plow any of it back into the NEXT projects. Then the guys who inherit the mess get the blame when the house of cards folds up.
As I understand it:
Bell Labs was a case in point of how basic research pays off. During the consolidation of the Bell System their size was pumped up partly as a regulatory scam: Once it had the mandated monopoly, the system could get their rates adjusted to make something like 6% profit on anything they spent on the telephone system - including research and development. So Bell Labs' real job was to spend as much money a possible on research on anything vuagely related to telephony. It looked like a no-lose: Spend a dollar, charge the telephoning public $1.06.
But it "failed" totally: From the first year of this hack onward through the disvestiture and beyond, patent licensing and the like brought in more than Bell Labs spent. Awwwwww... B-)
Xerox PARC, on the other hand, appears to have been an accounting error. (No by the same team that also mis-accounted lease revenue from their mainframe business, thought it was losing money, folded it, sued IBM for antitrust, and got laughed out of court when IBM's accounting team got hold of their books and found that it had actually been wildly profitable.)
Seems that early Xerox machines used electromechanical logic to control the device. Lots of relays and moving parts, all expensive. One of Xerox PARC's first projects was to build a control panel using one of those new-fangled microcomputer chips. This saved them a LOT of money on each machine. And PARC kept getting credited with part of this savings. So even though corporate never managed to productize most of the stuff they came up with, they looked profitable on the books. (They DID actually sell some rights to windowing systems and mice to Apple, Though. B-) )
[quoting the Washington Post reporting on some poll or other]:
:-b
Ominously, the poll showed some increased support in Muslim countries for suicide bombings and other forms of violence; 82 percent of Jordanians, 40 percent of Moroccans, 41 percent of Pakistanis and 15 percent of Turks said such violence could be justified.
Yeah, right.
And I bet if you worded it correctly you could get a similar fraction of the US population to say that there might just be SOME scenario where it would be OK to nuke Russia or China.
Sorry, AC. I don't believe a word of the news reports on polling results. Even when the polls are in English. The characterizations the reporters make of them don't usually have much to do with the actual questions - even when the newsies didn't commission a poll to push the respondents into supporting the editors' position on the subject.
THIS one was no doubt in Aarbic, not English. Before I'd trust it to say ANYTHING about how Muslims feel about blasting or burning innocent civilians to death by the thousands I'd want the original wording - along with translations and commentary on the meaning of it by at least one prominent member of each of the major madabs in each area where the poll was taken who is also a native speaker of the local dialect.
There's ONE pollster I'd trust to run such a poll, get results that have any meaning, and report them honestly: Zogby. But even there I'd only give weight to HIS own interpretation of his results, not some reporter's.
Meanwhile, my handy Sufi says I'm right and they're wrong. So there.
Just think about it: A personal computer is not a box that exists to DO something. It's a box that exists purely to be customized to do whatever the user wants.
Look how THAT caught on.
The difference is that if you sue the Federal Government (as opposed to a State or private organization) they get to decide if you are allowed to sue them or not. That may seem unfair on the face of it, but considering the number of people that make a living from deep-pockets lawsuits it's not hard to understand (there aren't any deeper pockets than the U.S. Government.)
That's a useful effect. But it's not the reason for the rule.
The reason that's in there is so that when the fed tries to exercise its constitutional authority on some OTHER big pockets (i.e. a state), it doesn't get tied up in a bunch of legal crud.
The fed, however, DOES let such suits proceed much of the time.
And after the CIA ripped off a company's database, installed spyware, and marketed it cheap to the world's banks (wrecking the market for the original authors), and the fed refused to be sued, resulting in the database's company folding, Congress got pissed and passed some laws to prevent it happening again. (That's why you see that bit about federal agencies in ELUAs - to set up a suit they can't wiggle out of if an agency tries to pull it again.)
I don't know if this would keep the agencies from ducking the suits, but it might. (IANAL, and haven't even read the darn gobbledegook directly.) Let's see what shows up on groklaw.
First of all, the federal government would probably stomp a hole in MS's skull for anti-trust concerns [....]
Of course they DID win antitrust suits against Microsoft - TWICE. And both times they slapped Microsoft's wrist, told it to play nice, and sent it back out to the playground.
Maybe if Microsoft tells the craziest guy in his gang to punch the principal in the snoot things will be different?
(Cross your fingers. "Third time's the charm.")
Now what's the penalty for being a "three time loser" in antitrust? "Three strikes and you're out?"
We still do - it's why we were able to know that the Iranians were lying about not creating weapons-grade material.
