Funny since the PDF you linked to mentions "The Face Book". Page 11 under the "Entire Agreement" Section
Yes, I know the PDF I liked to includes that. My point is that FB doesn't think that's actually the agreement that Zuckerberg signed. Let's pose a hypothetical situation:
Ceglia posts an ad on craigslist asking for programming help on a new startup called StreetFax. Zuckerberg responds, and they work out a contract that looks similar to the one that was in the filing, except there was never any mention of "The Face Book", "The Page Book" or anything *other* than the StreetFax stuff. They both sign it, Zuckerberg gets his money, Ceglia gets the StreetFax stuff, and then they part ways sometime in 2003 and never speak again.
Fast forward to June of 2010, and suddenly Ceglia is providing a poor quality copy of a contract that looks similar to the one Zuckerberg had signed 7 years ago, but now includes stuff about Face Book that was never on the original contract, and is claiming majority ownership of a billion-dollar company as a result of it.
Now, it may be that the contract in the filing is the actual contract between Zuckerberg and Ceglia, but I find it equally plausible that the contract is a forgery. Zuckerberg's faults have been highlighted here on numerous occasions, and Ceglia apparently was under investigation for alleged fraud with an unrelated business of his. I'm not sure who to believe, but the details of the case that have been made public makes it obvious that no one other than Ceglia and Zuckerberg probably know the truth, and neither has been able to make a convincing case thus far. I will grant that at this point Ceglia obviously has made a more convincing case than Facebook/Zuckerberg, but I'm sure that neither has tipped their entire hand at this point. The fact that Facebook is also going after other avenues (e.g. arguing that even if the contract was the one they both signed, that it isn't valid or the statute of limitations is up, etc.) makes me think that they don't have good evidence that the contract in the filing is a forgery, even if that is what they believe.
So, if they're going to argue that, then Zuckerberg should have kept a copy of his contract so that his lawyers could actually see it, and they would be smacking him hard up the back of the head for suggesting that they should make such a claim.
He was 18 at the time, and it was 7 years ago. I don't think it's unreasonable to think that he didn't keep good care of the contract, especially if it was only for the streetfax stuff.
Why isn't Zuckerberg claiming the contract is a forgery?
I think they're heading in that direction. Facebook has asked to see an original copy of the contract and Ceglia's lawyer told them to "come visit my office and I'll show it to you." Sounds fishy.
I looked into the merits of the case and it seems to me that Ceglia has a valid case and the defendants are using SCO-like defense tactics already at this early point.
Ceglia's attorney, Paul Argentieri, says bring it on. He says he offered to show Facebook's attorneys the original contract if they were willing to visit his offices in Hornell, N.Y. "They didn't take me up on my offer," Argentieri told DailyFinance. "They had three weeks to get ready to answer this very basic question. I was surprised by their hesitation."
Ceglia's attorney Argentieri says he will gladly explain why his client waited six- and-a-half years to file his lawsuit when the parties go to trial. He added that his client has plenty of other documents and canceled checks with Zuckerberg's signature.
That certainly makes Ceglia sound more like SCO than facebook here -- Ceglia's attorney is basically saying "we have lots of evidence but we won't show it to you."
The timing just doesn't make sense...Zuckerberg was alleged to have stolen the idea from the ConnectU guys in the fall of 2003, was't he? If this contract is actually legitimate, that would have shown that Zuckerberg had the idea long before he had contact with ConnectU -- or am I misremembering the time frame of the ConnectU stuff?
According to the book The Facebook Effect, Zuckerberg only owns 24% of the company at this point. Or at least that's the citation on that number on Facebook's wikipedia page.
Also shows you how absolutely useless the "due diligence" is that VCs perform. Zuckerberg probably never disclosed this old contract from the early days to any of his investors or they would have forced a settlement with this guy before things got this far. I've gotta think they are squirming just a bit right now.
