>"If you insist on paper you're tying elections to an old technology," he told internetnews.com.
In the name of the Flying Spaghetti Monster, either it works or it doesn't.
I was just talking to a friend who consults on DP for banks, having worked her way up from being a teller. They keep multiple records of everything and crosscheck everything. Double-checking begins at the earliest stages of data rollup. Humans look over the results from machines.
Paper "trails" do have the drawback that apparently voters never look at them. Currently I lean toward optical scan, filled in by the voter and not by machine, with readers set to reject invalid ballots with helpful error messages ("Looks like you voted twice for Congress")and trigger a shred-it-log-it-replace-it procedure.
o Exploits will be in older code. o The first "exploits" announced will be simply userland Trojans, as will most that follow. o Old-style remote exploits will be unusual and dramatically rarer than we're used to. o Nobody will notice the difference. The media will lump all problems together and the reports will boil down to "LOL V1st4 pwned".
MS has hunted down unsafe APIs and banned crypto algorithms that are damaged (MD5) or that nobody can figure out how to use correctly (RC4). They compile with stack canaries. They've added address space layout randomization. A large number of people in Canada will forever snarl at me in derision for saying this, but Microsoft is beginning to absorb lessons from the success of OpenBSD.
It's never going to be the same, of course. There's not enough money in the world to audit Microsoft's cetacean code base to OpenBSD standards and I can't believe the design of Windows would support privilege separation.
The paper points to a lack of serious integrity checking, which by itself opens up a whole family of crypto attacks. But another bad part is that the "switch" can rewrite PIN block formats. Of which there are too many. The attack sequence is to translate the PIN block into a format which doesn't contain an account number (but which does contain random data, the designers weren't quite that stupid), and then translate that block into another format which does include an account number. Which means you supply an account number in the API. Which means you have an encrypted block with a known account number, one of 10,000 possible PINs, and no padding or randomization. Which means brute force will get you the PIN really fast.
There's another lesson about allowing too many options in your crypto protocols:
It is well known that when several PIN block formats are available the security of the whole system degrades to the security of the weakest PIN block format. The attacks demonstrate that reformatting capability between different PIN block formats allows an attacker to abuse weaknesses of both formats.
A spokesman pointed out that you'd have to be on the same LAN segment as the Hardware Security Module to launch this attack. Considering that a technician with an infected laptop once put viruses onto ATMs, this is less comforting than it might be.
(Still trying to wrap my head around every "switch" (router?) in the network decrypting and re-encrypting the PIN block. These being systems outside the control of the data owner).
The almost successful Sumitomo Mitsui wire transfer heist is a spectacular example. The crooks posed as cleaning staff, installed keyloggers, used each set of credentials to log in and steal a higher set, and wound up with S.W.I.F.T. passwords. No word on what stopped them from transferring the intended 220 million pounds, but there are some human reviews of transfers that large.
Guess #1: there's a regulatory requirement to go through the motions but nobody really wanted the results.
Guess #2: an executive was smug about security procedures and thought "I can brag at the next meeting about how I did such a great job that even a professional penetration tester couldn't get in".
It's like an airlock: two doors in series, only one of which can open at a time. Crooks hate things that could slow down a getaway and if you implement your access check on someone in the middle with both doors locked, well, if they're a crook you've got them in custody.
"Nailing" seems an inadequate term to describe building a house, and "lying" seeems inadequate for the long-planned and carefully parlayed scams that end up with people FTP'ing their source repository to Kevin Mitnick and thinking it's a good idea.
All of which is really a distraction. Sticky notes on the monitors? If someone's that close they can install a hardware keylogger in a matter of seconds or RAT and rootkit the machine with a live CD in a few minutes. The only security improvement you get from taking down the sticky notes is against casual or opportunistic attacks, which is not nothing, but face the fact that physical access means Game Over.
>The main problem is that Iraq is an artificial state, with little real unifying history, religion, or any common identity.
If anything you understate the case. The British took a canvas sack and sewed it shut with two Siamese cats and a dog inside. Now we're holding the bag.
>By having a large troop presence in the center of the Middle East, the pentagon intends to keep Iran, Syria, Saudi Arabia, and all the major powers in line.
It's been US policy since the end of World War 2 to keep effective veto power over the Middle East, preventing either foreign control or hostile local control. The massive bases planned for Iraq by the Project for a New American Century were a continuation of that policy, though invading a country to put the bases in place was something of an innovation. In fact that was probably the top priority: after all, the oil was flowing already, at "market" prices, and in greater quantities than today. Which brings us to a quibble:
>the US troops serve their purpose, guarding the valuable oil pipelines.
