I'm fairly sure that MS has actually indulged in functionally identical practices, at least back when BeOS was regarded as a potential player.
With Intel, the deal was "Use our chips, and ours only, and there will be large 'marketing assistance' in it for you".
With Microsoft, the deal was "Use our OS, and ours only, and your per-unit OEM price will be competitive. Otherwise, you might find that your competitors are paying less for our product, or even end up buying licences at retail..."
The structuring was a bit different; but the net effect is pretty much identical. In both cases, though, my understanding is that these deals were tempered by a certain degree of realism. Back when Intel was joking around with P4s, and AMD was rocking the Opterons, you could get an Opteron server from Dell(particularly in 4 sockets and above, FSB vs. hypertransport was just a joke), just not a desktop or laptop. Similarly, while they would certainly hold your hand toward Windows Server whatever, you could get your servers bare or with Redhat licences. Even in the present day, though, the desktop/laptop linux offerings are pretty perfunctory.
Just to note: The model mentioned in the above article, the Latitude E6400, has a reputation for legendary suckage even in otherwise pretty satisfied dell shops. It is also on its 27th publicly released BIOS revision(released in the last week or so). Given that it is pretty much straight Intel silicon(unless you went with the Nvidia graphics option), I have no idea what could have gone so badly wrong with the design. The chassis is actually pretty nice; but the board Just Ain't Right, as they say.
If you applied a suitably capacitive material to the bottoms of the buttons, and made sure that your case design allowed them all to land on the screen when pressed, your fake buttons could actually be real buttons....
As for the screen: the easy way, but the one most likely to be detected, would be to just let part of the iDevice screen show through, and draw everything there. Quick, simple; but even the dullest of tenured lifers knows that a TI doesn't have a backlit color very-high-resolution display. The more difficult; but higher hacker cred, option would be to have an actual TI calculator display, driven from a microcontroller, with the iDevice communicating with the microcontroller either through the serial lines on the link port or through a crude-but-functional optical interface consisting of one or more regions on the screen that the iDevice blinks light/dark and the case has a photodetector immediately above. If the refresh rate is high enough, you could go with a single one, if not, you could increase the number, maybe 8 to send entire bytes at a time... That would be simplex; but the case could flash back at the camera to provide a simplex data line in the other direction, giving you full bidirectional communication, without a single bit of visible iDevice...
The fact that graphing calculators of quite modest specs and build still cost so much is a gooey blob of saliva in the face of idealist theories of competition.
However, the fact that graphing calculators are still of quite modest specs isn't.
The market for calculators is, basically, tests. They might also be used for homework and the occasional foray into programming; but they are basically purchased for tests. In a testing environment, wifi and 16GB of internal storage are not, shall we say, of much use in maintaining a fair testing environment.
Even if you make the "If the test is good, flashcards won't help you, and neither will notes stored on a calculator/iPod/whatever" argument(which is arguably a lot truer at higher levels), that still doesn't address the issue of network connected devices.
Imagine the following: iPod touch/iPhone with camera, internet connection, some sort of web conferencing software. Pay 29.95 at the paypal portal and, for the duration of the test if you get stuck on a problem, take a picture of it, and a suitably educated person in India solves it and sends back an image of the solution. Win/win(sort of). The cheater can get past even "mere facts won't save you" questions, and someone in a lower cost of living country makes comparatively good money solving easy problems in their area of expertise. The test, of course, becomes useless.
Intentionally limited devices for pedagogical purposes are eminently sensible. It's just that it should be pretty simple to stamp out a TI-83(or 89, the hardware doesn't exactly differ wildly) for absolute peanuts, not $100 a pop.
Unless you have a pretty high resolution screen(which, given present economies of scale, is probably substantially more expensive than a color one), color is a perfectly good way to distinguish multiple lines, data points from one run vs. those of another, and so on and so forth. If you have high resolution, you can get away with crosshatching and using different symbols and things; but that just turns into pixel soup on a lower resolution device.
As for usability, I'm assuming that, if only because the ADA could otherwise torpedo their chance of being purchased by a single public school district, they'll have a "don't use red/green for important distinctions" mode available at least optionally, if not by default.
I suspect that their effort will be largely stillborne in the US, at least. Nobody buys TI because of their technological superiority(which isn't) or low prices(which aren't); but because they have a more or less self-perpetuating hold on being "standard". Some(horribly degenerate) textbooks even have step by step pictorial instructions consisting of sequences of TI-83 keys to press. Would you like fries with that factorization?
Given that color screens aren't necessary(though technological progress means that $130 calculators should have nice screens, or today's calculators should sell for ~$20), I doubt that this will help much, and anybody who is likely to be doing lots of stuff where color would be useful(graphs, plots, figures, etc.) is probably using a computer algebra system anyway, so that won't sell many calculators.
