We've got two separate discussions going here. Let's review the facts to sort things out...
Battery dies...
Apple says they want $255
Guy googles and finds a $50 battery
Ipod breaks when he's installing the battery
Guy gets pissed that changing a battery is such a PITA and makes a video
Apple drops $255 fee to $100
Guy's video soaks his bandwidth
You offer guy access to gobs of tax-payer funded bandwidth if he adds a stipulation to his web page.
You get pissed when he takes you up on the bandwidth but doesn't make the change that you wanted him to make.
Let's make a couple of things clear
A - it's not your bandwidth Dave - it's taxpayer supported bandwidth.
B - Maybe the guy is still torqued because he's out an iPod?
It's beyond me where the hell you get off giving away that kind of bandwidth and then get high and mighty because Neistat is pissed off that he's out an iPod. In one day you gave away publicly funded bandwidth with a market value that's worth more than an iPod. Where I come from, people don't reckon it's a-ok for a company to soak them every time a battery dies. In your world where you have access to so much bandwidth that you can give it away like candy, maybe it is.
Even a dewar flask isn't a perfect insulator. On Apollo 13, one of the survival problems the astronauts had was just keeping warm enough. There wasn't enough fuel to spin the craft after the oxy tank exploded to keep the temperatures evenly distributed across the LEM. The astronauts got very cold as a result.
I'm surprised that you would admit to giving away tax supported bandwidth. The best deal I've seen on bandwidth has been around $65 for 40 gigs/month and you're admitting you not only gave away.7 terabytes of bandwidth in one day but would have continued to grant them access indefinitely had they put up a link you stipulated.
That strikes me as incredibly wasteful of public tax dollars that you can just give it away like that. Either you have too much publicly funded bandwidth or you have no business giving away a scarce resource to promote Apple.
The fact remains that you can reasonably expect some li-ion batteries to die after one year of service, that some did die, and that Apple is just now, 2 years into the product life cycle, making some provision for people whose battery has died. I fail to see how that's anyway ok for the people who either forked out $255, shelved their iPods, or busted their iPod trying to replace the battery because Apple didn't want to make it easy for the consumer to change a battery.
With business ethics like that, it's no wonder Apple's share of the PC market has shrunk from 75% to 3%. They keep this up and the same thing will happen to the iPod.
As to "incriminating", I'm not too sure. Do they really prove that Dave Schroeder offered to give away terabytes of bandwidth to promote Apple's website?
Digikey has been mentioned a couple of times as an iPod battery source. I went to their website to check it out and couldn't find a battery that matched the iPod's. Do you have a Digikey part number?
That's not true. Until Nov 14, 2003, Apple wanted $255 to service iPod's with dead batteries. PDASmart, the $50 source you refer to, didn't source the battery until June 2003, 20 months after the iPod's introduction. The battery has a lifespan between 1 and 2 years. That means that there are people out there on the wrong side of the Bell Curve who have shelved their iPods because they didn't think paying Apple $255 to replace a battery on a $400 item was a fair shake. May not be a problem to you but ask those people why don't you?
Roger Simon, a screenwriter, said his screeners have either been delivered by an indifferent Fedex driver who didn't care who signed, or just plain left on his doorstep. Given that, he suggests that just possibly, the MPAA's agenda on this issue isn't piracy.
but if its going twice as fast, can we expect it to burn more fuel, too?
Not if it flies 4 times higher than Concorde did. Not a lot of drag at 210,000 feet. Hypersoar was supposed to periodically ignite its engines to climb to the peak and coast for most of a cycle - sort of like a roller coaster. Only problem is, the designer works at Lawrence Livermore Lab. The lab has an unfortunate history of over promising and under delivering.
Why they didn't go the Malaysia or Philippines as you mentioned I don't know. Perhaps the current unrest in the world?
Because the Germans guaranteed the necessary loans. AMD's balance sheet has been so bloody of years past, that there wasn't anyway that AMD would have been able to swing the loans without a consigner. Aren't a lot of parents out there who can co-sign a $2.4 Billion note. From the article...
Press reports quoted regional government sources in the state of Saxony as anticipating that AMD will receive a debt guarantee from the German governement by next week. Other sources report that the Saxonian government has already provided guarantees for 80 percent of the required loans.