Pipelines are almost impossible to defend and nobody there can keep them open. The worst problem is that there's money in keeping them closed: shifting the output to tanker trucks makes retail hijacking and extortion simple.
>I think the people in charge of this whole charade knew this was going to happen all along.
It looks like there were too many grand ideas floating around and they got in each other's way. "Regime change" conducted according to amoral realpolitik would have left one of Saddam's generals or torturers in charge of a puppet government that would keep the oil pumping, keep Iran out, and generally meet all the US strategic goals. But the "spread democracy" idealists/fantasists/take-your-pick-but-no-way-con servatives got hold of the occupation and dismantled the Baathist power structure, and never said to the army "If you sneak back now we'll pay you".
>Did America have "overwhelming government infrastructure" in 1789?
Yes.
America had governors who were serving for a limited term, parliamentarians who were heir to centuries of English practice, and an entire common law system to back it up. There was a widespread understanding of human rights, which is a kind of infrastructure. Local governments had revenue sources and a civil service who generally didn't steal the revenue, a kind of infrastructure Iraq is missing. There was a common sense of national identity, something the Federalist Papers pointed out as reason to unify. Iraq has never had that.
Seriously, the worst we could possibly do is Yet Another mass extinction event, and we don't look close to that. The planet ran fine and sustained a lot of life without ice on the poles. The changes coming down the pike are going to hurt us a lot worse than they'll hurt the planet: we're trying to feed six billion humans using climate-sensitive crops.
Nuclear plants generate electricity. This is very good, because it means they're displacing coal, but they're not a solution to vehicle propulsion, home heating, or industrial process heat.
Convert to electric cars and trucks, and then you're talking.
>What would be the purpose of lobbying a school to insert more pictures rather than equations? To what end?
Textbook contracts are big money. Crap is cheaper to produce than quality textbooks are. The decision makers on school boards and curriculum committees have lots of possible motives, only a few of which relate to educating students. It may be easier to sell them a picture book with a congenial message than to sell them an equation-filled book they don't even understand themselves.
The next step beyond that actually saves effort, namely installing compact fluorescents. They don't need to be changed as often. For overhead light bulbs that would be worth it even if they used the same amount of electricity as incandescents.
LED home lighting is still for special effects, except that I've installed an LED bulb for that one fixture every house has which it's suicidal to try to reach. It sips a grand total of three watts and illuminates the stairs to the point of safety even with everything else burned out. I felt weird paying $30 for a light bulb until my wife pointed out "You mean $30 for specialized and highly effective safety equipment?"
>how long is it going to be before it becomes to not be profitable to make music?
It's already unprofitable to *make* music, and pirates aren't the problem. Look at the terms of a normal record label contract. The label gets to write checks to themselves against your hypothetical royalties for any "service" they choose to provide at whatever price pleases them. The money is in *distributing* music, and will be until the economics of moving bits around finally destroy the business of moving plastic discs around.
It's already unprofitable to make music, and that's never stopped a musician. It might stop opportunists and scam artists. List the great music of history, list the music written for a profit motive, and meditate upon the intersection of the two sets.
Stop singing "Happy Birthday"? Stop ripping purchased CDs to our iPods? We've heard a music executive who has access to lawyers say that doing so is infringement. Stop making temporary copies in RAM? Court precedent says that's copying.
And then get sued anyway, because the RIAA demonstrably doesn't care whether the defendant has never owned a computer?
Contractual: you must honor commitments. If you've promised God to pray five times a day, if you're promised your wife to stay with her, then those are moral obligations. On this theory, ask whether you've promised to work on the project yet.
Textual: you follow what your holy book says. There has just got to be some Bible verse against Flash.
Compassionate: you ask whether people are hurt or helped by your actions. Will the company be better off? The customer? The customer's customers? Start thinking about those last and you have a reason not to work on tobacco ads, for example.
Reciprocal: you follow the Golden Rule. What would you hope for if you were one of the parties affected by your decision?
Foreseeable consequences: what will follow from your actions?
Arbitrary crap: you grew up around people who thought something was immoral, like say interracial marriage, and you've never checked the idea against any kind of principle. This is the most common approach.
Whatever standard you use, there has to be a set of priorities to go with it. You're using the job to support your family. You'd have to quit if the place were kidnapping the homeless and turning them into Soylent Green, but for much else you have to balance against your family's well being. You have talked this over with your wife??