That is correct, Intel has done nothing to cripple PCIe on their parts(except some of the newer Atoms which, whether as a cost saving measure or to keep people from attaching capable GPUs only have a few external lanes available), so any discrete GPU will work just fine.
Trouble is, on a board the size of the Macbook Air's, and with a battery to match, a discrete graphics chip(plus some RAM for it) is going to be a real layout/power/thermal trick.
By using the Nvidia chipset, with integrated GPU, Apple got a (low end) Nvidia GPU along with the usual northbridge/southbridge stuff in almost the same space and power envelope as the Intel chipset would have occupied. If them move from Core2 to i3, 5, or 7, they will benefit some from Intel's consolidation of some chipset functions onto the CPU; but they will be forced to either go with Intel's integrated graphics, which don't suit their purposes(especially since the Air will likely be restricted to a lower clocked part by thermal/power considerations, and Intel integrated graphics performance varies according to the performance of the core it is integrated with), or dedicate board space and power/thermal headroom that they probably don't have to a discrete GPU.
In larger laptops, it isn't a huge deal, and the advantages of the i5s and i7s over the Core2s basically sell the new platform; but the Air just doesn't have much internal space.
I don't know of any reasonably standardized retail offering, probably because of the rarity of consumer-level demand, and the fact that the service would cost a factor of ten what the next generation of the device would cost(except, as here, where there was no next generation...); but circuit board rework is an established market(for more complex/expensive boards, reworking ones that are slightly defective after automated production is cost effective, and for prototyping/debugging/preproduction it can be an essential part of the process), for which all sorts of equipment vendors exist, and BGA rework is just a subset of that, so it is quite probable that there exists an engineering or prototyping house that would be happy to rework your board, for the right money.
Probably not a "Here is the web store, supply credit card number and we'll send you a box, send in your device and it'll be back in a week", ColourWare style operation; you'd likely have to actually call somebody and talk it out and you probably wouldn't like the price; but board rework isn't secret lore or anything, just a part of the production process that you don't usually see.
On the actual forum threads, they were discussing locations to steal +5v power for the new USB peripherals. Apparently there was some Japan-only daughtercard offering whose connector cable was still present on international models that had a line, as well as some motherboard locations. That only solves the power problem, of course, not the "Windows sees an unpowered hub with 2 amps worth of devices hanging off of it" problem; but a substantial percentage of cheap 'n cheerful unpowered hubs lie horribly about their capabilities, including generally reporting themselves to the OS as being powered hubs.
On the minus side, this is all kinds of horribly spec-violating, and can turn what should be a trivial debugging exercise into a mess of hardware swapping; but it certainly does reduce the number of visible issues(which is presumably why the bottom-feeder market does it) and comes in quite handy if your crazy hacking plan involves a means of powering the devices independently...
I don't know what the sales figures are(or even if Steve cares, he took the iPod mini out back and shot it when it was a literal best-seller and flash RAM was still pretty pricey just so he could replace it with something smaller...); but I'd be unsurprised if its the end of the line for the MacBook Air.
Intel's newer processors don't work, by design, with Nvidia's chipsets, and Apple is really hyping the GPU acceleration for their OSX-running products. This means that there really aren't any huge upgrade options easily available. The Core2 is still pretty decent; but it isn't new and Intel isn't making them any faster now than they were a while ago. Waiting for Nvidia and Intel to achieve legal togetherness might take a while. Cramming a discrete GPU onto a board and battery that size would be a fair challenge. RAM is cheaper and denser now, so you could certainly bump the onboard RAM a bit; but that'd be quite the underwhelming upgrade.
Perhaps more importantly, the thinnest CEO in the industry now has the iPad, and has shown a longstanding trend of disregarding what people say they want. Something based on an upclocked smartphone board is always going to be thinner and lighter and longer running than all but the most heroic novelty PC designs, and(for those not on the iDevice bandwagon) the full Macbook Pro is only slightly thicker and heavier than the Air is, visually deceptive tapering aside, and is more featureful, more powerful, and cheaper.
It is certainly possible that Apple will do some sort of update to the Air; but it would also be plausible if it turned out that they consider the Air to have been replaced by the iPad. The Air was always a bit underspecced and overpriced for running serious Mac applications, and there is now a lighter and cheaper option for running the email and web and light productivity/gaming stuff. There is also the incrementally heavier; but much more capable and somewhat cheaper Pro, for those who require the full Mac capability.
Were this the PC market, where there are numerous vendors, and most vendors have more chassis variations than Apple has product lines, this wouldn't be an issue. This being Apple, though, getting squeezed like that might well be a death sentence.
Sociopaths are quite uncommon(I think the number hovers a bit under 1%, possibly tossing in a few hardline narcissists and 'serious bad news not otherwise specified' types); but the trouble is that almost everyone who isn't a candidate for sainthood has a sense of empathy calibrated pretty well for the size of the primate groups that we lived in 100,000 years ago. In a world of 6.5 billion, global trade routes, and a nest of externalities, that doesn't help nearly as much as it might.