If I was a German taxpayer, I'd be wondering how AMD can possibly pay off the note.
There are so many different ways to hack a machine's output that you can never know for sure that the machine's product is correct. Ordinarily, it doesn't happen because you need access and motive to do it. Your spreadsheet results are acceptably reliable because there isn't someone who has an interest in skewing the spreadsheet's output. You recover from the occasional hardware glitch and go on.
A vote is something else...there's lots of motivation to steal an election. There isn't any way of knowing, given today's operating systems, that no one has either hacked the code in ROM or loaded a hook that'll modify the vote as desired. For every measure you propose to thwart theft, there's a counter measure. That's just the intentional attacks. There are hardware failures to contend with as well. There isn't a straightforward way to backup a vote and know for certain that the backup is accurate. Distributed tallying/backup just introduces another error source.
Voting is an activity that is best left to humans doing the tallying. When properly implemented, it's trustworthy unlike what we're currently doing. I know this is/. heresy but there are tasks where a technological solution should not be applied - voting is one of them.
More importantly, it may not have been Windows' fault given that tho box was under a denial of service attack. What the failure shows is relying on the Internet to deliver emergency messages is poor system design regardless of what OS is running. It also shows that the power grid can be brought down from a desktop when trees are falling due to high winds.
If this DOS wasn't intentionally targeting the power grid, I wouldn't be at all surprised that the next one does intentionally target the grid.
The network is expected to be available to 723,000 residents in 248,000 households and 34,500 businesses. Prices would vary considerably depending on the service, though basic high-speed Internet access is expected to cost about $28 a month.
and a little later says
As of October, only 180,300 homes had direct access to fiber optic lines; 64,700 were actually connected, according to Render, Vanderslice & Associates, a market research firm in Tulsa, Okla.
So, 248,000 households + 34,500 business + a growth factor gives about 300,000 potential customers. Using the current light-to-the-curb base of 180,300 with 64,700 takers says that about 1 in 3 people will buy the service if it's available. Apply that 1 in 3 to the 300,000 potential customers (assuming no overlap...) and you've got a client base of about 100,000 sales. Multiply that by $28 per month gives $2,800,000 per month or about $67,000,000 per year. If you had zero costs, that would be a revenue stream of about 16% on your $420,000,000.
But of course, there are costs. Figure 2/3 of the revenue goes to operating expenses leaving a healthy 33% margin and you're looking at $22,000,000 to pay down the $420,000,000 loan. That's about a 5% revenue stream to pay back a 6% note, leaving you 1% in the hole.
Granted, these are back of the envelope numbers but if my 33% margin guess is optimistic, Utopia will be known as Dystopia.
Lorentz showed that small perturbations could lead to unexpected outcomes. If the system you're modeling has a growing or shrinking non-linear component, you can't make any long term forecasts as to how the model will behave.
What's worse is making statements like know with certainty when there are so many unknowns. The only people who know with certainty are charlatans selling snake oil.
You can chant all you want but you're missing Lorenz's key point. His finding wasn't confined to weather systems. His observation applied to any system that can be described by self-referential, non-linear equations. He should have won the Nobel in Economics because self-referential, non-linear equations are the bread and butter of economists.
Your point is valid - one sample doesn't represent a population. So let me sketch in a few more points...
Club of Rome. Back in the early 70's there was an outfit called the Club of Rome. They ran their computer models and claimed that the world would run out of gas in 1985. Got a lot of press when they made their claim. None of them were around in 1985 to admit they were wrong.
Ehrlich and Simon. In 1980, Julian Simon (Univ. of Maryland) challenged Paul Ehrlich (Stanford) to a bet. Erhlich had been harping on how with the exploding world population we were going to run out resources real soon and prices were going to skyrocket. Simon called him on it and let him choose any list of goods he wanted. If the basket of goods 10 years later was higher, Ehrlich would win, if the basket was cheaper (in real dollars) then Simon would win. Oil, gold, silver, wheat, and a lot of other goods went into the basket. Simon was right - the basket's price declined over the 10 year span.