Or maybe there was nothing to disclose, because Zuckerberg's only contract with Ceglia was for $1,000 to work on Streetfax, with no mention of facebook at any time. Would you want to give $10 to $20 million to any person who came up with a supposed contract with Zuckerberg? If facebook just started handing out money like this, you'd get more people coming out of the woodwork claiming similar contracts, when all they did is create a similar contract using the signature from the supposed contract currently available at http://i.cdn.turner.com/money/2010/07/13/technology/facebook_lawsuit/facebook_filing.pdf?
well maybe not now, when it's "worth" so much, but back then when he was just a thieving punk kid with no money the relationship was beneficial to him. Contracts don't need to be written by lawyers to be legally binding.
Zuckerberg had turned down an offer to work with Microsoft on MP3 playing software ("Synapse") for hundreds of thousands of dollars when he was still in high school. While I know he worked with the Ceglia guy on streetfax for something like $1,000, I doubt he would have sold half of an idea like facebook for merely $1,000, especially since Ceglia was offering absolutely nothing other than the $1,000 -- Zuckerberg would be doing all the work.
Note how they are not denying the existence of the agreement or the authenticity of the signatures on that agreement.
They aren't denying that Zuckerberg had a contract to work with Ceglia. What they are denying is that the contract included any mention of facebook. They believe that the actual contract was only for the StreetFax.com work, and not facebook.
It could well be a protest against electronic ballots or sabotage by a competitor. If someone wanted to demonstrate publicly beyond the shadow of doubt that they could control the outcome of an election by hacking the ballot machines, this was the ideal way to do it. They don't even need to own up to it. At the least, the ballot machine maker has a lot of explaining to do.
Exactly. Have a result like this in a relatively unimportant primary -- little damage is done if the result isn't overturned (the Republican will likely win regardless of the challenger), but it also calls into question the validity of electronic voting systems. I would imagine that there are quite a few slashdotters who would be in favor of such a thing, based on previous discussions about the voting machines.
What if the goal isn't to actually affect the outcome of an election, but merely to cause such a ludicrous result that everyone calls for an investigation and finds out how bad these electronic voting machines are? Think of it -- rig a relatively unimportant election (as many have said, the Republican is most likely going to win the seat regardless of the opponent(s)), and suddenly mainstream media is once again reporting on the fallibility of electronic voting machines.
I'd even be happy with 200ppi on a computer display.
200ppi would probably be sufficient, given the average distance of a monitor compared to your iPhone. If they are saying that 326ppi at 1 foot is sufficient, then 200ppi at about 2 feet would actually be better. I would think that most people have a 21" monitor at least 2 feet away from their eyes. At 200ppi, a 2160p monitor (3840 x 2160) would be about 22" diagonal, with only 4x the pixels of a normal 1080p display.
Now, what I want is a 21' monitor with the same dpis, instead of crappy 1080p resolutions, no matter the monitor size.
Assuming you meant a 21inch monitor (at 16x9 aspect ratio), that would be about 18.3" wide and 10.3" tall. At the iPhone's 326 PPI (note that PPI is not per square inch), that would result in a display running at approximately 6,000 x 3375, or over 20 million pixels. At 60 Hz, assuming 3 bytes per pixel, that's about 29Gbit/s uncompressed. HDMI tops out at 10.2Gbit/s, DVI at 7.92Gbit/s (dual-link), so you'd need to come up with some other way to push all those pixels to the display or have the display accept multiple HDMI/DVI inputs (the resolution would be comparable to a 3x3 setup of normal-resolution monitors, which is more than something like eyeFinity can handle at this juncture).
If you're talking about a 21 feet wide diagonal screen, then you're looking at 72,000 x 40,500. I think that would be a little excessive.
I would actually argue it is more of a "stupid tax" but I also feel that way about most forms of gambling.
Participating in a lottery is not necessarily a "stupid" proposition, depending on the circumstances. I'm also not talking about cases where the payout of a particular lottery is great enough that the expected value of a lottery ticket is greater than the cost of a ticket. I'm speaking more to the personal utility of a particular sum of money (let's call it f(x), where "x" represents a sum of money, and the result is the utility of that sum of money). For a particular person, that function is most likely not linear. In other words, f(2x) is not necessarily 2f(x). For many people, f(lotto payout)/f(cost of ticket) can be many times greater than (lotto payout)/(cost of ticket). Given a sufficient level of this, the expected utility value of playing the lotto may end up being larger than the cost of the ticket, even if the expected dollar value of playing is still less than the cost of a ticket.
