Downloading and burning a copy of Windows is legal (with a huge number of complecated caveats and exceptions). Providing a copy for others to download and burn can sometimes also be legal, with more restrictions and exceptions.
Case closed. You ARE an idiot.
For example, under your understanding of the law, how do you explain this court ruling, in which providing copies of sim city and word perfect and other copyrighten software to any and all downloaders was ruled legal ?
This was a legal loophole before the No Electronic Theft Act was passed in 1997 (in response to this case). Besides, he was still violating copyright law. He just couldn't be CRIMINALLY prosecuted for it. If one of the software companies sued him, he would certainly be found guilty.
Then, you can examine the statements of Microsoft or the BSA and make up your own now educated mind.
I suppose misinterpreting the law is useful if you want to feel better about breaking it, but only attorneys are actually qualified to give legal opinions. The US code is meaningless if you don't know all the case law surrounding it. That's why lawyers get paid a lot. How about you go ask one and then get back to me?
They need not have worried because running the program was already covered under fair-use, which is another large portion of US law you should familiarize yourself with.
I think you need to familiarize yourself with that area of the law. Care to cite some court cases in which running a program without a license was judged to be fair use? You are aware that the US Code does not mean a thing, right? The US legal system is based on precedent, not code. That's why we have lawyers.
In they 1990's Congress added provisions to the code allowing you to transfer your music onto CD's and tapes. (And added a tax on music handling hardware and media which goes directly to the major labels.)
What exactly does music have to do with any of this? I thought we were discussing software.
When you download and install something, you are implicitly acknowledging your acceptance of whatever license the software is under. Otherwise, you would be breaking copyright law. A click-wrap license is necessary to make it easier to prove that you were aware of that agreement, and makes enforcing the license significantly easier, whether it's GPL or Microsoft's EULA. Without it, you would have to do more work to prove willful infringement.
So under what license was the installation performed? I'm not aware of any specific legal precedent clarifying this one way or the other, but the black-letter law appears pretty clear. Perhaps you can cite some counterexamples?
Generally, the EULA is presented before any actual copying occurs. My understanding is that software that is not copied onto the hard drive and runs directly from the CD-ROM does not need to be licensed, but installing it to a hard drive is not permitted without a license.
Also, you obviously haven't actually bothered reading the GPL. Here is section 8:
8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License.
If distribution in your country is not permitted, then you cannot use the software legally.
Yes, it is copyrighted, but use is not one of the rights reserved to the copyright holder by copyright law.
You need to download a copy of the program before you can use it and/or copy it from the distribution medium to the hard drive.
Under copyright law, the copyright holder has no right to prevent you from reading their book, listening to their song or running their program. US copyright law even has a specific exemption for copies made in the normal process of using the program.
So, if I buy a burned copy of Windows XP off of my friend and install it, it's legal? Or share one legal copy with 10 different people? Don't think so. That exemption is nowhere near as extensive as you think it is. The courts' interpretation of it is very narrow. It doesn't include things like installing the program onto a hard drive.
The GPL is not an End User License Agreement. The license specifically says it does not in any way restrict running the program, and in fact it really doesn't apply to end users at all. The GPL applies to those who want to modify and/or distribute the program, not to those who merely use it.
Well, it permits copying, among other things. Copying is generally necessary to use a program.
If you merely use the program, there is neither any need for you to accept or abide by the terms of the GPL, nor, in fact, any way for you to abide by the terms, because the GPL doesn't specify any terms.
Actually, it has a couple of restrictions. One is the patent clause: if you live in a country in which the software would violate a patent, you may be prohibited from using the software. The other is the warranty clause: "SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION."
And you are an idiot. The copyright laws certainly cover copying, that's why they are called COPYright laws. The act of downloading and running the program is COPYING. If the GPL does not grant you this right, then you are doing it illegally, because nothing else grants you that right, either. By your logic, downloading and burning a copy of Windows XP is legal, as long as I don't enter into a contract with Microsoft. That's not the way it works.
The GPL is most certainly a contract. You must agree to it in order to lawfully install or use the piece of software in question. GPLed software is still copyrighted, and is the property of its author. You merely get a license to use it, modify it, and distribute it. It's not different from any other EULA, other than giving you slightly more rights in exchange for certain obligations (such as providing a copy of the source code if you redistribute it). If you don't abide by the terms of the GPL, you can be sued for copyright infringement, including monetary damages. A click-wrap agreement is a good way to make enforcement easier.