Hmmm.
Now that you mention it, I also recall that it's how we first detected the Chernobyl incident, even before the CIA infrared satellite noticed the hot spot from the missing roof.
Step 2) Armageddon doesn't come. No end of the world. The faithful will be left with a corrected view of reality.
That would be about step 4). Step 2) is the whole Islamic population going on a Jihad. Step 3) is the destruction of most of humanity in the resulting world war.
That's close enough to armageddon for me even without any second-comings and heavenly hosts.
The atheist is not an idol worshiper which is what the unfaithful were when Mohammed received the message. Only a modern interpretation would have it that way. I doubt early Islam could have even conceived of someone who doesn't believe in anything divine at all.
Sorry, but that's wrong. There were plenty of agnositcs and athiests at the time. "Show me" predates Missouri by a bunch.
They ought to be monitoring the air right now, as they did in the cold war, for signs of radioactive material released during the CONSTRUCTION of bombs.
(Hmmm... Maybe they ARE, and saw some, which is why this is showing up now.)
This is actually done with PREVENTION in mind. Given an existing legitamite threat, this is well-spent money.
Interestingly, that was also the intent behind the doctrine of Mutual Assured Destruction (MAD) and the arms race.
Interestingly, it appears to have worked. Death rates from wars had been high and generally rising for most of recorded history. They took a nosedive right after WW II, and have been miniscule by comparison ever since.
(Yes I know a lot of people have been killed in wars since then. But you young whipper-snappers have NO idea what things like WW II were like. Or WW I, or the US Civil War, etc.)
Downside is that to make it work a President has to appear sane enough to be elected but just crazy enough that he might push the button. (The best example of that, IMHO, was probably Reagan - no doubt due to his acting experience. B-) But maybe Kennedy had him beat. He had to reestablish enough credibility to pull it off 18 months after backing out of the Bay of Pigs invasion.)
Drop a nuke on Mecca first ... And make it clear their God is dead or never existed like everyone else's.
Bad idea. The destruction of Mecca by the Infidels is part of their armageddon scenario. Which continues, by the way, with the second coming of Jesus (whom they refer to as the prophet Issa, a particluarly holy man, whom they believe went bodily to heaven and will be back shortly before the end).
Playing into that scenario would essentially require the bulk of the Islamic world (most of which consider terrorism to be heresy) to go on a holy war against the bombers and their allies.
Given that (if I recall correctly) there's over a Billion of 'em last count, and they DO beileve that dying in a war to defend the faith is a ticket to paradise, this would be very very bad.
By the way, It's not "their" God. It's "our" God. Assuming you and I are both Christian and/or Jewish. (Of course that might be problematic, given your statment about the non-existence and/or death of God.)
"Allah" is just Arabic for "God" - specifically the Arabic pronounciation of the word that Hebrew pronounces "Yahweh", which became "Jehova" in English translations. It's the word that is used by Arabic-speaking Muslums, Jews, and Christians alike when referring to God.
You know, if you really believe there IS no God, or that God is dead, then you're playing into another part of the scenario. Their version of armageddon is the war between the UNfaithful and the "people of the book" - members of EVERY divinely-inspired religion, along with everybody who converts to any of 'em along the way (with Jesus back to give the last word on it all).
Drop that bomb and you're exactly what they've been waiting for.
Nip it in the bud.
You're about 1,500 years too late.
But maybe we can nip YOUR idea in the bud. Before you set off WW III in the form of the sixth Crusade.
If in fact they did only reselect 10 ballots to test, then it is a *lucky* chance. It's a small population indeed that 10 individuals can provide an accurate representation of.
Again you're misrepresenting the sampling process.
Unlike polling, the sampling is NOT to determine the probable outcome. The sampling in this case is to determine, beyond a reasonable doubt, that the machines are not miscounting the ballots.
To do this you only have to sample enough votes to detect ONE error - then follow up if you find one.
What's the acceptable error rate? Most people might throw out a number like one in a million where voting is concerned. I can't say I'm inclined to disagree.
Since you're talking millions of votes I suspect you're talking about a national election.
So what's the sample size you need to achieve, say, a 99% confidence that there is less than one error per million votes in a national election?
I'll give you a hint: It's a LOT less than 999,999 per million.
But, no, one in a million is NOT necessarily the acceptable error rate.
The acceptable error rate is any error rate that is low enough that it does not swing the election.
Thus close elections get closer scrutiny.
