IN a stunning development, Osbourne computers announces the future development of the Osbourne2. Sales of of the just released Osboune1 plummet. Osbourne goes out of bussiness and the #2 is never released.
I'd like to get myself straight on this. But it's not clear that halftoning is correct. The three color print heads are distinct heads. I don't see any reason why the dots can't be superimposed. Can you please elaborate.
oopsie you're right! but that's still multi-gigabyte capacity. As for the other comments about it taking more than one dot to make an image pixel, they don't understand that I'm computing the maximum data capacity not the image reproduction capacity. And the calculation was supposed to be the ideal one not one considering imperfections.
Now if we were to assume that he could get an extra factor of 256 somewhere, like for example he had 48 Bit color depth, used 6 color inks, and could get 4 times as many pixels per inch using the best possible printers, then we're closing in on the right value.
And as you know newspapers get things wrong. Maybe that was for a two sided sheet of paper. Maybe he meant bits and they wrote bytes.
The point is the claim is not patently absurd like the blogs claim.
The blogs proclaiming this to be a scam seem pretty feeble reasoning. But it's not too hard to see if the numbers add up themselves. First let's suppose we have a very fine color printer and a very fine color scanner that can print at say 4096 DPI in RGB with 24 bits of color. And we'll consider an 8x11 sheet of paper: 4096*4096*256*256*256*8*11
this is 24 Million Giga Bits, or 3 Million Gigabytes. Now he's going to have to redundantly encode this in some error correcting way since paper and color resolution is imperfect. So assume he makes it 10,000 fold redundant with his circles and triangles. that's 300GB just like he claimes.
Okay now I'm replying to my own post. what I said was right. But the application is not for computer chips but for much hotter systems. Namely the application is for burining propane at 600 degrees C and converting that to electricity. In theory the themodynamic efficiency would be max of 50%. They claim that inpractice they might achieve 20 to 30%. "The result is a solid state energy conversion chip that can operate at temperatures of up to 600 degrees celcius and deliver absolute efficiencies in terms of how much heat energy is converted to electricity of between 20 and 30 percent."
Now 20 to 30% conversion of a stored chemical fuel to electricity RIGHT ON A MICRO CHIP without any mechinaical engine is great. Good energy density even if you are giving up 80% of the energy. The only trick is figuring out how to chill the backside. But if you are only looking for small amounts of power maybe ambient chilling or convection is not so bad. Maybe you could even burn a little more chemical enerfy to power a turbine to cool it off.
Anyhow the uses for this are not microchips but very hot systems. And that's what makes it different from conventional peltier coolers: it's compact, monlithic, and runs so hot it can get good efficiency.
T_low is probably at best above room tempertature or around 300 Kelvin. let's assume T_high is 57 degrees C or about 330 degrees kelvin. So that gives: Efficiency = 1-(T_low/T_high)
= 10%
So if they have no losses at all and it's a perfect heat engine they can recover 10% of the wasted energy as electricity. In reality I'd wager their losses will be 50% of may they can get back 5% of the heat energy.
It can't actually cause a noticalble dent in the chip temperature since that would reduce the temperature differential and it would stop working.
If you already have a high heat flux with a lot of cooling to hold the temperature at 57 degrees C then this won't change the amount out cooling you need. On the otherhand if you have a very precariously balanced system where reducing the heat flux 5% would make a very large change in the temperature it might conceivable be useful. I'm just haveing a hard time figuring out what the cicumstances that would lead to that are. Maybe some sort of unchilled embedded device that generates lots of power over a short interval.
As a peltier device were not restricted by this efficiency since energy is being supplied to act as a chiller. IN this case it could be a cooler. Just like any other peltier device.
I hate to say it but Cringely got it right and this article gets it wrong. Without net neutrality we move to spoke and wheel internet where the hubs are the high QOS cliques of the major carriers. all other paths joining nodes that are not in the intra-carrier cliques and thus getting first rank quality of service will be slow connections. As a result two things happen: the actual network capacity, compared to a peer-to-peer model goes down. and the number of players who can simultaneously be connected within one clique drops.