Uh, no. The "speed of light" is speed of light in vacuum. The speed of light in a glass fiber is much, much slower -- pretty close to the speed of an electromagnetic wave in a copper conductor. In addition, since the light reflects from the sides of the fiber at an angle, the actual path length is much larger than the length of the fiber cable. This means wave propagation in fiber is actually slower than in a copper pair.
Well, you've just gotten lucky. DSL lines are oversold more than 10 to 1. As in, a DSLAM with 300 ADSL ports might only have a single T3 going to it -- or worse. That's the real reason why ISPs don't like BitTorrent, Vonage, web servers, and other bandwidth-intensive things on residential connections.
If I get a plan from, say, Sprint, I pretty much have to choose a phone they sell with that plan. Sprint heavily subsidizes the phone, and discourages the whole bring-your-own phone thing. Therefore, if I'm a manufacturer, I care mostly about whether the provider is happy. If I have to pick between making the provider happy or making the user happy, I'll make the provider happy -- otherwise, they'll stop buying my phones. Adding features that compete with the provider's own services is not a good idea if I want to stay in business.
And yes, manufacturers often treat the retailer as their real customer, especially with companies like Wal-mart. If Wal-mart demands they make it cheap, they will cut corners on everything, even if that makes the consumers less satisfied. With music, Wal-mart can get the studios to release a specially censored version just for them. Do you think the manufacturers there really care about the consumer?
It's not even trusted computing. It's just that Linus wants his system to be successful, and forcing companies to do things they do not want to do is a great way to fail at that. There are plenty of reasons why companies don't want people to mess with their hardware -- control, competition, proprietary licensing, and many other reasons. If you are using a piece of software or hardware licensed from someone else and have to respect the NDA, you can't just go handing out source code left and right. For instance, wireless router chip companies have lots of competition (in China, for one), so they are notoriously tight-lipped.
Sure, but you don't get the cover art, per-song EQ settings, normalization levels, play counts, ratings, customized playlists, and other nice features you get transparently with iTunesDB on the iPod.
Considering that the sugar water business has been steadily making tons of money each year for the last 100 years or so, I'd say Jobs is in the right business. How many other products have existed for nearly a hundred years with only minor changes in the product line? Microsoft is kinda the Sam's Choice Dr. Thunder in that market -- close, but no cigar.
As far as PPC and other marginal PC architectures: they all needed to die a long time ago, and Apple made the right decision. Computers have matured to the point that a standard architecture is a good thing, and there's nothing really wrong with x86.
Let me just reiterate that my point is not that it is impossible to use compressed air to store energy and perhaps power a car. It's possible. My point is that fundamental thermodynamical constraints impose hard upper limits on the efficiency of that process. As in, it won't scale. With the pressures they are using, the energy density is horrible -- their car has a range of something like 50 miles moving less than 30 miles per hour, and that's without headlights, windshield wipers, ventilation, air conditioning, radio, and other things we expect from a car. Adding those things would reduce the range even further. This isn't something you can easily solve -- the tanks are already state of the art, so you can't jack up the pressure. On top of that, the efficiency is horrible and cannot be improved. When you are running the compressor, a large part of the energy input is dissipated as heat (this is proportional to the difference in pressures). If you want to increase the pressure to get better energy density, even more energy gets wasted by the compressor. This cannot be avoided in any way. In fact, this car is probably already far less efficient than an ICE when you consider losses in the compressor. They say it takes $2.50 to drive 50 miles, which is even worse than a much larger Prius.
In comparison, vehicles powered by batteries have much higher charging efficiency, far better energy density, and far fewer parts. Lithium-ion batteries can easily give you a range of a couple of hundred miles. Electric motors are pretty much maintenance-free. It's easier to add things like power steering or air conditioning. It's easy add regenerative braking, which boosts efficiency by a lot in stop and go traffic. Not to mention, you don't have tanks with enough ultra-high-pressure air to level a small building.
Well, there is no reason you couldn't make an air powered car if you wanted to. Higher pressure and lots of heat exchangers would allow you to use compressed gas to power an engine. The problem is that getting the required efficiency and range is nearly impossible. It would be cheaper and far more efficient to use batteries rather than compressed air.