This 'moral myopia' is extraordinarily useful, if we felt everybody's pain the way we feel that of close friends and kin we'd basically just spend all our time rocking back and forth in the fetal position, sobbing; but it also means that we can, in effect, act with impressive callousness if a modicum of intermediary structure distances us from what is happening.
Obviously people will abuse power given the chance; but don't the fools know that you are supposed to lie about your motives until you have the power?
And this is why sheep are harmless, wolves go through a brief period of dangerousness before being neutralized by the cops, and wolves in shepherd's are genuinely dangerous.
Well, we'll just have to hope that their competitors will implement the technology; because the odds of Alcatel doing a proper job are pretty much zero....
If everyone had one fire per property per occupancy, that would work just fine. As it is, the rate is far lower, so you'd need to recover much more than back fees(since the probability of a fire during an individual's occupancy of the property is less than 1, their total back fees will, if the insurance is priced properly, be a fraction of the cost of fighting a fire, corresponding to approximately to the fraction representing their risk). You would essentially have to collect a retail price, since the probability of a fire is 1 when the house is on fire.
I glossed over them not because they are insignificant, I find ethics to be both extremely important in practice and quite interesting as a branch of philosophy; but because I didn't want to focus on them in this context.
Ethical/moral debates in large online fora are generally an exercise in pain and virtual shouting. All smoke, no light, extremely frustrating. I also felt that I had nothing novel or interesting to contribute to the ethical side, which had already settled down into more or less two-position shouting by the time I got here today.
By contrast, I thought that the problem of externalities had not been adequately considered, so I focused on that without letting some half-baked attempt at rehashing a moral position cloud the point.
Now, personally, I find the unempathic behavior and ideology that extreme market orientation can reward as "rational" to be rather creepy, and actually not even as efficient as it alleges itself to be(the literature, both theoretical and empirical, on the value of high levels of social trust is fascinating stuff) and I would have much preferred to see the firefighters save the guy's house(with the proviso that I would also expect the guy to act similarly "irrationally" in helping others out if they needed it...)
Reciprocal altruism is both pleasant and surprisingly efficient, as long as levels of value-rational parasitism don't get too high, at which point things can get ugly.
I was a bit unclear there: As with health insurance, nobody(excluding a statistically insignificant number of people who are mentally questionable) will increase their direct risk of fire for that purpose. Even with firefighters, and full insurance, a house fire is a huge pain in the ass and a potential for loss of life.
However, increasing your property's risk of fire spread is a much more reasonable strategy. A few minor landscaping changes, a shrub here, a woodpile there, can control whether your yard is a firebreak or a fuel source, and in what directions. It would be dickish; but not necessarily illogical, to arrange your yard such that people with more to lose feel the need to pay for protection that also includes you.
To make a salient health analogy: in a great many contexts, STD treatment and prophilaxis are heavily subsidized or free, because these diseases are highly contagious and the threat encourages public support for taxpayer funded measures that reduce the ambient threat. By contrast, treatment for noncontagious conditions is, if subsidized at all, done indirectly as some insurance scheme, or on humanitarian grounds.
People don't get to chose the contagiousness of the disease that they have; but the net result is that people with contagious diseases do, in fact, motivate the population around them to subsidize treatment(or, in less humanitarian cases, involuntary confinement on public health grounds). People with less contagious, or noncontagious, diseases do not attract subsidies to the same level. Nobody wants to be sick; but contagious sick people get better subsidies than noncontagious ones do.
The unpleasant complicating factor is that his neighbor, who did pay, got inferior service because the firefighters let the fire grow and spread, rather than suppressing it. Now the guy who did pay the fee gets property damage. You can make moral arguments, if you wish, about saving the guy who didn't pay up; but the guy who did pay up has solid monetary reasons to be upset.
One of the reasons that, historically, certain private fire departments were suppressed, and replaced by tax-funded universal ones was that they were, shall we say, doing a little more than hoping that somebody would have a fire in order to justify their investment...
You would want to set the price ahead of time, based on a reasonable estimate of actual cost+convenience fee, just to avoid sleazy extortion setups; but it would be nice if there were a real retail cost in addition to an insurance option.
For most, risk pooling is a great function of insurance, and makes insurance worth it; but situations where the insured market squeezes out the retail market can get a bit dysfunctional(just look at the healthcare market, where the price paid by insurers is often wildly different from the price on the invoice, which is largely a fiction for anything much more expensive than basic drugs) .
The tricky part here is the externalities(as usual, externalities are a bitch).
So, the Cranicks don't pay, they don't get service. Simple enough. There might be some ethical objections; but the economics line up just fine.