And since you don't know if the election is close until after the election, you need a paper trail to be able to INCREASE your level of scrutiny after a close one without busting the budget staring at every pebble in the landslides.
Oh come on - noble Nixon resigned for the good of his office while Clinton was selfish in fighting impeachment?
I certainly wouldn't call the old Ratf***er "noble". And I celebrated when he left office and again when he finally made it apparent in an interview that he had been involved in more than the coverup.
But he already HAD a track record of sacrificing his presidency to preserve the integrity of the office.
He "lost" the election vs. Kennedy by the electoral votes of Illinois. And he "lost" Illinois by less than one vote per precinct - in Democratic Mayor Richard Daley's Chicago, one of the most corrupt city-eras in US history, where the graveyard voters and other phantoms often outnumber the live ones.
Like Gore, he COULD have demanded a recount and an investigation. And in his case (UNlike Gore) he certainly would have had the votes to win. And (also unlike Gore) he also would have turned over the log and exposed the rot and thriving vermin. Instead (still unlike Gore) he chose NOT to contest it.
I have no doubt that Nixon was a very not-nice person on a number of fronts. But it's pretty clear that, like Lincoln, he put some things ahead of his personal gain and position, with a big one being his perception of the interests of the USA.
Nixon was going to be thrown out for covering up a crime. [...]
And as for Clinton - following a multiyear, $70M witch hunt, they finally caught him lying about a blowjob. Should he have resigned for that?
Absolutely. Because of the WAY he lied about it.
He lied about demanding a blowjob from an employee on the witness stand in a trial. This resulted in denying the victim her right to due process and recovery for crimes against her.
This is PERJURY. This is OBSTRUCTION OF JUSTICE. Both are FELONIES. As a lawyer he was sworn to uphold the law. (And he was later disbarred for this crime.) As a governor, and later as the president, he was the chief law enforcement officer of the land. (For this you want a multiple felon, whose felonies consist specifically of subverting the law?)
Then he repeated the lie, and others like it, before congress, while his merry men slammed the reputations of that victim, and several others. Perjury again, while in the presidential office.
His lies in office deliberately smeared the reputations of his victims. By suckering womens' rights organizations into the smear campaign he marginalized them. (You'll notice that NOBODY takes them seriously any more, now that they've taken the side of the office sexual harasser-in-chief against his victims.) And he effectively gutted the workplace sexual-harassment laws by creating the "Clinton Defense": (If it's OK for the president to do it, it must be OK for every other boss in the country.)
You bet your boots he should have resigned. And the congresscritters should have convicted him after they impeached him, when he didn't.
Much as I disliked his actions as president, I could have forgiven Clinton for "lying about a blowjob" if he'd done it solely to protect the reputation of a voluntary partner. But Clinton "lied about a blowjob" to protect himself, and to SMEAR and further harass the women on whom he'd tried to force his lusts.
So let's just drop the "lied about a blowjob" line, shall we?
It's almost as if the US govt wants having about the lowest voter turnout in the western world. Get rid of the machines & replace them with simple hand counted 'tick the box' paper ballots & I bet the turnout increases at least 10%,
or do anything else to increase the perceived accuracy
then change the vote to saturday & I bet turnout increases at least another 10%.
IMHO the real cause of the low voter turnout is apathy and resignation. When the choice is voting for the puppet on the left and the puppet on the right, there's really little point in bothering, because both puppets are being controlled by the same people.
Whether it's tweedledum/tweedledumber choices, a perception that the election is rigged, or a perception that the district is so lopsided that the outcome is predtermined, it amounts to a feeling that the vote doesn't matter - so why bother. And all three of those are bad.
But there's a fourth reason for not voting - or at least not voting on a particular candidate or issue: If the voter doesn't really care which way it comes out. THEN he shouldn't vote. (To make a random choice, or worse - a systematic choice along with other voters - corrupts the process.)
This is because the purpose of an election is NOT to count "all the people's opinions". It is to head off civil war by figuring out in advance which side would win - effectively, to hold the civil war without getting a lot of people hurt or killed.
As such, people should not be voting unless the matter is important enough to them that they would actually put some effort into DOing something about it.
The election process WAS a reasonably good model: Time off from other pursuits to register and vote - about as much for the rank-and-file electorate as going down to a recruiting office. More for the people running the election - the noncoms of the army. Still more for the party functionaries (commissioned officers) and candidates (the general and general staff). Elegibility to vote going specifically to groups that have historically proven their ability and will to organize mass violence.