Now the providers like this. First, the guy with the biggest clique wins and it drives out the little guy competittion. Second, they don't care what your bandwidth is as long as they are the gate keeper and can charge you what it costs them plus a fixed profit. They have no strong incentive to build more bandwidth since as gate keepers their profit will be the same. It's not like there are suddenly be fewers internet users. As long as you can play some games you will be shelling out 49.99 per month--you wont decide well hey it's not fast enough so I wont use the internet at all. You'll still belly up. You might be willing to pay a premium for faster service, but unless all the other game players were willing to do so also then your speed limit in the game is not your connection but the connection to the other players on the slow links.
Now the way they can deliver better QOS to everyone is to maximally exploit all the interconnects they don't gate keep. Namley the the peer-to-peer connections that may span provider networks. If all those have high QOS there's more bandwidth for everyone. They just can't change you extra for it and since it allows competition and the small cliques can compete you are not slaved to one provider: you can move to the best value and still have good QOS. So there's incentive to the providers to provide faster and faster connetions at the lowest cost.
This sounds like horseshit. It's like something you would see in a factually absurd hollywood movie about a programmer uploading a virus into the power grid. How does this work in these games that someone is ever allowed to inject a code that can run on someone elses session? Why would they allow that. Spining rings appearing in my session from some one elses code and my computer runs the code if I touch them. Praise Tron. I assume there is some explanation for this but since I'm not a gamer I am without a clue.
This is a dupe from january. At the time it was predicted it would take 1 month before someone exploited it. The Sony DRM fiasco actually came after this was known. So it's entirely plausible that Sony actually did try to implement this because at the time they had not yet learned how bad agressive DRM was going to be for their bussiness
Silly statistics biased by observer error. Cleraly they did no weight their queries by the probability of a patient showing up with a given disease. If you are sick, then statistically, you have a cold in nearly all of cases of sickness. Therefore if, no matter what your symtoms were, I were to guess you had a cold, then I'd be able to correctly diagnose most patients. Instead they probably weighed all diseases uniformly.
Second, I would assume they always inputted the right symptoms and signs. But how many patients really display all the "right" signs and don't have some symptoms that are misleading. One of the hard problems in emergency medicine, since you don't have a case history, is knowing what symtoms are caused by the disease or injury. Is the patient's pulse fast or is it normally fast. Is the patient's blood pressure low or is it normallly low. Are they beligerent normally or might they have a concussion?
toss in some red herring symptoms and I'd bet these diagnoses become so promiscuous they are uselessly non-specific.
79 zune points costs 99 cents. and you have to buy them in increments of $5. jeezus! this is so diaboilical on many levels. First, MS gets to hold your float for $4.01 after you buy the first song. That's a lot of interest to MS over time. And then after you buy 5 songs then what? MS gets to keep the extra nickel you can't spend. So really songs cost $1.00 not $0.99 cents. MS pockets 1% extra. Finally, the detachment of points from dollars measn two things. First it measn your brain hurts when you are spending funny money with strange conversion rates that make it hard to figure on the fly. And second it means they can charge different groups different prices or raise prices. For example, everyone who buys zune points now may be paying $1.00 per song but if you buy the points next week it's say $1.50 a song. Or maybe you get a zune points discount if you buy a new computer with MS OS.
it all seems so deliberately complicated and multi-tiered like all of MS products. That can be a good bussiness strategy but it's not a user-freindly strategy. I predict people will prefer their music priced in way they can figure and don't have to work the angles to get the best prices.
Ummm think about it. If you pay $500 extra for Windows, did microsoft pay it or did you pay it? Same thing with car insurance. Microsoft is not profiting from the bot nets. They are profiting from skimping on security when they are the monopoly mono crop. I'd set higher standards for any monocrop maker. They are benefiting from the monopoly position in their price structure but with it they have more responsibility to be safe. They do not pay for the damage they do to the internet. So this fixes it. They could not face this tax on their products if they had the apropos level of security on the their products. Their call.
I was arguing that windows was the source of the problem and MS was profiting by not having to pay for their share of the problem. A more apropos example is car insurance. If you drive a statitically risky car or belong to a statistically risky group your car insurance is higher. I'm not taking microsoft I'm charging a fee for using windows just as the State forces you to pay car insurance to use your car.
thanks for the only meaningful response. Yeah $500 ws just for drama. But I think there would be no harm is starting low and ramping up till the problem is tolerable. So discovery otf the appropriate level of tax is self determining.