I suppose that you could try and argue that people whose utility function of money is not linear are "stupid," but I would argue that not taking into account the utility a particular sum of money would give you would be the "stupid" move.
To further expand upon this point, let's take a reverse case. Suppose I offered you a $1,000 ticket with billion-to-one odds (10E9) that paid 1 quadrillion dollars (10E15) if you won (and you were only allowed to buy one ticket). The expected value of the ticket is $1,000,000, so the odds should indicate you would take that offer. I would imagine that most "smart" people, if given that opportunity, would not take it, because their expected utility value of playing would actually be lower than the utility value of $1,000.
A bit off topic, but how can one be unable to get a bank account? A ten second Google search shows that there are accounts in existence with no fee or minimum balance.
Well, there's two things at play -- people might either not want a bank account, or they can't get one. All bank accounts probably require some sort of proof of identity, which someone like an illegal alien or fugitive might not have access to (or wouldn't want to give). And, despite FDIC insurance, there are people who don't trust banks with their money.
Changing the font size or DPI settings in Windows wreaks havoc on many programs. Some mainstream applications handle it nicely, but a change to either setting destroys a number of industry applications that my clients use.
What really needs to happen is that the DPI of the higher-resolution monitors needs to be around twice that of the current set of monitors. For example, my current monitor is 24" at 1920x1200. If you can make a 24" at 3840x2400, you then need two things:
1) The OS needs to know that your DPI is twice that of "normal" standards.
2) The application needs to indicate whether it is DPI aware or not (by default, it is not, so older applications work correctly). This could even be on an individual UI element level. Anything that is not DPI aware just gets four pixels on your display per "application pixel".
Net result is that stuff that can handle the increased DPI is smoother, stuff that doesn't looks pretty much the same as it would on a 1920x1200 monitor. If the DPI is not an integer multiple of a "standard" DPI, you'll get the "blurriness" that you get when you run an LCD in a non-native resolution (except of course if you run in half the resolution of native).
And this theory is utter crap. An error rate of 1:10,000 (yes, that's the rate given about 200 complaints in 2M cars) is way too high to be caused by random electromagnetic interference. When was the last time you saw your processor kick out a wrong bit?
Ok, so maybe true interference isn't the culprit, but I've had a number of processors kick out a ton of wrong bits. It could be that my power supply is bad, a memory chip is bad, or I overclocked my processor too much. If you look at a computer as a whole, I'd agree that software accounts for the majority of issues I see, but I've also seen a number of hardware issues as well. I would guess (and it's really just a guess, I could be way off) that the software itself is less complex than the stuff running on my PC (I would hope so, at least, because fool-proof testing on my PC would be impossible), and that the software is tested a lot more rigorously than the vast majority of the software that ends up on my PC. I'm sure the hardware is also tested, but my guess is that the hardware isn't as thoroughly tested as the software, because you can test the software once and sign off on it -- it (should) be identical in every car. The hardware, on the other hand, is something that would have to be tested on each car.
Otherwise, it's just another set of computer scientists looking over a few million lines of code they didn't write, trying to find a defect that has supposedly manifest itself less than a few hundred times out of million of cars and probably billions of miles driven.
You're confusing "electronic" with "software." One possible theory is that interference (internal or external) is causing signals between parts to become corrupted. My understanding (having RTFA) is that they are focusing on the electrical engineering aspects of it. I would imagine that NASA, needing to design and test equipment in the harsh environment of space, is pretty darn good at exactly that.
Harvard students can't grasp simple concepts such as "puck goes in the net, not the opposing team pepband's trombone section".
They were probably tired of scoring so many goals against Cornell, so they wanted to change things up a bit. On a more serious note, where were you sitting that you got struck in the face? If it was at Harvard, the pep band sits off to the side and it would have to be a very random event to have a puck go off into the stands with such force that you couldn't dodge it. Or was this at Cornell or was it before they put nets up behind the goals?
Wait, so the Govt. could clone the kids, but they went through extra steps to steal the *real* kids, and make the clones die quickly? Did they make a video game that Uwe Boll can't ruin the script of when he makes the movie?