Global Warming. There's a graph that oxygen isotope variations over the past 500,000 years. The data are derived from ocean sediment cores and are a useful surrogate for temperatures. The isotope ratios are inversely proportional to temperatures and closely track interglacial warming periods. I can't find the original paper but you can see the data graphed on page 10 of this NOAA document. The graph shows a roughly 100,000 year cycle to global temperatures with periods warmer than today. We're right on track right now towards a global uptick just like the past 4 times over the past 500,000 years. We weren't around 410,000, 320,000, 200,000 or 120,000 years ago to cause the last temperature spikes but they happened anyway. The earth may be getting warmer right now but it's not likely that it's due to us pumping CO2 into the atmosphere. Look at the CO2 and Methane data and the temperature correlation isn't all that great.
Weather forecasting. In 1963, Lorenz demonstrated that you can't make reliable long term weather forecasts even when you have a perfect weather model and all the data accurate to 6 decimal places. It was a key finding and yet you still have people making global warming forecasts for the next 100 years as if Lorenz hadn't already demonstrated they can't possibly know what they're talking about.
The environmental movement has done a lot of good in making us take stock of how we're disposing of our waste. Los Angeles air and San Francisco Bay are a lot cleaner today than they would have otherwise been had it not been for the hoopla. But at the same time, you have to be very skeptical when someone tries to tell you we're going to destroy the earth if we keep doing what we're doing. In geologic terms, we've been around for a brief moment and the earth has managed some amazingly self-destructive feats without us and yet here we are.
Why do people think environmentalists would be biased, anyway?
Possibly because they admit it?
In John McPhee's Encounters with the Archdruid, David Bower, the former director of the Sierra Club, admits he just made his numbers up. McPhee asks Bower where he found the data for the 'The U.S. has 6% of the world's population but consumes 40% of the world's resourcess' quote. Bower's response was it sounded about right.
This isn't a troll but some of you may think I'm being intentionally inflamatory. All I can tell you is I think what I'm about to write is true.
I think using computers to count vote is a mis-application of technology. My reasons are:
Security. None of the operating systems and hardware in use are designed from the ground up to be secure. The reason is that security and ease-of-use are at loggerheads - get more of one you lose some of the other. One of the key features of every OS I've worked on is the ability to install a daemon somewhere in the message queue so you can remap devices to other purposes. For example, keyboard drivers are easily changed to morph a 'p' into a long sequence of instructions. No matter how well you try to detect a daemon/hook/wedge or whatever you want to call it, if the developer is intent on inserting his code and there are provisions for mapping into user space (I've yet to run on an OS that that couldn't be done) the code can be inserted. That means that open source, closed source, audited source, tested source are all susceptible to modification by a malacious bit of code. It just requires access. Touch screen/punch card/optical scan - it doesn't matter - if you're relying on a computer to do the tally and you can't guarantee that no one has inserted a daemon, you don't have a secure vote.
Little gained. A lot of "improvements" to what's out there right now discuss the idea of a voter-inspectable audit trail. Voter uses a computer to vote and the computer produces a paper ballot that the user can inspect to make sure the computer isn't cheating. There are two things wrong here. First if a computer is going to tally the paper ballot, you're back to point 1. You've just moved the location of the fraud. If the computer is going to tally and the paper is just a backup, then in most cases, a fraud will go undetected. If the fraud is small enough to be within the bounds of statistical uncertainty but large enough to sway the vote, you're not going to catch it unless you hand count the entire population of ballots. Secondly, you're in essence using a machine to mark a piece of paper which a human can just as easily do - you haven't gained anything by introducing the voting machine into the mix.
I think the Canadians who just use a paper and pencil and cross-checked human counters to tally the vote have it right. The whole system is very simple. You mark your ballot, put it in a box. When the poll closes, at least 3 pairs of eyes look at it, one person is the election official, the other two are from opposing parties. When all 3 agree what the vote is, it's tallied as such. They can cross check tallies as they go so you're not running into a transcription problem down the road. The precinct reports its tallies to a higher level up the tree and the results are published so that the three (or more) counters can check the tally was accurately registered at the next level. Anyone who wants to can check the process from start to finish. Open, transparent, accurate and simple. Contrast that to encrypted keys, password maintenance, static discharge induced miscounts, lack of audit trails and the rest of the mess that characterizes the spectrum of American voting techniques and you have to ask - why the hell do we bother using machines to do this when we can do a better job by hand?