Similarly, letting them pay $75 at time of use is a no-go. Fighting a fire costs way more than $75. That's an insurance price, not a retail service price. If you allow people to buy insurance after they need it, you either go bankrupt or the cost of insurance ends up equaling the retail cost of service. You then lose the risk-pooling function of insurance. Now, for things as potentially valuable as houses and their contents, it would be sensible to have an actual retail price(set ahead of time, and publically known, to prevent extortion) that an uninsured person could pay to save a burning building, there are probably a fair few situations where the price of fighting the fire is lower than the cost of replacing the structure, so being able to pay a retail cost of approximately actual cost+service fee would be sensible for both householder and firefighting company.
However, here is where things get unpleasant: Because the Cranicks didn't pay, the firefighters allowed the fire to burn merrily, growing and spreading until it hit somebody who had paid. Now, since the paying householder's property is on fire, they likely suffered some thousands or 10s of thousands in direct combustion, smoke, and water damage. They paid, and they got shitty service. Had the firefighters used the Cranicks property to fight and stop the fire, they could have saved their customers from any damage, and done a much better job of serving them, the ones who actually paid.
Of course, if it becomes known that firefighters will fight fires around an insured property, the obvious strategy is for property owners to club together, buy insurance for 1 plot and get insurance for all for only $75/n. The fire department couldn't support itself on that. If they tried to offer two tiers, a $75 "Fires fought on your property only" and a more expensive "Fires that threaten your property fought", then this creates a perverse incentive: If I live next to a wealthy looking neighbor, I can get him to buy my fire insurance for me just by making my property more dangerous to his. Don't want to encourage that.
This is why firefighting, like certain public health measures, is very hard to elegantly force into a market model unless you are so far in the sticks that each man really is an island. Fires spread, just like diseases. Whether or not the firefighters come to my neighbor's aid matters to me(aside from any debatable moral stuff); because the raging inferno that is his burning house just needs the wind to shift for my house to be next. Even if I've paid my fee, having thousands in water damage from the firefighters, plus smoke and any combustion that occurs before they get there isn't really satisfying. I'd really rather have them fight the fire where it starts, and never have to suffer it myself, rather than insist that everyone pay, and let pockets of fire spread until they endanger me. Same way, even if I don't give a fuck about the life of the guy making my sandwich at the deli, and I don't care how poor he is, I sure do care about what immunizations he has, and whether he can take sick days; because his germs are getting into my food supply.
That is the real complexity of this story, in my opinion. There are some moral questions, but those are debatable, and there really should be a retail price set; but that is a bookkeeping matter; but if I were the insured householder I'd be absolutely livid about this. I paid my dues, and I get lousy service because they are trying to make a point? You could have completely protected my property; but chose to let a nearby building become a danger to it, when I pay you to protect my property? WTF?
The correct solution is pressure sensors on the wheel and IR eye tracking. If the driver shows inadequate attention given the speed, level of swerving, and measured conditions, if available, a small explosive charge propels their body into the steering wheel/windshield at a velocity directly proportional to the one that a bystander would experience if hit by them...
The dickwaving over who has the biggest supercomputer seems largely like hype stirred up to enhance IBM's shareholder value; but if you are going to make a dubiously sensible investment in expensive toys, doing it with borrowed money is, in some ways, preferable to doing it with real money.
The huge debts that sovereign nations tend to rack up trigger the same moral instincts that petty consumer debt does; but it isn't at all clear that they work anything like the same way, economically.
China: "Dear US, we are cashing in the giant pile of debt you owe us."
US: "Shucks, China, it looks like we spent all our money on increasingly elaborate pyramid schemes and shitty exurbs that nobody wants. Anyway, thanks for all the free stuff over the years, and I hope you don't find the sudden transition from high-employment export economy to moderate-unemployment internal economy too jarring... TTLGTG!"
Let me translate IBM's statement into a more sincere, less carefully spun, form, and you'll see why an uncaring profit-maximizing multinational is wrapping itself in the flag:
PR: "But now, an IBM representative has said to a Washington, DC forum, 'You have sovereign nations making material investments of a tremendous magnitude to basically eat our lunch, eat our collective lunch.'"
Translation: "But now, an IBM lobbyist has said to a Washington, DC forum 'Other countries are doling out sweetheart contracts to manufacturers and designers of expensive computers. Give us a giant pile of money or the chinks win."
The ionizing portion of sunlight is pretty slim. A little UV and not much else. Certainly not zero potential for cellular damage(but neither is simple cell division in a dark box entirely free of mutagens. Entropy just gets you sometimes)...
Sunlight is powerful in a zergling rush sort of way; but the energy levels of each of the various photons are pretty low.
Slashdot Mirror
I'm fairly sure that MS has actually indulged in functionally identical practices, at least back when BeOS was regarded as a potential player.