It's been a while since I read the books, but my recollection is that the cloning process was not very good and resulted in the early death, which is why they couldn't use the cloned kids in the program (otherwise they could just find the best and make an army of clones).
Pretty much. Apophis, at that distance, basically has the same gravitational pull as a can of soda at 10m away. Apophis would have to be about 230 meters away to have the same gravitational acceleration as the moon does.
Solar panels are great until they get dirty or worse damaged by micro-meteorites.
Simple solution, taken right from the nuclear reactor concept -- just put the solar panels behind a few feet of lead, that will surely keep all the dust and micro-meteorites out!
Hmm. Well, the paper's argument is like saying that, if the average number of bugs (across all software and methodologies) in N lines of code is X, then somebody's claim that they have written a piece of software with M bugs in Y lines of code, where M/Y << N/X is bogus.
This is patently ridiculous. If I write a relatively small piece of software where I have carried out a formal mathematical proof of the algorithms used in that software, I should obtain a much better bug ratio than the industry average, which includes work done by code monkeys working 90 hour work weeks.
To apply what the article is saying to your analogy, it would be refuting your "bug-free code" by the fact that the "formal mathematical proofs" you are using may in fact be flawed. So by basing your "proof" on the things that themselves might have bugs in them, then it's quite possible that your software has bugs.
A much better analogy using software would be the following:
Suppose you write some code that has a 99.9% chance of being bug-free. You could then state that this program has a 99.9% chance of being bug free. However, if you now use a compiler that has a 1% bug rate, you can no longer say that your compiled program is a 99.9% chance of being bug free. At best, you can say that it has a 99% chance of being bug free. In much the same way, the original calculations were done assuming certain things were 100% accurate. The point of this article is that those certain things that are assumed to be 100% accurate, when actually empirically examined, are only correct 99.99% of the time. So if your "axioms" are only 99.99% correct, then you cannot prove anything with those axioms to be more than 99.99% correct.
Slashdot Mirror
Funny since the PDF you linked to mentions "The Face Book". Page 11 under the "Entire Agreement" Section
Yes, I know the PDF I liked to includes that. My point is that FB doesn't think that's actually the agreement that Zuckerberg signed. Let's pose a hypothetical situation:
Ceglia posts an ad on craigslist asking for programming help on a new startup called StreetFax. Zuckerberg responds, and they work out a contract that looks similar to the one that was in the filing, except there was never any mention of "The Face Book", "The Page Book" or anything *other* than the StreetFax stuff. They both sign it, Zuckerberg gets his money, Ceglia gets the StreetFax stuff, and then they part ways sometime in 2003 and never speak again.
Fast forward to June of 2010, and suddenly Ceglia is providing a poor quality copy of a contract that looks similar to the one Zuckerberg had signed 7 years ago, but now includes stuff about Face Book that was never on the original contract, and is claiming majority ownership of a billion-dollar company as a result of it.
Now, it may be that the contract in the filing is the actual contract between Zuckerberg and Ceglia, but I find it equally plausible that the contract is a forgery. Zuckerberg's faults have been highlighted here on numerous occasions, and Ceglia apparently was under investigation for alleged fraud with an unrelated business of his. I'm not sure who to believe, but the details of the case that have been made public makes it obvious that no one other than Ceglia and Zuckerberg probably know the truth, and neither has been able to make a convincing case thus far. I will grant that at this point Ceglia obviously has made a more convincing case than Facebook/Zuckerberg, but I'm sure that neither has tipped their entire hand at this point. The fact that Facebook is also going after other avenues (e.g. arguing that even if the contract was the one they both signed, that it isn't valid or the statute of limitations is up, etc.) makes me think that they don't have good evidence that the contract in the filing is a forgery, even if that is what they believe.
So, if they're going to argue that, then Zuckerberg should have kept a copy of his contract so that his lawyers could actually see it, and they would be smacking him hard up the back of the head for suggesting that they should make such a claim.
He was 18 at the time, and it was 7 years ago. I don't think it's unreasonable to think that he didn't keep good care of the contract, especially if it was only for the streetfax stuff.
Why isn't Zuckerberg claiming the contract is a forgery?
I think they're heading in that direction. Facebook has asked to see an original copy of the contract and Ceglia's lawyer told them to "come visit my office and I'll show it to you." Sounds fishy.