There are lots of times that tech is part of a solution. Then there are times, like vote counting, where it is part of the problem. It may be retro and old fashioned but I think it's time we just used paper and pen again. It worked all the way up to the sixties and the country managed then. If our parents and grandparents could manage it, shouldn't we be able to hand count as well?
Hell, these ain't new. I modded my Mazda Rx-4 (yep, 4) in the 70's by turning the car roof into a speaker with a piezo transducer. You could get them at the old Heathkit stores.
Problem then, and probably now, is though they were good at reproducing high frequency, the bass notes weren't so great. You still needed a big old fashioned bass driver if you wanted chest thumping bass.
I was at Dr. Varadarajan's presentation at the O'Reilly conference. One of the attendees asked him what the university paid. He said that VPI paid the standard academic pricing for the quantity, no special deal. VPI got the same price that anyone could get if they were ordering 1100 units of that configuration for any educational customer.
For me, that suffices as documentation. My problem with your original post is it struck me as a "did not" response to a "did too" post.
It's not important for me to believe VPI botched something up. What I prefer is when I read a slashdot post that asserts a postion that the asserter include some documentation that demonstrates they know what they're talking about.
Why? Because you say so? The fact is that Wired reported VT paid full price. Just because you say it's not true doesn't mean anything. Pony up some documentation as to what you're saying and then it'll be a canard.
If someone collects the Xprize before it expires next year, it may be the way to go. The total amount of money spent by all the competitors exceeds the value of the Xprize but the competitors are trying anyway.
Tech prizes go way back. Parliment issued a prize to John Harrison for developing an accurate chronometer. The guy had zero credentials to do it - he was a cabinet maker - but he beat out everyone else and solved a long standing puzzle because of the prize. Paul MacReady won two sequential prizes for developing human powered aircraft. Lindberg won the Oertig prize for crossing the Atlantic.
If Nasa put up a series of substantial prizes for an aircraft capable of reaching LEO, Geostationary Orbit, Lunar Orbit and Lunar landing and Return, I'll bet we'd see a huge surge in space flight for a fraction of what we're spending today for shuttle flights. Nasa may not like the lack of control a prize implies but it would certainly encourage innovation.
Slashdot Mirror
- Battery dies...
- Apple says they want $255
- Guy googles and finds a $50 battery
- Ipod breaks when he's installing the battery
- Guy gets pissed that changing a battery is such a PITA and makes a video
- Apple drops $255 fee to $100
- Guy's video soaks his bandwidth
- You offer guy access to gobs of tax-payer funded bandwidth if he adds a stipulation to his web page.
- You get pissed when he takes you up on the bandwidth but doesn't make the change that you wanted him to make.
Let's make a couple of things clear- A - it's not your bandwidth Dave - it's taxpayer supported bandwidth.
- B - Maybe the guy is still torqued because he's out an iPod?
It's beyond me where the hell you get off giving away that kind of bandwidth and then get high and mighty because Neistat is pissed off that he's out an iPod. In one day you gave away publicly funded bandwidth with a market value that's worth more than an iPod. Where I come from, people don't reckon it's a-ok for a company to soak them every time a battery dies. In your world where you have access to so much bandwidth that you can give it away like candy, maybe it is.Even a dewar flask isn't a perfect insulator. On Apollo 13, one of the survival problems the astronauts had was just keeping warm enough. There wasn't enough fuel to spin the craft after the oxy tank exploded to keep the temperatures evenly distributed across the LEM. The astronauts got very cold as a result.
That strikes me as incredibly wasteful of public tax dollars that you can just give it away like that. Either you have too much publicly funded bandwidth or you have no business giving away a scarce resource to promote Apple.
The fact remains that you can reasonably expect some li-ion batteries to die after one year of service, that some did die, and that Apple is just now, 2 years into the product life cycle, making some provision for people whose battery has died. I fail to see how that's anyway ok for the people who either forked out $255, shelved their iPods, or busted their iPod trying to replace the battery because Apple didn't want to make it easy for the consumer to change a battery.
With business ethics like that, it's no wonder Apple's share of the PC market has shrunk from 75% to 3%. They keep this up and the same thing will happen to the iPod.
As to "incriminating", I'm not too sure. Do they really prove that Dave Schroeder offered to give away terabytes of bandwidth to promote Apple's website?