With Intel, the deal was "Use our chips, and ours only, and there will be large 'marketing assistance' in it for you".
With Microsoft, the deal was "Use our OS, and ours only, and your per-unit OEM price will be competitive. Otherwise, you might find that your competitors are paying less for our product, or even end up buying licences at retail..."
The structuring was a bit different; but the net effect is pretty much identical. In both cases, though, my understanding is that these deals were tempered by a certain degree of realism. Back when Intel was joking around with P4s, and AMD was rocking the Opterons, you could get an Opteron server from Dell(particularly in 4 sockets and above, FSB vs. hypertransport was just a joke), just not a desktop or laptop. Similarly, while they would certainly hold your hand toward Windows Server whatever, you could get your servers bare or with Redhat licences. Even in the present day, though, the desktop/laptop linux offerings are pretty perfunctory.
It's just like they told you when you were a kid: "[blue collar] crime does not pay".
Just to note: The model mentioned in the above article, the Latitude E6400, has a reputation for legendary suckage even in otherwise pretty satisfied dell shops. It is also on its 27th publicly released BIOS revision(released in the last week or so). Given that it is pretty much straight Intel silicon(unless you went with the Nvidia graphics option), I have no idea what could have gone so badly wrong with the design. The chassis is actually pretty nice; but the board Just Ain't Right, as they say.
If you applied a suitably capacitive material to the bottoms of the buttons, and made sure that your case design allowed them all to land on the screen when pressed, your fake buttons could actually be real buttons....
As for the screen: the easy way, but the one most likely to be detected, would be to just let part of the iDevice screen show through, and draw everything there. Quick, simple; but even the dullest of tenured lifers knows that a TI doesn't have a backlit color very-high-resolution display. The more difficult; but higher hacker cred, option would be to have an actual TI calculator display, driven from a microcontroller, with the iDevice communicating with the microcontroller either through the serial lines on the link port or through a crude-but-functional optical interface consisting of one or more regions on the screen that the iDevice blinks light/dark and the case has a photodetector immediately above. If the refresh rate is high enough, you could go with a single one, if not, you could increase the number, maybe 8 to send entire bytes at a time... That would be simplex; but the case could flash back at the camera to provide a simplex data line in the other direction, giving you full bidirectional communication, without a single bit of visible iDevice...
Hackaday here I come!
Oh, that is cold, especially when "reasonable accommodation" is so damn cheap for that particular condition.
The fact that graphing calculators of quite modest specs and build still cost so much is a gooey blob of saliva in the face of idealist theories of competition.
However, the fact that graphing calculators are still of quite modest specs isn't.
The market for calculators is, basically, tests. They might also be used for homework and the occasional foray into programming; but they are basically purchased for tests. In a testing environment, wifi and 16GB of internal storage are not, shall we say, of much use in maintaining a fair testing environment.
Even if you make the "If the test is good, flashcards won't help you, and neither will notes stored on a calculator/iPod/whatever" argument(which is arguably a lot truer at higher levels), that still doesn't address the issue of network connected devices.
Imagine the following: iPod touch/iPhone with camera, internet connection, some sort of web conferencing software. Pay 29.95 at the paypal portal and, for the duration of the test if you get stuck on a problem, take a picture of it, and a suitably educated person in India solves it and sends back an image of the solution. Win/win(sort of). The cheater can get past even "mere facts won't save you" questions, and someone in a lower cost of living country makes comparatively good money solving easy problems in their area of expertise. The test, of course, becomes useless.
Intentionally limited devices for pedagogical purposes are eminently sensible. It's just that it should be pretty simple to stamp out a TI-83(or 89, the hardware doesn't exactly differ wildly) for absolute peanuts, not $100 a pop.
Unless you have a pretty high resolution screen(which, given present economies of scale, is probably substantially more expensive than a color one), color is a perfectly good way to distinguish multiple lines, data points from one run vs. those of another, and so on and so forth. If you have high resolution, you can get away with crosshatching and using different symbols and things; but that just turns into pixel soup on a lower resolution device.
As for usability, I'm assuming that, if only because the ADA could otherwise torpedo their chance of being purchased by a single public school district, they'll have a "don't use red/green for important distinctions" mode available at least optionally, if not by default.
I suspect that their effort will be largely stillborne in the US, at least. Nobody buys TI because of their technological superiority(which isn't) or low prices(which aren't); but because they have a more or less self-perpetuating hold on being "standard". Some(horribly degenerate) textbooks even have step by step pictorial instructions consisting of sequences of TI-83 keys to press. Would you like fries with that factorization?
Given that color screens aren't necessary(though technological progress means that $130 calculators should have nice screens, or today's calculators should sell for ~$20), I doubt that this will help much, and anybody who is likely to be doing lots of stuff where color would be useful(graphs, plots, figures, etc.) is probably using a computer algebra system anyway, so that won't sell many calculators.