A little bit of electricity? Just so you know, air compressors are not very efficient devices. They waste a lot of heat. If you want to store 2 kWh of energy as compressed air, you'll need to use maybe 10 kWh of electricity to compress it. Second, compressed air takes has extremely low energy density. A small shop air compressor (compressing to about 150psi) usually has a 10-20 gal tank just to store enough air to run something like an air drill for a minute or two. This is for a tank made of very thick steel. Even if you increased the pressure in the tank 100 times (which is completely impossible from an engineering viewpoint) it would still not have nearly enough energy to power a car (although you could run a drill for an hour). Not to mention, a compressor that can dump a few hundred kWh into a tank even overnight would take up more space than the car and would be loud enough to hear from a mile away. You'd need like a 100hp compressor, those generally take up a large room. Here is a picture of one.
Not to mention, running a car on compressed air is pretty simple, provided you have a source of compressed air. You could even use a regular gas engine without any modifications -- simply force high pressure air into the intake. The reason nobody does it is because doing so would require an ungodly amount of compressed air. There's not much you can do to improve the efficiency, either.
My conclusion: the site is a scam, attempting to extract money from dumb investors. Note how they are focusing on silly things like the benefits of not using gas (obvious) and how they will arrange the seatbelts (irrelevant) while thoroughly avoiding any description of the actual technology. Their tanks are supposedly good to 300 bar (~4300 psi), which is realistic (that's what scuba tanks or CNG tanks are rated for). However, this is not nearly enough energy to power a car. Hell, it's barely enough energy to power a car if you store NATURAL GAS in the tanks and BURN IT.
300 bar = 30 MPa = 30 MJoules / m^3 = 8.3 kWh/m^3. If you somehow managed to put 3 cubic meters of air tanks on that thing (that's about 800 gallons -- a HUGE air tank), you would have as much energy as ONE gallon of gas. To compress that air, you would use up several times that amount, because going from atmospheric pressure to 4500 psi will release a ton of waste heat. You would also never be able to get that energy out in any reasonable length of time because the air will become very, very cold when it expands.
If that technology really worked, we would have had air-powered cars 150 years ago -- it's a steam engine that's hooked up to an air tank instead of a boiler. The problem is, there is no way it could possibly work.
Any system stays the way it is if you don't touch it. That's not the point. If you install Windows and don't ever do anything with it, it will not need touching at all. Neither does Solaris. Hell, Solaris is far more reliable than Linux. In this case, administration consists of things like installing software, doing backups, performing upgrades and security updates, and so on. And if you shove a server in a closet and forget about it, it will die on you at the worst possible time. Either it will run out of disk space, hackers will break into it, or it will experience a hardware failure. From personal experience, I can assure you that Linux does need fixing quite often when used as a workstation. Try installing a custom driver for something, and then performing an automatic update. At least with Windows, I don't need to upgrade my kernel every week like I do with Ubuntu.
Support contracts don't decrease workload for the IT department, unless said contract includes people that are physically present at your location (in which case that contract will cost far more than a few employees). Companies like IBM can provide this level of support, but it's far too expensive to be practical. Redhat's "support" consists of answering questions by phone. This isn't going to reduce the IT department's workload by any appreciable amount. If your IT department consists of one person, it won't let you get rid of him.
What does your post say, then? Your complaint against windows seems to be that: 1. Stupid users break it. and 2. It needs a system administrator who helps users fix problems. How does Linux attack either of these two problems?
Total and utter BS. I spend about 3x as much time adminning my Ubuntu box than I spend on 3 windows setups (laptop, vmware, computer at work where I have admin privs). Linux doesn't magically set itself up, configure itself, and fix itself. And don't tell me it's because I'm a Windows fanboy. My computer at home has been running various flavors of Linux since 1999. And for every Word document you can't open with Word 97 (or whatever your co-worker was using), you can find a hundred that don't format correctly in OpenOffice. In fact, I've never seen a single document look exactly the same in Word and in Openoffice. I'm not even talking about Powerpoint presentations or complicated Excel spreadsheets -- those are a total trainwreck.