Digikey has been mentioned a couple of times as an iPod battery source. I went to their website to check it out and couldn't find a battery that matched the iPod's. Do you have a Digikey part number?
That's not true. Until Nov 14, 2003, Apple wanted $255 to service iPod's with dead batteries. PDASmart, the $50 source you refer to, didn't source the battery until June 2003, 20 months after the iPod's introduction. The battery has a lifespan between 1 and 2 years. That means that there are people out there on the wrong side of the Bell Curve who have shelved their iPods because they didn't think paying Apple $255 to replace a battery on a $400 item was a fair shake. May not be a problem to you but ask those people why don't you?
What's really stupid about this is had Apple:
- Made it easy to change a battery and...
- Sourced the battery from the beginning.
nobody would have been complained.Roger Simon, a screenwriter, said his screeners have either been delivered by an indifferent Fedex driver who didn't care who signed, or just plain left on his doorstep. Given that, he suggests that just possibly, the MPAA's agenda on this issue isn't piracy.
Not if it flies 4 times higher than Concorde did. Not a lot of drag at 210,000 feet. Hypersoar was supposed to periodically ignite its engines to climb to the peak and coast for most of a cycle - sort of like a roller coaster. Only problem is, the designer works at Lawrence Livermore Lab. The lab has an unfortunate history of over promising and under delivering.
Because the Germans guaranteed the necessary loans. AMD's balance sheet has been so bloody of years past, that there wasn't anyway that AMD would have been able to swing the loans without a consigner. Aren't a lot of parents out there who can co-sign a $2.4 Billion note. From the article...
If I was a German taxpayer, I'd be wondering how AMD can possibly pay off the note.A vote is something else...there's lots of motivation to steal an election. There isn't any way of knowing, given today's operating systems, that no one has either hacked the code in ROM or loaded a hook that'll modify the vote as desired. For every measure you propose to thwart theft, there's a counter measure. That's just the intentional attacks. There are hardware failures to contend with as well. There isn't a straightforward way to backup a vote and know for certain that the backup is accurate. Distributed tallying/backup just introduces another error source.
Voting is an activity that is best left to humans doing the tallying. When properly implemented, it's trustworthy unlike what we're currently doing. I know this is /. heresy but there are tasks where a technological solution should not be applied - voting is one of them.
More importantly, it may not have been Windows' fault given that tho box was under a denial of service attack. What the failure shows is relying on the Internet to deliver emergency messages is poor system design regardless of what OS is running. It also shows that the power grid can be brought down from a desktop when trees are falling due to high winds.
If this DOS wasn't intentionally targeting the power grid, I wouldn't be at all surprised that the next one does intentionally target the grid.
The Sahara is not anywhere near South Africa - they're at opposite ends of the world's 2nd largest continent.
But of course, there are costs. Figure 2/3 of the revenue goes to operating expenses leaving a healthy 33% margin and you're looking at $22,000,000 to pay down the $420,000,000 loan. That's about a 5% revenue stream to pay back a 6% note, leaving you 1% in the hole.
Granted, these are back of the envelope numbers but if my 33% margin guess is optimistic, Utopia will be known as Dystopia.
Glad to hear Linus is covered. Who is covering Stallman's legal expenses?
The Xcor faq answers a lot of questions but skips the most important one. How much will it cost to develop the Xerus?
What's worse is making statements like know with certainty when there are so many unknowns. The only people who know with certainty are charlatans selling snake oil.
You can chant all you want but you're missing Lorenz's key point. His finding wasn't confined to weather systems. His observation applied to any system that can be described by self-referential, non-linear equations. He should have won the Nobel in Economics because self-referential, non-linear equations are the bread and butter of economists.
The environmental movement has done a lot of good in making us take stock of how we're disposing of our waste. Los Angeles air and San Francisco Bay are a lot cleaner today than they would have otherwise been had it not been for the hoopla. But at the same time, you have to be very skeptical when someone tries to tell you we're going to destroy the earth if we keep doing what we're doing. In geologic terms, we've been around for a brief moment and the earth has managed some amazingly self-destructive feats without us and yet here we are.
Possibly because they admit it?