The more I think about this however the more I think it should be dome like car insurance. To get an IP addrress you shoulld pay a spam protection fee. Then you get citations for bad behaviour. running an open relay, 2 demerits. running windows 1 demerit, actually spamming 10 demerits. Then your spam protection fee goes up if you are a bad internet driver or engage in risky behaiour like running windows. And I would not pick on widows either. namely I'd define any OS that had a zombie/installed users base ratio that was high as risky OS. The ISPs would of course have to decide how they were going to determine what your spam protection fee was. But we have the same problem with car insurance.
To explain this further lets contrast this with an alternative implementation of the concept. Suppose instead of adding $500 to the price of all MS Windows OS (and I'm just pulling $500 out of my ass here to make it dramtic) we instead say it's a user responisbility. SO instead we let users forego the $500 tax as long as they post a bond of 10x the tax amount that they will forfiet if their computer becomes a spambot. They of course would not actually post the bond itself but instead would buy insurance.
Now in the end if this were a workable system, it should actually come out to be the same mostly since in the end the total amount collected as tax or forfietied has to be the same. that is insurance rate would turn out to be the same net cost to the average consumer as the tax. The difference is that careful users might decide to forego the insruance and would never have to pay and sloppy ones would pay for the rest. However in practice the bond idea is unworkbale. First people would cheat on it. It's impossible to enforce efficiently and would end up disenfrachinsing people for mistakes they could not afford to fix. It's analogous to the considerations that lead to no-fault car insurance systems. Sometimes just having everyone pay makes sense because it is easier to enforce.
Of course it would quicky occur that people would cheat and use bootleg tax free software. But this is not a problem per se. It just means that we would charge the tax at the source. MS would pay it directly not the retail store. If this made MS software probihitively expensive MS would be moved to solve the problem.
Let's put a $500 tax on all copies of windows OS. Wait! this is not flamebait. Here me out.
The tragedy of the commons is what occurs when there is no limit on use of public resource but iindividuals do not bear the consquence of abuse in a way that would make them modify their behaviour for the common good. The historic solution is to put a fee for admission that promotes optimal use. Now as we have all heard over and over that most propose e-mail stamp plans all fail for one reason or another. Indeed there's that ubiquitous and hilarious form letter someone always posts on slashdot whenever the latest unworkbale plan is proposed that exaplains why it won't work.
So my plan is not to have some micro payment scheme but to simply tax the origin of abuse directly. Windows Operating systems are essentially responsible for all Spam. Now if microsoft had put more effrot into securing their system then windows would have cost more to develop. So instead they are getting rich off of this since the costs of the consequences are not being borne by microsoft. Therefore there is needed a fee. The fee would be applied to cover the cost of rigorous anti-spam actions by ISPs or whomever was the appropriate cop. Alternatively it could have the effect of detering excessive monocropong of operating systems, like Windows, that makes it ripe for epidemics like this
Now before someone says well it's not microsoft's fault, their software is just as good as Linux, mac, amiga, Beos..., let me say that does not matter. Microsoft gets a market advantage and cost structure advantage by meing the mono-crop operating system. Therefore regardless of whether there security is comparabel to some other, they have a greater responsibility and a greater finaincial wherewithall to make their software be more secure. It is precisley fair to treat a monopoly with a different set of stnadards if that monopoly position is 1) the source of the problem 2) they are getting financial gain from being a monopoly.
So rather than flaming me, tell me why this is not a proper anlaysis of the problem and a possible approach to solving it. Yes it's radical. But according to earthlink I get 2000 spam messages a week. and according to this article 3/4 of the mail out there is spam. Radical solutions are called for.
I guess you have never gazed into the abyss of embrace and extend. You have say an internet browser that can display all the pages that another browser can plus some the other cannot. Which do you use?
Suppose MS got companies to give away free Zune music for reward points or something. Then you have the stack of zune tunes you need a zune player for. Now which is going to be your nextplayer?