I looked into the merits of the case and it seems to me that Ceglia has a valid case and the defendants are using SCO-like defense tactics already at this early point.
You obviously didn't look too hard. From dailyfinance.com:
Ceglia's attorney, Paul Argentieri, says bring it on. He says he offered to show Facebook's attorneys the original contract if they were willing to visit his offices in Hornell, N.Y. "They didn't take me up on my offer," Argentieri told DailyFinance. "They had three weeks to get ready to answer this very basic question. I was surprised by their hesitation."
Ceglia's attorney Argentieri says he will gladly explain why his client waited six- and-a-half years to file his lawsuit when the parties go to trial. He added that his client has plenty of other documents and canceled checks with Zuckerberg's signature.
That certainly makes Ceglia sound more like SCO than facebook here -- Ceglia's attorney is basically saying "we have lots of evidence but we won't show it to you."
The timing just doesn't make sense...Zuckerberg was alleged to have stolen the idea from the ConnectU guys in the fall of 2003, was't he? If this contract is actually legitimate, that would have shown that Zuckerberg had the idea long before he had contact with ConnectU -- or am I misremembering the time frame of the ConnectU stuff?
According to the book The Facebook Effect, Zuckerberg only owns 24% of the company at this point. Or at least that's the citation on that number on Facebook's wikipedia page.
Also shows you how absolutely useless the "due diligence" is that VCs perform. Zuckerberg probably never disclosed this old contract from the early days to any of his investors or they would have forced a settlement with this guy before things got this far. I've gotta think they are squirming just a bit right now.
Or maybe there was nothing to disclose, because Zuckerberg's only contract with Ceglia was for $1,000 to work on Streetfax, with no mention of facebook at any time. Would you want to give $10 to $20 million to any person who came up with a supposed contract with Zuckerberg? If facebook just started handing out money like this, you'd get more people coming out of the woodwork claiming similar contracts, when all they did is create a similar contract using the signature from the supposed contract currently available at http://i.cdn.turner.com/money/2010/07/13/technology/facebook_lawsuit/facebook_filing.pdf?
well maybe not now, when it's "worth" so much, but back then when he was just a thieving punk kid with no money the relationship was beneficial to him. Contracts don't need to be written by lawyers to be legally binding.
Zuckerberg had turned down an offer to work with Microsoft on MP3 playing software ("Synapse") for hundreds of thousands of dollars when he was still in high school. While I know he worked with the Ceglia guy on streetfax for something like $1,000, I doubt he would have sold half of an idea like facebook for merely $1,000, especially since Ceglia was offering absolutely nothing other than the $1,000 -- Zuckerberg would be doing all the work.
Note how they are not denying the existence of the agreement or the authenticity of the signatures on that agreement.
They aren't denying that Zuckerberg had a contract to work with Ceglia. What they are denying is that the contract included any mention of facebook. They believe that the actual contract was only for the StreetFax.com work, and not facebook.
The contract is included in the lawsuit, available here: http://i.cdn.turner.com/money/2010/07/13/technology/facebook_lawsuit/facebook_filing.pdf
It could well be a protest against electronic ballots or sabotage by a competitor. If someone wanted to demonstrate publicly beyond the shadow of doubt that they could control the outcome of an election by hacking the ballot machines, this was the ideal way to do it. They don't even need to own up to it. At the least, the ballot machine maker has a lot of explaining to do.
Exactly. Have a result like this in a relatively unimportant primary -- little damage is done if the result isn't overturned (the Republican will likely win regardless of the challenger), but it also calls into question the validity of electronic voting systems. I would imagine that there are quite a few slashdotters who would be in favor of such a thing, based on previous discussions about the voting machines.
so what would they have to gain by doing this?
What if the goal isn't to actually affect the outcome of an election, but merely to cause such a ludicrous result that everyone calls for an investigation and finds out how bad these electronic voting machines are? Think of it -- rig a relatively unimportant election (as many have said, the Republican is most likely going to win the seat regardless of the opponent(s)), and suddenly mainstream media is once again reporting on the fallibility of electronic voting machines.
I'd even be happy with 200ppi on a computer display.