Dude, I actually work with the analog/hardware side of DSL every day. I know how it works. I know what the limitations are. Even the best DSL hardware does not work beyond 10-15 kilofeet, and even that requires all kinds of tricks and very high performance from every part of the circuit. The signal on an adjacent pair would be far worse than it is on the original pair even at 15 kilofeet. You'll also get crosstalk from 20-40 other DSL lines. Your line probably isn't terminated properly, so you'll get all kinds of bouncing, ringing, and other crap. The cable is a long distributed capacitor, so the signals would be smeared to hell instead of cross-coupling nicely. The noise and interference will jam whatever is left of the original signal. If it was possible to recover any data under these conditions, DSL could be easily made to work well beyond 15 kilofeet.
So you're saying that a couple of dozen Linux workstations and a bunch of servers won't need ANYONE to service them? Total bullshit. If you are getting by with just one admin and one assistant, it doesn't get much better than that. Besides, with Linux, the only difference would be that your IT guy would be reading newsgroups and mailing lists instead of wading through the knowledge base. Linux breaks, too, and Redhat's support contracts aren't all that great. I think their "support contract" consists of helping you fix things yourself over the phone. Now factor in the hidden costs of solving interoperability problems with Linux (considering it can't open Office documents natively and isn't compatible with MSIE-only websites and intranet applications), and I think Linux would require considerably more support. There is a reason techies usually make bad managers -- your preference for a technology severely distorts your perspective, and you end up making bad decisions.
Not to mention, it's the cost of an illegal copy of WinXP. OEM copies purchased separately from a PC are illegal. If you get a BSA audit, your company will get reamed big-time. And dell charges $150 just to _upgrade_ to XP Pro. Not to mention, XP comes with _zero_ support. You have to pay something like $40 per incident if you want to call Microsoft.
In any case, you should be picking software based solely on _merits_: as in, cost vs. benefit. I am not sure why the author even mention Qt, since that's a commercial package. Since you mention switching to C#, you obviously had no business using Qt in the first place, since your app needs to run only on Windows. Obviously, a Windows-only app should use the API directly rather than through a wrapper, if you are comfortable with that API. On the other hand, for cross-platform development, Qt is quite a bargain. It saves lots of development time while only costing $3,300 per developer and being royalty free. Sometimes open-source solutions are better, sometimes commercial solutions are better. You shouldn't have to justify your decision with religious beliefs (as in "but at least we aren't supporting evil microsoft"). If Windows works better than Linux for your application, you should use it.
Slashdot Mirror
Most people here hate Sony after their DRM fiasco and for their general stupidity. If you don't like that the editors are biased, go read digg.
Downloading and burning a copy of Windows is legal (with a huge number of complecated caveats and exceptions). Providing a copy for others to download and burn can sometimes also be legal, with more restrictions and exceptions.
Case closed. You ARE an idiot.
For example, under your understanding of the law, how do you explain this court ruling, in which providing copies of sim city and word perfect and other copyrighten software to any and all downloaders was ruled legal ?
This was a legal loophole before the No Electronic Theft Act was passed in 1997 (in response to this case). Besides, he was still violating copyright law. He just couldn't be CRIMINALLY prosecuted for it. If one of the software companies sued him, he would certainly be found guilty.
Then, you can examine the statements of Microsoft or the BSA and make up your own now educated mind.
I suppose misinterpreting the law is useful if you want to feel better about breaking it, but only attorneys are actually qualified to give legal opinions. The US code is meaningless if you don't know all the case law surrounding it. That's why lawyers get paid a lot. How about you go ask one and then get back to me?
They need not have worried because running the program was already covered under fair-use, which is another large portion of US law you should familiarize yourself with.
I think you need to familiarize yourself with that area of the law. Care to cite some court cases in which running a program without a license was judged to be fair use? You are aware that the US Code does not mean a thing, right? The US legal system is based on precedent, not code. That's why we have lawyers.
In they 1990's Congress added provisions to the code allowing you to transfer your music onto CD's and tapes. (And added a tax on music handling hardware and media which goes directly to the major labels.)
What exactly does music have to do with any of this? I thought we were discussing software.
So under what license was the installation performed? I'm not aware of any specific legal precedent clarifying this one way or the other, but the black-letter law appears pretty clear. Perhaps you can cite some counterexamples?
Generally, the EULA is presented before any actual copying occurs. My understanding is that software that is not copied onto the hard drive and runs directly from the CD-ROM does not need to be licensed, but installing it to a hard drive is not permitted without a license.
Also, you obviously haven't actually bothered reading the GPL. Here is section 8:
If distribution in your country is not permitted, then you cannot use the software legally.