That is correct, Intel has done nothing to cripple PCIe on their parts(except some of the newer Atoms which, whether as a cost saving measure or to keep people from attaching capable GPUs only have a few external lanes available), so any discrete GPU will work just fine.
Trouble is, on a board the size of the Macbook Air's, and with a battery to match, a discrete graphics chip(plus some RAM for it) is going to be a real layout/power/thermal trick.
By using the Nvidia chipset, with integrated GPU, Apple got a (low end) Nvidia GPU along with the usual northbridge/southbridge stuff in almost the same space and power envelope as the Intel chipset would have occupied. If them move from Core2 to i3, 5, or 7, they will benefit some from Intel's consolidation of some chipset functions onto the CPU; but they will be forced to either go with Intel's integrated graphics, which don't suit their purposes(especially since the Air will likely be restricted to a lower clocked part by thermal/power considerations, and Intel integrated graphics performance varies according to the performance of the core it is integrated with), or dedicate board space and power/thermal headroom that they probably don't have to a discrete GPU.
In larger laptops, it isn't a huge deal, and the advantages of the i5s and i7s over the Core2s basically sell the new platform; but the Air just doesn't have much internal space.
I don't know of any reasonably standardized retail offering, probably because of the rarity of consumer-level demand, and the fact that the service would cost a factor of ten what the next generation of the device would cost(except, as here, where there was no next generation...); but circuit board rework is an established market(for more complex/expensive boards, reworking ones that are slightly defective after automated production is cost effective, and for prototyping/debugging/preproduction it can be an essential part of the process), for which all sorts of equipment vendors exist, and BGA rework is just a subset of that, so it is quite probable that there exists an engineering or prototyping house that would be happy to rework your board, for the right money.
Probably not a "Here is the web store, supply credit card number and we'll send you a box, send in your device and it'll be back in a week", ColourWare style operation; you'd likely have to actually call somebody and talk it out and you probably wouldn't like the price; but board rework isn't secret lore or anything, just a part of the production process that you don't usually see.
On the actual forum threads, they were discussing locations to steal +5v power for the new USB peripherals. Apparently there was some Japan-only daughtercard offering whose connector cable was still present on international models that had a line, as well as some motherboard locations. That only solves the power problem, of course, not the "Windows sees an unpowered hub with 2 amps worth of devices hanging off of it" problem; but a substantial percentage of cheap 'n cheerful unpowered hubs lie horribly about their capabilities, including generally reporting themselves to the OS as being powered hubs.
On the minus side, this is all kinds of horribly spec-violating, and can turn what should be a trivial debugging exercise into a mess of hardware swapping; but it certainly does reduce the number of visible issues(which is presumably why the bottom-feeder market does it) and comes in quite handy if your crazy hacking plan involves a means of powering the devices independently...
I don't know what the sales figures are(or even if Steve cares, he took the iPod mini out back and shot it when it was a literal best-seller and flash RAM was still pretty pricey just so he could replace it with something smaller...); but I'd be unsurprised if its the end of the line for the MacBook Air.
Intel's newer processors don't work, by design, with Nvidia's chipsets, and Apple is really hyping the GPU acceleration for their OSX-running products. This means that there really aren't any huge upgrade options easily available. The Core2 is still pretty decent; but it isn't new and Intel isn't making them any faster now than they were a while ago. Waiting for Nvidia and Intel to achieve legal togetherness might take a while. Cramming a discrete GPU onto a board and battery that size would be a fair challenge. RAM is cheaper and denser now, so you could certainly bump the onboard RAM a bit; but that'd be quite the underwhelming upgrade.
Perhaps more importantly, the thinnest CEO in the industry now has the iPad, and has shown a longstanding trend of disregarding what people say they want. Something based on an upclocked smartphone board is always going to be thinner and lighter and longer running than all but the most heroic novelty PC designs, and(for those not on the iDevice bandwagon) the full Macbook Pro is only slightly thicker and heavier than the Air is, visually deceptive tapering aside, and is more featureful, more powerful, and cheaper.
It is certainly possible that Apple will do some sort of update to the Air; but it would also be plausible if it turned out that they consider the Air to have been replaced by the iPad. The Air was always a bit underspecced and overpriced for running serious Mac applications, and there is now a lighter and cheaper option for running the email and web and light productivity/gaming stuff. There is also the incrementally heavier; but much more capable and somewhat cheaper Pro, for those who require the full Mac capability.
Were this the PC market, where there are numerous vendors, and most vendors have more chassis variations than Apple has product lines, this wouldn't be an issue. This being Apple, though, getting squeezed like that might well be a death sentence.