In John McPhee's Encounters with the Archdruid, David Bower, the former director of the Sierra Club, admits he just made his numbers up. McPhee asks Bower where he found the data for the 'The U.S. has 6% of the world's population but consumes 40% of the world's resourcess' quote. Bower's response was it sounded about right.
I think using computers to count vote is a mis-application of technology. My reasons are:
- Security. None of the operating systems and hardware in use are designed from the ground up to be secure. The reason is that security and ease-of-use are at loggerheads - get more of one you lose some of the other. One of the key features of every OS I've worked on is the ability to install a daemon somewhere in the message queue so you can remap devices to other purposes. For example, keyboard drivers are easily changed to morph a 'p' into a long sequence of instructions. No matter how well you try to detect a daemon/hook/wedge or whatever you want to call it, if the developer is intent on inserting his code and there are provisions for mapping into user space (I've yet to run on an OS that that couldn't be done) the code can be inserted. That means that open source, closed source, audited source, tested source are all susceptible to modification by a malacious bit of code. It just requires access. Touch screen/punch card/optical scan - it doesn't matter - if you're relying on a computer to do the tally and you can't guarantee that no one has inserted a daemon, you don't have a secure vote.
- Little gained. A lot of "improvements" to what's out there right now discuss the idea of a voter-inspectable audit trail. Voter uses a computer to vote and the computer produces a paper ballot that the user can inspect to make sure the computer isn't cheating. There are two things wrong here. First if a computer is going to tally the paper ballot, you're back to point 1. You've just moved the location of the fraud. If the computer is going to tally and the paper is just a backup, then in most cases, a fraud will go undetected. If the fraud is small enough to be within the bounds of statistical uncertainty but large enough to sway the vote, you're not going to catch it unless you hand count the entire population of ballots. Secondly, you're in essence using a machine to mark a piece of paper which a human can just as easily do - you haven't gained anything by introducing the voting machine into the mix.
I think the Canadians who just use a paper and pencil and cross-checked human counters to tally the vote have it right. The whole system is very simple. You mark your ballot, put it in a box. When the poll closes, at least 3 pairs of eyes look at it, one person is the election official, the other two are from opposing parties. When all 3 agree what the vote is, it's tallied as such. They can cross check tallies as they go so you're not running into a transcription problem down the road. The precinct reports its tallies to a higher level up the tree and the results are published so that the three (or more) counters can check the tally was accurately registered at the next level. Anyone who wants to can check the process from start to finish. Open, transparent, accurate and simple. Contrast that to encrypted keys, password maintenance, static discharge induced miscounts, lack of audit trails and the rest of the mess that characterizes the spectrum of American voting techniques and you have to ask - why the hell do we bother using machines to do this when we can do a better job by hand?There are lots of times that tech is part of a solution. Then there are times, like vote counting, where it is part of the problem. It may be retro and old fashioned but I think it's time we just used paper and pen again. It worked all the way up to the sixties and the country managed then. If our parents and grandparents could manage it, shouldn't we be able to hand count as well?
The lemmings morphed into Roadies
Problem then, and probably now, is though they were good at reproducing high frequency, the bass notes weren't so great. You still needed a big old fashioned bass driver if you wanted chest thumping bass.
For me, that suffices as documentation. My problem with your original post is it struck me as a "did not" response to a "did too" post.
It's not important for me to believe VPI botched something up. What I prefer is when I read a slashdot post that asserts a postion that the asserter include some documentation that demonstrates they know what they're talking about.
Why? Because you say so? The fact is that Wired reported VT paid full price. Just because you say it's not true doesn't mean anything. Pony up some documentation as to what you're saying and then it'll be a canard.
Tech prizes go way back. Parliment issued a prize to John Harrison for developing an accurate chronometer. The guy had zero credentials to do it - he was a cabinet maker - but he beat out everyone else and solved a long standing puzzle because of the prize. Paul MacReady won two sequential prizes for developing human powered aircraft. Lindberg won the Oertig prize for crossing the Atlantic.
If Nasa put up a series of substantial prizes for an aircraft capable of reaching LEO, Geostationary Orbit, Lunar Orbit and Lunar landing and Return, I'll bet we'd see a huge surge in space flight for a fraction of what we're spending today for shuttle flights. Nasa may not like the lack of control a prize implies but it would certainly encourage innovation.