Yes, it is copyrighted, but use is not one of the rights reserved to the copyright holder by copyright law.
You need to download a copy of the program before you can use it and/or copy it from the distribution medium to the hard drive.
Under copyright law, the copyright holder has no right to prevent you from reading their book, listening to their song or running their program. US copyright law even has a specific exemption for copies made in the normal process of using the program.
So, if I buy a burned copy of Windows XP off of my friend and install it, it's legal? Or share one legal copy with 10 different people? Don't think so. That exemption is nowhere near as extensive as you think it is. The courts' interpretation of it is very narrow. It doesn't include things like installing the program onto a hard drive.
The GPL is not an End User License Agreement. The license specifically says it does not in any way restrict running the program, and in fact it really doesn't apply to end users at all. The GPL applies to those who want to modify and/or distribute the program, not to those who merely use it.
Well, it permits copying, among other things. Copying is generally necessary to use a program.
If you merely use the program, there is neither any need for you to accept or abide by the terms of the GPL, nor, in fact, any way for you to abide by the terms, because the GPL doesn't specify any terms.
Actually, it has a couple of restrictions. One is the patent clause: if you live in a country in which the software would violate a patent, you may be prohibited from using the software. The other is the warranty clause: "SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION."
And you are an idiot. The copyright laws certainly cover copying, that's why they are called COPYright laws. The act of downloading and running the program is COPYING. If the GPL does not grant you this right, then you are doing it illegally, because nothing else grants you that right, either. By your logic, downloading and burning a copy of Windows XP is legal, as long as I don't enter into a contract with Microsoft. That's not the way it works.
The GPL is most certainly a contract. You must agree to it in order to lawfully install or use the piece of software in question. GPLed software is still copyrighted, and is the property of its author. You merely get a license to use it, modify it, and distribute it. It's not different from any other EULA, other than giving you slightly more rights in exchange for certain obligations (such as providing a copy of the source code if you redistribute it). If you don't abide by the terms of the GPL, you can be sued for copyright infringement, including monetary damages. A click-wrap agreement is a good way to make enforcement easier.
Uh, no. The "speed of light" is speed of light in vacuum. The speed of light in a glass fiber is much, much slower -- pretty close to the speed of an electromagnetic wave in a copper conductor. In addition, since the light reflects from the sides of the fiber at an angle, the actual path length is much larger than the length of the fiber cable. This means wave propagation in fiber is actually slower than in a copper pair.
Well, you've just gotten lucky. DSL lines are oversold more than 10 to 1. As in, a DSLAM with 300 ADSL ports might only have a single T3 going to it -- or worse. That's the real reason why ISPs don't like BitTorrent, Vonage, web servers, and other bandwidth-intensive things on residential connections.
If I get a plan from, say, Sprint, I pretty much have to choose a phone they sell with that plan. Sprint heavily subsidizes the phone, and discourages the whole bring-your-own phone thing. Therefore, if I'm a manufacturer, I care mostly about whether the provider is happy. If I have to pick between making the provider happy or making the user happy, I'll make the provider happy -- otherwise, they'll stop buying my phones. Adding features that compete with the provider's own services is not a good idea if I want to stay in business.
And yes, manufacturers often treat the retailer as their real customer, especially with companies like Wal-mart. If Wal-mart demands they make it cheap, they will cut corners on everything, even if that makes the consumers less satisfied. With music, Wal-mart can get the studios to release a specially censored version just for them. Do you think the manufacturers there really care about the consumer?
It's not even trusted computing. It's just that Linus wants his system to be successful, and forcing companies to do things they do not want to do is a great way to fail at that. There are plenty of reasons why companies don't want people to mess with their hardware -- control, competition, proprietary licensing, and many other reasons. If you are using a piece of software or hardware licensed from someone else and have to respect the NDA, you can't just go handing out source code left and right. For instance, wireless router chip companies have lots of competition (in China, for one), so they are notoriously tight-lipped.
Sure, but you don't get the cover art, per-song EQ settings, normalization levels, play counts, ratings, customized playlists, and other nice features you get transparently with iTunesDB on the iPod.
Considering that the sugar water business has been steadily making tons of money each year for the last 100 years or so, I'd say Jobs is in the right business. How many other products have existed for nearly a hundred years with only minor changes in the product line? Microsoft is kinda the Sam's Choice Dr. Thunder in that market -- close, but no cigar.