If you were digging a mine and hit a vein of old computers, wouldn't you think of it as good ore?
Commercial silver and gold deposits are measured in ounces per ton. Commercial copper stays in production at 2% and companies seem to consider a find of 0.13% worth reporting. Plastics would be a problem but then you always have tailings from a mine. Now imagine that the ore doesn't require digging, and in fact people will deliver it to you and pay you to take it.
If people wanted quality images, wouldn't they be up in arms against the cable companies with failing amplifiers? Wouldn't they demand lower compression? Wouldn't they reject the TVs that default to a contrast setting that causes blooming? Wouldn't they reject sets with phosphors that don't match the color standards? Wouldn't they be rejecting sets that *have the wrong color for white*?
>anyone who doesn't feel HD is a worthwhile upgrade SERIOUSLY needs to get their eyes checked.
If you've fallen to the level of insulting customers then it's time to step back and re-evaluate your product. The last time I encountered a salesman like that... well, I'd already made up my mind about the product but he removed any lingering doubts.
>It's not an early adopter device any more, but it hasn't even come close to reaching critical mass in the general populace yet.
There's a marketing book that's worth reading, and it's about this exact situation. Products do not move smoothly from early adopters to early majority. There's a pit in between the two that many products fall into.
The book, "Crossing the Chasm", explains that you have to make the transition to your new product as smooth and slick as teflon on teflon, or normal people will never generate good word of mouth. An example of a brilliant success at this is the Toyota Prius, which spends a significant amount of software simulating the artifacts of a 20th-century car, just to allow buyers to slide right into it without an adjustment.
If the HD industry were poised for success you'd see plug-and-play installations that didn't require setup by a consultant, no obstructive DRM, and standardized cabling.
He had to provide gap-free radio coverage on a factory floor full of large metal objects that cast RF shadows. The usual answer is the RF equivalent of indirect lighting. Instead of single antennas or groups of antennas, you use Radiax, coax with expensively engineered precision leakage from the shielding. You loop that around the ceiling, and then the whole factory floor is bathed in RF.
Radiax was a bit spendy for the project so he went to Radio Shack and got a spool of their regular coax. It did the job fine.
Slashdot Mirror
>"If you insist on paper you're tying elections to an old technology," he told internetnews.com.
In the name of the Flying Spaghetti Monster, either it works or it doesn't.
I was just talking to a friend who consults on DP for banks, having worked her way up from being a teller. They keep multiple records of everything and crosscheck everything. Double-checking begins at the earliest stages of data rollup. Humans look over the results from machines.
Paper "trails" do have the drawback that apparently voters never look at them. Currently I lean toward optical scan, filled in by the voter and not by machine, with readers set to reject invalid ballots with helpful error messages ("Looks like you voted twice for Congress")and trigger a shred-it-log-it-replace-it procedure.
>"It's quite hard to destroy the Earth."
Does that statement make anyone else nervous? I mean, does that sound like experience talking?
o Exploits will be in older code.
o The first "exploits" announced will be simply userland Trojans, as will most that follow.
o Old-style remote exploits will be unusual and dramatically rarer than we're used to.
o Nobody will notice the difference. The media will lump all problems together and the reports will boil down to "LOL V1st4 pwned".
MS has hunted down unsafe APIs and banned crypto algorithms that are damaged (MD5) or that nobody can figure out how to use correctly (RC4). They compile with stack canaries. They've added address space layout randomization. A large number of people in Canada will forever snarl at me in derision for saying this, but Microsoft is beginning to absorb lessons from the success of OpenBSD.
It's never going to be the same, of course. There's not enough money in the world to audit Microsoft's cetacean code base to OpenBSD standards and I can't believe the design of Windows would support privilege separation.
There's another lesson about allowing too many options in your crypto protocols:
A spokesman pointed out that you'd have to be on the same LAN segment as the Hardware Security Module to launch this attack. Considering that a technician with an infected laptop once put viruses onto ATMs, this is less comforting than it might be.
(Still trying to wrap my head around every "switch" (router?) in the network decrypting and re-encrypting the PIN block. These being systems outside the control of the data owner).
The almost successful Sumitomo Mitsui wire transfer heist is a spectacular example. The crooks posed as cleaning staff, installed keyloggers, used each set of credentials to log in and steal a higher set, and wound up with S.W.I.F.T. passwords. No word on what stopped them from transferring the intended 220 million pounds, but there are some human reviews of transfers that large.