Sociopaths are quite uncommon(I think the number hovers a bit under 1%, possibly tossing in a few hardline narcissists and 'serious bad news not otherwise specified' types); but the trouble is that almost everyone who isn't a candidate for sainthood has a sense of empathy calibrated pretty well for the size of the primate groups that we lived in 100,000 years ago. In a world of 6.5 billion, global trade routes, and a nest of externalities, that doesn't help nearly as much as it might.
This 'moral myopia' is extraordinarily useful, if we felt everybody's pain the way we feel that of close friends and kin we'd basically just spend all our time rocking back and forth in the fetal position, sobbing; but it also means that we can, in effect, act with impressive callousness if a modicum of intermediary structure distances us from what is happening.
Obviously people will abuse power given the chance; but don't the fools know that you are supposed to lie about your motives until you have the power?
And this is why sheep are harmless, wolves go through a brief period of dangerousness before being neutralized by the cops, and wolves in shepherd's are genuinely dangerous.
Well, we'll just have to hope that their competitors will implement the technology; because the odds of Alcatel doing a proper job are pretty much zero....
If everyone had one fire per property per occupancy, that would work just fine. As it is, the rate is far lower, so you'd need to recover much more than back fees(since the probability of a fire during an individual's occupancy of the property is less than 1, their total back fees will, if the insurance is priced properly, be a fraction of the cost of fighting a fire, corresponding to approximately to the fraction representing their risk). You would essentially have to collect a retail price, since the probability of a fire is 1 when the house is on fire.
I glossed over them not because they are insignificant, I find ethics to be both extremely important in practice and quite interesting as a branch of philosophy; but because I didn't want to focus on them in this context.
Ethical/moral debates in large online fora are generally an exercise in pain and virtual shouting. All smoke, no light, extremely frustrating. I also felt that I had nothing novel or interesting to contribute to the ethical side, which had already settled down into more or less two-position shouting by the time I got here today.
By contrast, I thought that the problem of externalities had not been adequately considered, so I focused on that without letting some half-baked attempt at rehashing a moral position cloud the point. Now, personally, I find the unempathic behavior and ideology that extreme market orientation can reward as "rational" to be rather creepy, and actually not even as efficient as it alleges itself to be(the literature, both theoretical and empirical, on the value of high levels of social trust is fascinating stuff) and I would have much preferred to see the firefighters save the guy's house(with the proviso that I would also expect the guy to act similarly "irrationally" in helping others out if they needed it...)
Reciprocal altruism is both pleasant and surprisingly efficient, as long as levels of value-rational parasitism don't get too high, at which point things can get ugly.
I was a bit unclear there: As with health insurance, nobody(excluding a statistically insignificant number of people who are mentally questionable) will increase their direct risk of fire for that purpose. Even with firefighters, and full insurance, a house fire is a huge pain in the ass and a potential for loss of life.
However, increasing your property's risk of fire spread is a much more reasonable strategy. A few minor landscaping changes, a shrub here, a woodpile there, can control whether your yard is a firebreak or a fuel source, and in what directions. It would be dickish; but not necessarily illogical, to arrange your yard such that people with more to lose feel the need to pay for protection that also includes you.
To make a salient health analogy: in a great many contexts, STD treatment and prophilaxis are heavily subsidized or free, because these diseases are highly contagious and the threat encourages public support for taxpayer funded measures that reduce the ambient threat. By contrast, treatment for noncontagious conditions is, if subsidized at all, done indirectly as some insurance scheme, or on humanitarian grounds.
People don't get to chose the contagiousness of the disease that they have; but the net result is that people with contagious diseases do, in fact, motivate the population around them to subsidize treatment(or, in less humanitarian cases, involuntary confinement on public health grounds). People with less contagious, or noncontagious, diseases do not attract subsidies to the same level. Nobody wants to be sick; but contagious sick people get better subsidies than noncontagious ones do.
The unpleasant complicating factor is that his neighbor, who did pay, got inferior service because the firefighters let the fire grow and spread, rather than suppressing it. Now the guy who did pay the fee gets property damage. You can make moral arguments, if you wish, about saving the guy who didn't pay up; but the guy who did pay up has solid monetary reasons to be upset.
One of the reasons that, historically, certain private fire departments were suppressed, and replaced by tax-funded universal ones was that they were, shall we say, doing a little more than hoping that somebody would have a fire in order to justify their investment...
You would want to set the price ahead of time, based on a reasonable estimate of actual cost+convenience fee, just to avoid sleazy extortion setups; but it would be nice if there were a real retail cost in addition to an insurance option.
For most, risk pooling is a great function of insurance, and makes insurance worth it; but situations where the insured market squeezes out the retail market can get a bit dysfunctional(just look at the healthcare market, where the price paid by insurers is often wildly different from the price on the invoice, which is largely a fiction for anything much more expensive than basic drugs) .
The tricky part here is the externalities(as usual, externalities are a bitch). So, the Cranicks don't pay, they don't get service. Simple enough. There might be some ethical objections; but the economics line up just fine.