200ppi would probably be sufficient, given the average distance of a monitor compared to your iPhone. If they are saying that 326ppi at 1 foot is sufficient, then 200ppi at about 2 feet would actually be better. I would think that most people have a 21" monitor at least 2 feet away from their eyes. At 200ppi, a 2160p monitor (3840 x 2160) would be about 22" diagonal, with only 4x the pixels of a normal 1080p display.
Now, what I want is a 21' monitor with the same dpis, instead of crappy 1080p resolutions, no matter the monitor size.
Assuming you meant a 21inch monitor (at 16x9 aspect ratio), that would be about 18.3" wide and 10.3" tall. At the iPhone's 326 PPI (note that PPI is not per square inch), that would result in a display running at approximately 6,000 x 3375, or over 20 million pixels. At 60 Hz, assuming 3 bytes per pixel, that's about 29Gbit/s uncompressed. HDMI tops out at 10.2Gbit/s, DVI at 7.92Gbit/s (dual-link), so you'd need to come up with some other way to push all those pixels to the display or have the display accept multiple HDMI/DVI inputs (the resolution would be comparable to a 3x3 setup of normal-resolution monitors, which is more than something like eyeFinity can handle at this juncture).
If you're talking about a 21 feet wide diagonal screen, then you're looking at 72,000 x 40,500. I think that would be a little excessive.
I would actually argue it is more of a "stupid tax" but I also feel that way about most forms of gambling.
Participating in a lottery is not necessarily a "stupid" proposition, depending on the circumstances. I'm also not talking about cases where the payout of a particular lottery is great enough that the expected value of a lottery ticket is greater than the cost of a ticket. I'm speaking more to the personal utility of a particular sum of money (let's call it f(x), where "x" represents a sum of money, and the result is the utility of that sum of money). For a particular person, that function is most likely not linear. In other words, f(2x) is not necessarily 2f(x). For many people, f(lotto payout)/f(cost of ticket) can be many times greater than (lotto payout)/(cost of ticket). Given a sufficient level of this, the expected utility value of playing the lotto may end up being larger than the cost of the ticket, even if the expected dollar value of playing is still less than the cost of a ticket.
I suppose that you could try and argue that people whose utility function of money is not linear are "stupid," but I would argue that not taking into account the utility a particular sum of money would give you would be the "stupid" move.
To further expand upon this point, let's take a reverse case. Suppose I offered you a $1,000 ticket with billion-to-one odds (10E9) that paid 1 quadrillion dollars (10E15) if you won (and you were only allowed to buy one ticket). The expected value of the ticket is $1,000,000, so the odds should indicate you would take that offer. I would imagine that most "smart" people, if given that opportunity, would not take it, because their expected utility value of playing would actually be lower than the utility value of $1,000.
A bit off topic, but how can one be unable to get a bank account? A ten second Google search shows that there are accounts in existence with no fee or minimum balance.
Well, there's two things at play -- people might either not want a bank account, or they can't get one. All bank accounts probably require some sort of proof of identity, which someone like an illegal alien or fugitive might not have access to (or wouldn't want to give). And, despite FDIC insurance, there are people who don't trust banks with their money.
It is mostly bad UI.
Changing the font size or DPI settings in Windows wreaks havoc on many programs. Some mainstream applications handle it nicely, but a change to either setting destroys a number of industry applications that my clients use.
What really needs to happen is that the DPI of the higher-resolution monitors needs to be around twice that of the current set of monitors. For example, my current monitor is 24" at 1920x1200. If you can make a 24" at 3840x2400, you then need two things:
1) The OS needs to know that your DPI is twice that of "normal" standards.
2) The application needs to indicate whether it is DPI aware or not (by default, it is not, so older applications work correctly). This could even be on an individual UI element level. Anything that is not DPI aware just gets four pixels on your display per "application pixel".
Net result is that stuff that can handle the increased DPI is smoother, stuff that doesn't looks pretty much the same as it would on a 1920x1200 monitor. If the DPI is not an integer multiple of a "standard" DPI, you'll get the "blurriness" that you get when you run an LCD in a non-native resolution (except of course if you run in half the resolution of native).
How do you sleep at night?