Guess #1: there's a regulatory requirement to go through the motions but nobody really wanted the results.
Guess #2: an executive was smug about security procedures and thought "I can brag at the next meeting about how I did such a great job that even a professional penetration tester couldn't get in".
It's like an airlock: two doors in series, only one of which can open at a time. Crooks hate things that could slow down a getaway and if you implement your access check on someone in the middle with both doors locked, well, if they're a crook you've got them in custody.
"Nailing" seems an inadequate term to describe building a house, and "lying" seeems inadequate for the long-planned and carefully parlayed scams that end up with people FTP'ing their source repository to Kevin Mitnick and thinking it's a good idea.
My explanation of why you *should* write down your password. Bruce Scheier has made the same point.
All of which is really a distraction. Sticky notes on the monitors? If someone's that close they can install a hardware keylogger in a matter of seconds or RAT and rootkit the machine with a live CD in a few minutes. The only security improvement you get from taking down the sticky notes is against casual or opportunistic attacks, which is not nothing, but face the fact that physical access means Game Over.
>The main problem is that Iraq is an artificial state, with little real unifying history, religion, or any common identity.
n servatives got hold of the occupation and dismantled the Baathist power structure, and never said to the army "If you sneak back now we'll pay you".
If anything you understate the case. The British took a canvas sack and sewed it shut with two Siamese cats and a dog inside. Now we're holding the bag.
>By having a large troop presence in the center of the Middle East, the pentagon intends to keep Iran, Syria, Saudi Arabia, and all the major powers in line.
It's been US policy since the end of World War 2 to keep effective veto power over the Middle East, preventing either foreign control or hostile local control. The massive bases planned for Iraq by the Project for a New American Century were a continuation of that policy, though invading a country to put the bases in place was something of an innovation. In fact that was probably the top priority: after all, the oil was flowing already, at "market" prices, and in greater quantities than today. Which brings us to a quibble:
>the US troops serve their purpose, guarding the valuable oil pipelines.
Pipelines are almost impossible to defend and nobody there can keep them open. The worst problem is that there's money in keeping them closed: shifting the output to tanker trucks makes retail hijacking and extortion simple.
>I think the people in charge of this whole charade knew this was going to happen all along.
It looks like there were too many grand ideas floating around and they got in each other's way. "Regime change" conducted according to amoral realpolitik would have left one of Saddam's generals or torturers in charge of a puppet government that would keep the oil pumping, keep Iran out, and generally meet all the US strategic goals. But the "spread democracy" idealists/fantasists/take-your-pick-but-no-way-co
>Did America have "overwhelming government infrastructure" in 1789?
Yes.
America had governors who were serving for a limited term, parliamentarians who were heir to centuries of English practice, and an entire common law system to back it up. There was a widespread understanding of human rights, which is a kind of infrastructure. Local governments had revenue sources and a civil service who generally didn't steal the revenue, a kind of infrastructure Iraq is missing. There was a common sense of national identity, something the Federalist Papers pointed out as reason to unify. Iraq has never had that.
>the fact is that the Earth is Dying(tm)
Has Netcraft confirmed it?
Seriously, the worst we could possibly do is Yet Another mass extinction event, and we don't look close to that. The planet ran fine and sustained a lot of life without ice on the poles. The changes coming down the pike are going to hurt us a lot worse than they'll hurt the planet: we're trying to feed six billion humans using climate-sensitive crops.
Nuclear plants generate electricity. This is very good, because it means they're displacing coal, but they're not a solution to vehicle propulsion, home heating, or industrial process heat.
Convert to electric cars and trucks, and then you're talking.
>It's real easy to say, "We can keep our principles and take their money" but history shows that that's just not true.
History going back a long way: "Where the man's treasure is, there is his heart also" is from Matthew 6:21.
>What would be the purpose of lobbying a school to insert more pictures rather than equations? To what end?
Textbook contracts are big money. Crap is cheaper to produce than quality textbooks are. The decision makers on school boards and curriculum committees have lots of possible motives, only a few of which relate to educating students. It may be easier to sell them a picture book with a congenial message than to sell them an equation-filled book they don't even understand themselves.
The next step beyond that actually saves effort, namely installing compact fluorescents. They don't need to be changed as often. For overhead light bulbs that would be worth it even if they used the same amount of electricity as incandescents.