Any Journal article comprised of 1% plagiarism would be subject to law suits, apologies and the journal would face ostracism. This is intellectual theft somehow made possible by the anonymity of the Wiki. We do tolerate this in less professional venues. For example, amateur reader comments are not subject to this kind of scrutiny. Comment sites like slashdot are protected from that sort of thing. But a formal identifiable entity that generates citable articles in itself, which has pervasive plagiarism at the 1% level needs to be shut down or it's citations fixed. This is a terrible day for the otherwise marvelous wikipedia concept. Deep thought is needed to figure out how to create some process of assured attribution. It's a shame. Even with the plagiarism Wikipedia is still informative. It's just that we can't become permissive about plagiarism even if it is for a good cause.
Open Voting Consortium which needs your donations of time or money to develop the open source implementation of this, and Populex, a commercial product both work exactly as you describe. But plugging all the loopholes in it is trickier than you might think. Go to OVC's site and read up.
Slashdot Mirror
IN a stunning development, Osbourne computers announces the future development of the Osbourne2. Sales of of the just released Osboune1 plummet. Osbourne goes out of bussiness and the #2 is never released.
I'd like to get myself straight on this. But it's not clear that halftoning is correct. The three color print heads are distinct heads. I don't see any reason why the dots can't be superimposed. Can you please elaborate.
oopsie you're right! but that's still multi-gigabyte capacity. As for the other comments about it taking more than one dot to make an image pixel, they don't understand that I'm computing the maximum data capacity not the image reproduction capacity. And the calculation was supposed to be the ideal one not one considering imperfections.
Now if we were to assume that he could get an extra factor of 256 somewhere, like for example he had 48 Bit color depth, used 6 color inks, and could get 4 times as many pixels per inch using the best possible printers, then we're closing in on the right value.
And as you know newspapers get things wrong. Maybe that was for a two sided sheet of paper. Maybe he meant bits and they wrote bytes.
The point is the claim is not patently absurd like the blogs claim.
The blogs proclaiming this to be a scam seem pretty feeble reasoning. But it's not too hard to see if the numbers add up themselves. First let's suppose we have a very fine color printer and a very fine color scanner that can print at say 4096 DPI in RGB with 24 bits of color. And we'll consider an 8x11 sheet of paper:
4096*4096*256*256*256*8*11
this is 24 Million Giga Bits, or 3 Million Gigabytes.
Now he's going to have to redundantly encode this in some error correcting way since paper and color resolution is imperfect. So assume he makes it 10,000 fold redundant with his circles and triangles. that's 300GB just like he claimes.
Okay now I'm replying to my own post. what I said was right. But the application is not for computer chips but for much hotter systems. Namely the application is for burining propane at 600 degrees C and converting that to electricity. In theory the themodynamic efficiency would be max of 50%. They claim that inpractice they might achieve 20 to 30%.
"The result is a solid state energy conversion chip that can operate at temperatures of up to 600 degrees celcius and deliver absolute efficiencies in terms of how much heat energy is converted to electricity of between 20 and 30 percent."
Now 20 to 30% conversion of a stored chemical fuel to electricity RIGHT ON A MICRO CHIP without any mechinaical engine is great. Good energy density even if you are giving up 80% of the energy. The only trick is figuring out how to chill the backside. But if you are only looking for small amounts of power maybe ambient chilling or convection is not so bad. Maybe you could even burn a little more chemical enerfy to power a turbine to cool it off.
Anyhow the uses for this are not microchips but very hot systems. And that's what makes it different from conventional peltier coolers: it's compact, monlithic, and runs so hot it can get good efficiency.
T_low is probably at best above room tempertature or around 300 Kelvin. let's assume T_high is 57 degrees C or about 330 degrees kelvin. So that gives:
Efficiency = 1-(T_low/T_high)
= 10%
So if they have no losses at all and it's a perfect heat engine they can recover 10% of the wasted energy as electricity. In reality I'd wager their losses will be 50% of may they can get back 5% of the heat energy.
It can't actually cause a noticalble dent in the chip temperature since that would reduce the temperature differential and it would stop working.
If you already have a high heat flux with a lot of cooling to hold the temperature at 57 degrees C then this won't change the amount out cooling you need. On the otherhand if you have a very precariously balanced system where reducing the heat flux 5% would make a very large change in the temperature it might conceivable be useful. I'm just haveing a hard time figuring out what the cicumstances that would lead to that are. Maybe some sort of unchilled embedded device that generates lots of power over a short interval.