Most likely on a mattress stuffed with (hundred) dollar bills
And this theory is utter crap. An error rate of 1:10,000 (yes, that's the rate given about 200 complaints in 2M cars) is way too high to be caused by random electromagnetic interference. When was the last time you saw your processor kick out a wrong bit?
Ok, so maybe true interference isn't the culprit, but I've had a number of processors kick out a ton of wrong bits. It could be that my power supply is bad, a memory chip is bad, or I overclocked my processor too much. If you look at a computer as a whole, I'd agree that software accounts for the majority of issues I see, but I've also seen a number of hardware issues as well. I would guess (and it's really just a guess, I could be way off) that the software itself is less complex than the stuff running on my PC (I would hope so, at least, because fool-proof testing on my PC would be impossible), and that the software is tested a lot more rigorously than the vast majority of the software that ends up on my PC. I'm sure the hardware is also tested, but my guess is that the hardware isn't as thoroughly tested as the software, because you can test the software once and sign off on it -- it (should) be identical in every car. The hardware, on the other hand, is something that would have to be tested on each car.
Otherwise, it's just another set of computer scientists looking over a few million lines of code they didn't write, trying to find a defect that has supposedly manifest itself less than a few hundred times out of million of cars and probably billions of miles driven.
You're confusing "electronic" with "software." One possible theory is that interference (internal or external) is causing signals between parts to become corrupted. My understanding (having RTFA) is that they are focusing on the electrical engineering aspects of it. I would imagine that NASA, needing to design and test equipment in the harsh environment of space, is pretty darn good at exactly that.
Couldn't have put it more perfectly my self.
Harvard students can't grasp simple concepts such as "puck goes in the net, not the opposing team pepband's trombone section".
They were probably tired of scoring so many goals against Cornell, so they wanted to change things up a bit. On a more serious note, where were you sitting that you got struck in the face? If it was at Harvard, the pep band sits off to the side and it would have to be a very random event to have a puck go off into the stands with such force that you couldn't dodge it. Or was this at Cornell or was it before they put nets up behind the goals?
Wait, so the Govt. could clone the kids, but they went through extra steps to steal the *real* kids, and make the clones die quickly? Did they make a video game that Uwe Boll can't ruin the script of when he makes the movie?
It's been a while since I read the books, but my recollection is that the cloning process was not very good and resulted in the early death, which is why they couldn't use the cloned kids in the program (otherwise they could just find the best and make an army of clones).
Am I close to doing that right?
Pretty much. Apophis, at that distance, basically has the same gravitational pull as a can of soda at 10m away. Apophis would have to be about 230 meters away to have the same gravitational acceleration as the moon does.
Solar panels are great until they get dirty or worse damaged by micro-meteorites.
Simple solution, taken right from the nuclear reactor concept -- just put the solar panels behind a few feet of lead, that will surely keep all the dust and micro-meteorites out!
Hmm. Well, the paper's argument is like saying that, if the average number of bugs (across all software and methodologies) in N lines of code is X, then somebody's claim that they have written a piece of software with M bugs in Y lines of code, where M/Y << N/X is bogus.
This is patently ridiculous. If I write a relatively small piece of software where I have carried out a formal mathematical proof of the algorithms used in that software, I should obtain a much better bug ratio than the industry average, which includes work done by code monkeys working 90 hour work weeks.
To apply what the article is saying to your analogy, it would be refuting your "bug-free code" by the fact that the "formal mathematical proofs" you are using may in fact be flawed. So by basing your "proof" on the things that themselves might have bugs in them, then it's quite possible that your software has bugs.
A much better analogy using software would be the following:
Suppose you write some code that has a 99.9% chance of being bug-free. You could then state that this program has a 99.9% chance of being bug free. However, if you now use a compiler that has a 1% bug rate, you can no longer say that your compiled program is a 99.9% chance of being bug free. At best, you can say that it has a 99% chance of being bug free. In much the same way, the original calculations were done assuming certain things were 100% accurate. The point of this article is that those certain things that are assumed to be 100% accurate, when actually empirically examined, are only correct 99.99% of the time. So if your "axioms" are only 99.99% correct, then you cannot prove anything with those axioms to be more than 99.99% correct.