LED home lighting is still for special effects, except that I've installed an LED bulb for that one fixture every house has which it's suicidal to try to reach. It sips a grand total of three watts and illuminates the stairs to the point of safety even with everything else burned out. I felt weird paying $30 for a light bulb until my wife pointed out "You mean $30 for specialized and highly effective safety equipment?"
>how long is it going to be before it becomes to not be profitable to make music?
It's already unprofitable to *make* music, and pirates aren't the problem. Look at the terms of a normal record label contract. The label gets to write checks to themselves against your hypothetical royalties for any "service" they choose to provide at whatever price pleases them. The money is in *distributing* music, and will be until the economics of moving bits around finally destroy the business of moving plastic discs around.
It's already unprofitable to make music, and that's never stopped a musician. It might stop opportunists and scam artists. List the great music of history, list the music written for a profit motive, and meditate upon the intersection of the two sets.
>Stop committing copyright infringement maybe?
Stop singing "Happy Birthday"?
Stop ripping purchased CDs to our iPods? We've heard a music executive who has access to lawyers say that doing so is infringement.
Stop making temporary copies in RAM? Court precedent says that's copying.
And then get sued anyway, because the RIAA demonstrably doesn't care whether the defendant has never owned a computer?
make sure you understand the concept of "vexatious litigant", when a court might decide that you are one, and what can happen next.
What do you call someone who gets legal advice on Slashdot? "Inmate".
Contractual: you must honor commitments. If you've promised God to pray five times a day, if you're promised your wife to stay with her, then those are moral obligations. On this theory, ask whether you've promised to work on the project yet.
Textual: you follow what your holy book says. There has just got to be some Bible verse against Flash.
Compassionate: you ask whether people are hurt or helped by your actions. Will the company be better off? The customer? The customer's customers? Start thinking about those last and you have a reason not to work on tobacco ads, for example.
Reciprocal: you follow the Golden Rule. What would you hope for if you were one of the parties affected by your decision?
Foreseeable consequences: what will follow from your actions?
Arbitrary crap: you grew up around people who thought something was immoral, like say interracial marriage, and you've never checked the idea against any kind of principle. This is the most common approach.
Whatever standard you use, there has to be a set of priorities to go with it. You're using the job to support your family. You'd have to quit if the place were kidnapping the homeless and turning them into Soylent Green, but for much else you have to balance against your family's well being. You have talked this over with your wife??
If you were digging a mine and hit a vein of old computers, wouldn't you think of it as good ore?
Commercial silver and gold deposits are measured in ounces per ton. Commercial copper stays in production at 2% and companies seem to consider a find of 0.13% worth reporting. Plastics would be a problem but then you always have tailings from a mine. Now imagine that the ore doesn't require digging, and in fact people will deliver it to you and pay you to take it.
Why are you so sure there's a market?
... well, I'd already made up my mind about the product but he removed any lingering doubts.
If people wanted quality images, wouldn't they be up in arms against the cable companies with failing amplifiers? Wouldn't they demand lower compression? Wouldn't they reject the TVs that default to a contrast setting that causes blooming? Wouldn't they reject sets with phosphors that don't match the color standards? Wouldn't they be rejecting sets that *have the wrong color for white*?
>anyone who doesn't feel HD is a worthwhile upgrade SERIOUSLY needs to get their eyes checked.
If you've fallen to the level of insulting customers then it's time to step back and re-evaluate your product. The last time I encountered a salesman like that
>It's not an early adopter device any more, but it hasn't even come close to reaching critical mass in the general populace yet.
There's a marketing book that's worth reading, and it's about this exact situation. Products do not move smoothly from early adopters to early majority. There's a pit in between the two that many products fall into.
The book, "Crossing the Chasm", explains that you have to make the transition to your new product as smooth and slick as teflon on teflon, or normal people will never generate good word of mouth. An example of a brilliant success at this is the Toyota Prius, which spends a significant amount of software simulating the artifacts of a 20th-century car, just to allow buyers to slide right into it without an adjustment.
If the HD industry were poised for success you'd see plug-and-play installations that didn't require setup by a consultant, no obstructive DRM, and standardized cabling.
He had to provide gap-free radio coverage on a factory floor full of large metal objects that cast RF shadows. The usual answer is the RF equivalent of indirect lighting. Instead of single antennas or groups of antennas, you use Radiax, coax with expensively engineered precision leakage from the shielding. You loop that around the ceiling, and then the whole factory floor is bathed in RF.
Radiax was a bit spendy for the project so he went to Radio Shack and got a spool of their regular coax. It did the job fine.