As far as PPC and other marginal PC architectures: they all needed to die a long time ago, and Apple made the right decision. Computers have matured to the point that a standard architecture is a good thing, and there's nothing really wrong with x86.
Do you have a cell phone? Probabaly. Would you pay more than $60 a month for one? Unlikely. Why would they need to add more features?
As far as cell phone MANUFACTURERS, their customers are the cellphone COMPANIES, not you.
Well, not unless you use gtkpod or anapod or about half a million other programs.
Let me just reiterate that my point is not that it is impossible to use compressed air to store energy and perhaps power a car. It's possible. My point is that fundamental thermodynamical constraints impose hard upper limits on the efficiency of that process. As in, it won't scale. With the pressures they are using, the energy density is horrible -- their car has a range of something like 50 miles moving less than 30 miles per hour, and that's without headlights, windshield wipers, ventilation, air conditioning, radio, and other things we expect from a car. Adding those things would reduce the range even further. This isn't something you can easily solve -- the tanks are already state of the art, so you can't jack up the pressure. On top of that, the efficiency is horrible and cannot be improved. When you are running the compressor, a large part of the energy input is dissipated as heat (this is proportional to the difference in pressures). If you want to increase the pressure to get better energy density, even more energy gets wasted by the compressor. This cannot be avoided in any way. In fact, this car is probably already far less efficient than an ICE when you consider losses in the compressor. They say it takes $2.50 to drive 50 miles, which is even worse than a much larger Prius.
In comparison, vehicles powered by batteries have much higher charging efficiency, far better energy density, and far fewer parts. Lithium-ion batteries can easily give you a range of a couple of hundred miles. Electric motors are pretty much maintenance-free. It's easier to add things like power steering or air conditioning. It's easy add regenerative braking, which boosts efficiency by a lot in stop and go traffic. Not to mention, you don't have tanks with enough ultra-high-pressure air to level a small building.
Well, there is no reason you couldn't make an air powered car if you wanted to. Higher pressure and lots of heat exchangers would allow you to use compressed gas to power an engine. The problem is that getting the required efficiency and range is nearly impossible. It would be cheaper and far more efficient to use batteries rather than compressed air.
A little bit of electricity? Just so you know, air compressors are not very efficient devices. They waste a lot of heat. If you want to store 2 kWh of energy as compressed air, you'll need to use maybe 10 kWh of electricity to compress it. Second, compressed air takes has extremely low energy density. A small shop air compressor (compressing to about 150psi) usually has a 10-20 gal tank just to store enough air to run something like an air drill for a minute or two. This is for a tank made of very thick steel. Even if you increased the pressure in the tank 100 times (which is completely impossible from an engineering viewpoint) it would still not have nearly enough energy to power a car (although you could run a drill for an hour). Not to mention, a compressor that can dump a few hundred kWh into a tank even overnight would take up more space than the car and would be loud enough to hear from a mile away. You'd need like a 100hp compressor, those generally take up a large room. Here is a picture of one.
Not to mention, running a car on compressed air is pretty simple, provided you have a source of compressed air. You could even use a regular gas engine without any modifications -- simply force high pressure air into the intake. The reason nobody does it is because doing so would require an ungodly amount of compressed air. There's not much you can do to improve the efficiency, either.
My conclusion: the site is a scam, attempting to extract money from dumb investors. Note how they are focusing on silly things like the benefits of not using gas (obvious) and how they will arrange the seatbelts (irrelevant) while thoroughly avoiding any description of the actual technology. Their tanks are supposedly good to 300 bar (~4300 psi), which is realistic (that's what scuba tanks or CNG tanks are rated for). However, this is not nearly enough energy to power a car. Hell, it's barely enough energy to power a car if you store NATURAL GAS in the tanks and BURN IT.
300 bar = 30 MPa = 30 MJoules / m^3 = 8.3 kWh/m^3. If you somehow managed to put 3 cubic meters of air tanks on that thing (that's about 800 gallons -- a HUGE air tank), you would have as much energy as ONE gallon of gas. To compress that air, you would use up several times that amount, because going from atmospheric pressure to 4500 psi will release a ton of waste heat. You would also never be able to get that energy out in any reasonable length of time because the air will become very, very cold when it expands.
If that technology really worked, we would have had air-powered cars 150 years ago -- it's a steam engine that's hooked up to an air tank instead of a boiler. The problem is, there is no way it could possibly work.