As a peltier device were not restricted by this efficiency since energy is being supplied to act as a chiller. IN this case it could be a cooler. Just like any other peltier device.
So what's new here? Anyone have a clue/
I hate to say it but Cringely got it right and this article gets it wrong. Without net neutrality we move to spoke and wheel internet where the hubs are the high QOS cliques of the major carriers. all other paths joining nodes that are not in the intra-carrier cliques and thus getting first rank quality of service will be slow connections. As a result two things happen: the actual network capacity, compared to a peer-to-peer model goes down. and the number of players who can simultaneously be connected within one clique drops.
Now the providers like this. First, the guy with the biggest clique wins and it drives out the little guy competittion. Second, they don't care what your bandwidth is as long as they are the gate keeper and can charge you what it costs them plus a fixed profit. They have no strong incentive to build more bandwidth since as gate keepers their profit will be the same. It's not like there are suddenly be fewers internet users. As long as you can play some games you will be shelling out 49.99 per month--you wont decide well hey it's not fast enough so I wont use the internet at all. You'll still belly up. You might be willing to pay a premium for faster service, but unless all the other game players were willing to do so also then your speed limit in the game is not your connection but the connection to the other players on the slow links.
Now the way they can deliver better QOS to everyone is to maximally exploit all the interconnects they don't gate keep. Namley the the peer-to-peer connections that may span provider networks. If all those have high QOS there's more bandwidth for everyone. They just can't change you extra for it and since it allows competition and the small cliques can compete you are not slaved to one provider: you can move to the best value and still have good QOS. So there's incentive to the providers to provide faster and faster connetions at the lowest cost.
the article is exactly wrong
This sounds like horseshit. It's like something you would see in a factually absurd hollywood movie about a programmer uploading a virus into the power grid. How does this work in these games that someone is ever allowed to inject a code that can run on someone elses session? Why would they allow that. Spining rings appearing in my session from some one elses code and my computer runs the code if I touch them. Praise Tron. I assume there is some explanation for this but since I'm not a gamer I am without a clue.
This is a dupe from january. At the time it was predicted it would take 1 month before someone exploited it. The Sony DRM fiasco actually came after this was known. So it's entirely plausible that Sony actually did try to implement this because at the time they had not yet learned how bad agressive DRM was going to be for their bussiness
....general purpose voting machines.
at 200Km/hour would it not take less than 2 hours to reach outerspace?
Second, I would assume they always inputted the right symptoms and signs. But how many patients really display all the "right" signs and don't have some symptoms that are misleading. One of the hard problems in emergency medicine, since you don't have a case history, is knowing what symtoms are caused by the disease or injury. Is the patient's pulse fast or is it normally fast. Is the patient's blood pressure low or is it normallly low. Are they beligerent normally or might they have a concussion?
toss in some red herring symptoms and I'd bet these diagnoses become so promiscuous they are uselessly non-specific.
it all seems so deliberately complicated and multi-tiered like all of MS products. That can be a good bussiness strategy but it's not a user-freindly strategy. I predict people will prefer their music priced in way they can figure and don't have to work the angles to get the best prices.
Ummm think about it. If you pay $500 extra for Windows, did microsoft pay it or did you pay it? Same thing with car insurance. Microsoft is not profiting from the bot nets. They are profiting from skimping on security when they are the monopoly mono crop. I'd set higher standards for any monocrop maker. They are benefiting from the monopoly position in their price structure but with it they have more responsibility to be safe. They do not pay for the damage they do to the internet. So this fixes it. They could not face this tax on their products if they had the apropos level of security on the their products. Their call.
I was arguing that windows was the source of the problem and MS was profiting by not having to pay for their share of the problem. A more apropos example is car insurance. If you drive a statitically risky car or belong to a statistically risky group your car insurance is higher. I'm not taking microsoft I'm charging a fee for using windows just as the State forces you to pay car insurance to use your car.