Trying to get people to vote just to get a high participation level, or anything that reduces the perceived accuracy of the count, creates the risk that the losing side might think they can reverse it by force. This tends to destabilize the society.
(Also: It doesn't model assassination or asymmetric warfare {terrorism}. So those are still with us from time to time.)
On the flip side, this wouldn't jive with Diebold, because the CEO has already promised the next election to His Favorite Candidate. And, before the Rabid Right flames me for being pissed at the CEO of Diebold, remember -- I'm mad because it's a conflict of interest, not because he's a Republican. I'd be screaming murder if the Democrats, who I don't like much either, tried to pull the same crap.
I agree with you completely except for one thing: I think it makes a difference that he's a Republican.
My perception of the members of the parties is that the Republicans tend to be rule-bound and the Democrats to be "whatever works". So the Republicans are likely to react to a perception of error in the voting process by trying to fix it, while the Democrats have little incentive to do so in the districts they control (since it's still electing them, so who cares?)
But the perception that the system might be rigged by the Republicans gives a BIG incentive to the Democrats to work toward fixing it. (Or at least to make it possible to check whether the Republicans have rigged it. B-) ) Republicans, on the other hand, are likely to ALSO react by trying to fix it, both because of the rule-bound thing and because it makes them look bad and they care about that.
So with a Republican activist in control of the major e-voting machine manufacturer I expect BOTH parties to be trying to get it fixed ASAP. If it were a Democratic activist I'd expect the Democrats to let it ride.
(And maybe even the Republicans would let it ride, rather than raise a stink - much as Nixon resigned rather than further muddy the Presidential office, but Clinton fought regardless of the consequences.)
My objective is to get the elections as accurate as possible. So this makes the head of Diebold being a Republican a GOOD thing, IMHO. B-)
why couldn't the database on the back end be configured to flag any ballots that seemed irregular for inspection? for instance, if the counting machine recorded ballot #41768 as being entirely blank, this could be flagged [...]
Better yet - why is the mark read as "yes/no" rather than "yes/no/maybe" by the optics?
Back in 1966(!) I had a job that included operating an IBM optical mark reader. It did exactly that, grading each potential mark as black/grey/white. If there was a single mark on the paper that it considered grey, it could be programmed to kick the sheet out into a separate hopper for correction and reentry.
Of course for an election the stack in the separate hopper would be set aside for manual examination (to see if the "grey" marks were a light vote, an erased change, or a paper flaw) and manual tabulation.
Here we are 39 years later and the technology has gone BACKWARD on its way to incorporation into what is arguably the most important tabulation job in the country.
"Seeing statements about at least some of these errors being caught almost purely by chance is very disconcerting."
Seeing statistical sampling characterized as "found the errors purely by chance" in order to create FUD is more than disconcerting. It's appalling. And it's a prime example of "how to lie with statistics".
Yes, it's technically accurate, since chance was carefully DESIGNED INto the procedure. But the characterization uses a different meaning of "chance" to imply that the discovery of the errors was a lucky accident.
This is using a pun to tell a lie. In fact the procedure did EXACTLY WHAT IT WAS SUPPOSED TO DO - discover that there was a problem, and drive further investigation to characterize the problem and correct it, both in this and in future elections.
This event:
- Shows one reason a paper trail is needed. (It doesn't directly address deliberate software or database tampering.)
- Provides a counter-argument to claims that optically-scanned paper ballots are an acceptable substitute for machine-printed paper trail ballots. (An optically-marked ballot may look just fine but scan incorrectly.)
Touch-screen machines that print a voter-readable paper trail currently appear to be the most reliable solution for error-resistant and cheat-resistant elections.
His response was [...] a system that "anonymizes" data by an encryption technique called hashing. Because the data are scrambled, [...] secret watch lists can be distributed to private entities, all without fear--because they can't be read
Although this is a step in the right direction [...]
Actually it does NOTHING to prevent misuse of the lists as blacklists - or even vigilante hit lists.
As a blacklist: Put in the name/identifying information, turn the crank, out comes a yes/no. Dump on the subject if it's yes. You've just extracted the entries for everybody you interact with who's on the list, just in time to dump on 'em.
As a hit list: Put in the names/info of everybody in town (or everybody who appears in ANY public database). Turn the crank. Out comes a list list yes/nos. Discard the names that generated "no"s and you have your hit list. If you use an extensive enough database of people you end up extracting ALL from the "encrypted" list (plus any false-positives generated by the hash function).
So the hashing function does NOTHING to secure the list (except maybe pollute it with extra false-positives.)