Similarly, letting them pay $75 at time of use is a no-go. Fighting a fire costs way more than $75. That's an insurance price, not a retail service price. If you allow people to buy insurance after they need it, you either go bankrupt or the cost of insurance ends up equaling the retail cost of service. You then lose the risk-pooling function of insurance. Now, for things as potentially valuable as houses and their contents, it would be sensible to have an actual retail price(set ahead of time, and publically known, to prevent extortion) that an uninsured person could pay to save a burning building, there are probably a fair few situations where the price of fighting the fire is lower than the cost of replacing the structure, so being able to pay a retail cost of approximately actual cost+service fee would be sensible for both householder and firefighting company.
However, here is where things get unpleasant: Because the Cranicks didn't pay, the firefighters allowed the fire to burn merrily, growing and spreading until it hit somebody who had paid. Now, since the paying householder's property is on fire, they likely suffered some thousands or 10s of thousands in direct combustion, smoke, and water damage. They paid, and they got shitty service. Had the firefighters used the Cranicks property to fight and stop the fire, they could have saved their customers from any damage, and done a much better job of serving them, the ones who actually paid.
Of course, if it becomes known that firefighters will fight fires around an insured property, the obvious strategy is for property owners to club together, buy insurance for 1 plot and get insurance for all for only $75/n. The fire department couldn't support itself on that. If they tried to offer two tiers, a $75 "Fires fought on your property only" and a more expensive "Fires that threaten your property fought", then this creates a perverse incentive: If I live next to a wealthy looking neighbor, I can get him to buy my fire insurance for me just by making my property more dangerous to his. Don't want to encourage that.
This is why firefighting, like certain public health measures, is very hard to elegantly force into a market model unless you are so far in the sticks that each man really is an island. Fires spread, just like diseases. Whether or not the firefighters come to my neighbor's aid matters to me(aside from any debatable moral stuff); because the raging inferno that is his burning house just needs the wind to shift for my house to be next. Even if I've paid my fee, having thousands in water damage from the firefighters, plus smoke and any combustion that occurs before they get there isn't really satisfying. I'd really rather have them fight the fire where it starts, and never have to suffer it myself, rather than insist that everyone pay, and let pockets of fire spread until they endanger me. Same way, even if I don't give a fuck about the life of the guy making my sandwich at the deli, and I don't care how poor he is, I sure do care about what immunizations he has, and whether he can take sick days; because his germs are getting into my food supply.
That is the real complexity of this story, in my opinion. There are some moral questions, but those are debatable, and there really should be a retail price set; but that is a bookkeeping matter; but if I were the insured householder I'd be absolutely livid about this. I paid my dues, and I get lousy service because they are trying to make a point? You could have completely protected my property; but chose to let a nearby building become a danger to it, when I pay you to protect my property? WTF?
Pacifist weakling.
The correct solution is pressure sensors on the wheel and IR eye tracking. If the driver shows inadequate attention given the speed, level of swerving, and measured conditions, if available, a small explosive charge propels their body into the steering wheel/windshield at a velocity directly proportional to the one that a bystander would experience if hit by them...
Poetic justice through superior technology!
The dickwaving over who has the biggest supercomputer seems largely like hype stirred up to enhance IBM's shareholder value; but if you are going to make a dubiously sensible investment in expensive toys, doing it with borrowed money is, in some ways, preferable to doing it with real money.
The huge debts that sovereign nations tend to rack up trigger the same moral instincts that petty consumer debt does; but it isn't at all clear that they work anything like the same way, economically.
China: "Dear US, we are cashing in the giant pile of debt you owe us."
US: "Shucks, China, it looks like we spent all our money on increasingly elaborate pyramid schemes and shitty exurbs that nobody wants. Anyway, thanks for all the free stuff over the years, and I hope you don't find the sudden transition from high-employment export economy to moderate-unemployment internal economy too jarring... TTLGTG!"
Let me translate IBM's statement into a more sincere, less carefully spun, form, and you'll see why an uncaring profit-maximizing multinational is wrapping itself in the flag:
PR: "But now, an IBM representative has said to a Washington, DC forum, 'You have sovereign nations making material investments of a tremendous magnitude to basically eat our lunch, eat our collective lunch.'"
Translation: "But now, an IBM lobbyist has said to a Washington, DC forum 'Other countries are doling out sweetheart contracts to manufacturers and designers of expensive computers. Give us a giant pile of money or the chinks win."
The ionizing portion of sunlight is pretty slim. A little UV and not much else. Certainly not zero potential for cellular damage(but neither is simple cell division in a dark box entirely free of mutagens. Entropy just gets you sometimes)...
Sunlight is powerful in a zergling rush sort of way; but the energy levels of each of the various photons are pretty low.