The more I think about this however the more I think it should be dome like car insurance. To get an IP addrress you shoulld pay a spam protection fee. Then you get citations for bad behaviour. running an open relay, 2 demerits. running windows 1 demerit, actually spamming 10 demerits. Then your spam protection fee goes up if you are a bad internet driver or engage in risky behaiour like running windows. And I would not pick on widows either. namely I'd define any OS that had a zombie/installed users base ratio that was high as risky OS. The ISPs would of course have to decide how they were going to determine what your spam protection fee was. But we have the same problem with car insurance.
Now in the end if this were a workable system, it should actually come out to be the same mostly since in the end the total amount collected as tax or forfietied has to be the same. that is insurance rate would turn out to be the same net cost to the average consumer as the tax. The difference is that careful users might decide to forego the insruance and would never have to pay and sloppy ones would pay for the rest. However in practice the bond idea is unworkbale. First people would cheat on it. It's impossible to enforce efficiently and would end up disenfrachinsing people for mistakes they could not afford to fix. It's analogous to the considerations that lead to no-fault car insurance systems. Sometimes just having everyone pay makes sense because it is easier to enforce.
Of course it would quicky occur that people would cheat and use bootleg tax free software. But this is not a problem per se. It just means that we would charge the tax at the source. MS would pay it directly not the retail store. If this made MS software probihitively expensive MS would be moved to solve the problem.
The tragedy of the commons is what occurs when there is no limit on use of public resource but iindividuals do not bear the consquence of abuse in a way that would make them modify their behaviour for the common good. The historic solution is to put a fee for admission that promotes optimal use. Now as we have all heard over and over that most propose e-mail stamp plans all fail for one reason or another. Indeed there's that ubiquitous and hilarious form letter someone always posts on slashdot whenever the latest unworkbale plan is proposed that exaplains why it won't work.
So my plan is not to have some micro payment scheme but to simply tax the origin of abuse directly. Windows Operating systems are essentially responsible for all Spam. Now if microsoft had put more effrot into securing their system then windows would have cost more to develop. So instead they are getting rich off of this since the costs of the consequences are not being borne by microsoft. Therefore there is needed a fee. The fee would be applied to cover the cost of rigorous anti-spam actions by ISPs or whomever was the appropriate cop. Alternatively it could have the effect of detering excessive monocropong of operating systems, like Windows, that makes it ripe for epidemics like this
Now before someone says well it's not microsoft's fault, their software is just as good as Linux, mac, amiga, Beos..., let me say that does not matter. Microsoft gets a market advantage and cost structure advantage by meing the mono-crop operating system. Therefore regardless of whether there security is comparabel to some other, they have a greater responsibility and a greater finaincial wherewithall to make their software be more secure. It is precisley fair to treat a monopoly with a different set of stnadards if that monopoly position is 1) the source of the problem 2) they are getting financial gain from being a monopoly.
So rather than flaming me, tell me why this is not a proper anlaysis of the problem and a possible approach to solving it. Yes it's radical. But according to earthlink I get 2000 spam messages a week. and according to this article 3/4 of the mail out there is spam. Radical solutions are called for.
Suppose MS got companies to give away free Zune music for reward points or something. Then you have the stack of zune tunes you need a zune player for. Now which is going to be your nextplayer?
Just watch, soon Zune will play play-for-sure but other players won't play zune format music.
147/12000 = 1.2%
cute but this has nothing to do with plagiarism. Press releases are meant to be copied.
I ask because apparently You did not actually graduate high school yet if you can't understand what the difference is between cited and uncited text.
Any Journal article comprised of 1% plagiarism would be subject to law suits, apologies and the journal would face ostracism. This is intellectual theft somehow made possible by the anonymity of the Wiki. We do tolerate this in less professional venues. For example, amateur reader comments are not subject to this kind of scrutiny. Comment sites like slashdot are protected from that sort of thing. But a formal identifiable entity that generates citable articles in itself, which has pervasive plagiarism at the 1% level needs to be shut down or it's citations fixed. This is a terrible day for the otherwise marvelous wikipedia concept. Deep thought is needed to figure out how to create some process of assured attribution. It's a shame. Even with the plagiarism Wikipedia is still informative. It's just that we can't become permissive about plagiarism even if it is for a good cause.
Open Voting Consortium which needs your donations of time or money to develop the open source implementation of this, and Populex, a commercial product both work exactly as you describe. But plugging all the loopholes in it is trickier than you might think. Go to OVC's site and read up.