Any system stays the way it is if you don't touch it. That's not the point. If you install Windows and don't ever do anything with it, it will not need touching at all. Neither does Solaris. Hell, Solaris is far more reliable than Linux. In this case, administration consists of things like installing software, doing backups, performing upgrades and security updates, and so on. And if you shove a server in a closet and forget about it, it will die on you at the worst possible time. Either it will run out of disk space, hackers will break into it, or it will experience a hardware failure. From personal experience, I can assure you that Linux does need fixing quite often when used as a workstation. Try installing a custom driver for something, and then performing an automatic update. At least with Windows, I don't need to upgrade my kernel every week like I do with Ubuntu.
Support contracts don't decrease workload for the IT department, unless said contract includes people that are physically present at your location (in which case that contract will cost far more than a few employees). Companies like IBM can provide this level of support, but it's far too expensive to be practical. Redhat's "support" consists of answering questions by phone. This isn't going to reduce the IT department's workload by any appreciable amount. If your IT department consists of one person, it won't let you get rid of him.
What does your post say, then? Your complaint against windows seems to be that:
1. Stupid users break it.
and
2. It needs a system administrator who helps users fix problems.
How does Linux attack either of these two problems?
Total and utter BS. I spend about 3x as much time adminning my Ubuntu box than I spend on 3 windows setups (laptop, vmware, computer at work where I have admin privs). Linux doesn't magically set itself up, configure itself, and fix itself. And don't tell me it's because I'm a Windows fanboy. My computer at home has been running various flavors of Linux since 1999. And for every Word document you can't open with Word 97 (or whatever your co-worker was using), you can find a hundred that don't format correctly in OpenOffice. In fact, I've never seen a single document look exactly the same in Word and in Openoffice. I'm not even talking about Powerpoint presentations or complicated Excel spreadsheets -- those are a total trainwreck.
Dude, I actually work with the analog/hardware side of DSL every day. I know how it works. I know what the limitations are. Even the best DSL hardware does not work beyond 10-15 kilofeet, and even that requires all kinds of tricks and very high performance from every part of the circuit. The signal on an adjacent pair would be far worse than it is on the original pair even at 15 kilofeet. You'll also get crosstalk from 20-40 other DSL lines. Your line probably isn't terminated properly, so you'll get all kinds of bouncing, ringing, and other crap. The cable is a long distributed capacitor, so the signals would be smeared to hell instead of cross-coupling nicely. The noise and interference will jam whatever is left of the original signal. If it was possible to recover any data under these conditions, DSL could be easily made to work well beyond 15 kilofeet.
So you're saying that a couple of dozen Linux workstations and a bunch of servers won't need ANYONE to service them? Total bullshit. If you are getting by with just one admin and one assistant, it doesn't get much better than that. Besides, with Linux, the only difference would be that your IT guy would be reading newsgroups and mailing lists instead of wading through the knowledge base. Linux breaks, too, and Redhat's support contracts aren't all that great. I think their "support contract" consists of helping you fix things yourself over the phone. Now factor in the hidden costs of solving interoperability problems with Linux (considering it can't open Office documents natively and isn't compatible with MSIE-only websites and intranet applications), and I think Linux would require considerably more support. There is a reason techies usually make bad managers -- your preference for a technology severely distorts your perspective, and you end up making bad decisions.
Not to mention, it's the cost of an illegal copy of WinXP. OEM copies purchased separately from a PC are illegal. If you get a BSA audit, your company will get reamed big-time. And dell charges $150 just to _upgrade_ to XP Pro. Not to mention, XP comes with _zero_ support. You have to pay something like $40 per incident if you want to call Microsoft.
In any case, you should be picking software based solely on _merits_: as in, cost vs. benefit. I am not sure why the author even mention Qt, since that's a commercial package. Since you mention switching to C#, you obviously had no business using Qt in the first place, since your app needs to run only on Windows. Obviously, a Windows-only app should use the API directly rather than through a wrapper, if you are comfortable with that API. On the other hand, for cross-platform development, Qt is quite a bargain. It saves lots of development time while only costing $3,300 per developer and being royalty free. Sometimes open-source solutions are better, sometimes commercial solutions are better. You shouldn't have to justify your decision with religious beliefs (as in "but at least we aren't supporting evil microsoft"). If Windows works better than Linux for your application, you should use it.