Where is 'innocent until proven guilty' these days?
Where it has always been, in the hands of the jury. The doctrine that you are considered innocent until proven guilty beyond a reasonable doubt does not imply that every single piece of evidence introduced into the case must be able to single handedly prove that guilt.
I can guess at why the prosecution would want to get this evidence in the record and why it would be admitted.
Sure, to prejudice the jury by introducing facts that they aren't going to understand.
if there is other evidence suggesting you have bad files and you have intentionally made those file unreadable, the tools you used to do that are possibly relevant.
OK, but what if there isn't evidence suggesting you intentionally made those files undreadable? "The court didn't say that police had unearthed any encrypted files"
Kind of like pointing out the defendant owned a shredder, there was huge pile of shredded paper by it, and the "smoking gun" documents are no where to be found.
In this case though, they seem to have merely pointed to the existence of the shredder in the first place. Of course, I haven't read the actual ruling, as I can't find it. But from what I read in the article it seems that erred in allowing the evidence. Whether or not that amounts to reversible error I won't even speculate, though.
This wasn't about an insanity defense, which is an affirmative defence which would have had to have been raised at the beginning of the trial. Rather, this is about the mens rea element of the law.
I know you've heard that "ignorance is no excuse" for the law, but that's a gross oversimplification. Rather than explain mens rea to you, I'll point you to the Wikipedia article, which does a better job of it than I can.
In New York v. Ferber, 458 U.S. 747, 765 (1982), the Supreme Court upheld a broad child pornography statute but cautioned that, "[a]s with obscenity laws, criminal responsibility may not be imposed without some element of scienter on the part of the defendant."
Don't exaggerate. This ruling wasn't that anyone with PGP has criminal intent. It was that a judge admitting the fact that the criminal had PGP into evidence wasn't reversible error. If pretty much every Linux distribution had GPG included, and you had such a Linux distribution, it would be quite easy for the defense to put this into evidence.
Locking your doors is a normal and irrelevant thing-- just like encrypting your data before your send it through the internet.
If all he had on his computer was a web browser with HTTPS, that'd be true. But he had PGP, which isn't very common at all. Should we trust a jury to make the right decision when given this evidence (along with presumably a defense showing just how many other reasons he could have had PGP)? I really don't know. I suppose it depends in part on how strongly the defense objected to the evidence before it was presented. Since he wasn't actually using the PGP to encrypt anything I would probably lean toward not admitting the evidence. But is this reversible error sufficient to overturn a ruling? I can't say it is without knowing the details of the case.
The particular reason for encrypting certain documents may be so that your little sister won't read your diary. You don't care if she reads your term paper, so it isn't encrypted.
Anything's possible. It's up to the jury to decide whether that's reasonable. Maybe the guy didn't even have a sister. If he did, and the sister regularly used the computer, then maybe this is something the defense could bring up.
TFA doesn't say that the man used encryption to cover up illegal activity, but that the very existence of PGP itself indicated criminal intent, which is ludicrous.
If that's true, then I'd suggest TFA is wrong. Her's what I read:
"We find that evidence of appellant's Internet use and the existence of an encryption program on his computer was at least somewhat relevant to the state's case against him," Judge R.A. Randall wrote in an opinion dated May 3.
Without knowing more about the case, I can't say if I agree or not.
You can't say that windows is as cross plaform as java, because windows programs don't run on java.
Huh?
The point was that wine doesn't provide cross platform for windows binaries, because you are tied to a subset of windows programs, and just x86 processors.
I believe that a program which runs on multiple operating systems is cross platform. It doesn't need to run on multiple architectures. But that's just semantics. Regardless of what you want to call it, what Java provides is not very useful.
You can do whatever you want with C++, but the fact is that you will have a hard time finding people doing good code for C++, and it's much easier to find good java coders.
If that's true it's pretty sad. But I'd rather spend the time and money finding someone who can code well in C++ than release a product in java just because I'm too cheap to build things right.
With Java, safe comes for free.
Safe comes at the cost of slow, bulky, and difficult to install.
You just keep repeating the same preconceptions that make you think java is useless, and I shouldn't try to keep you from doing it.
I'm not sure what it is I'm repeating, but if I'm repeating something it's because you haven't addressed it. Moreover, I can't prove that java is useless. In fact, I admit it has at least one use - applets and other sandboxed software. But as for the rest, I'm just not convinced.
I find Java is a very powerful and versatile laguage, with no measurable shortcomings compared to other commonly used languages, and lots of advantages as a platform.
The difference in speed is a measurable shortcoming. The difference in size of the entire package (including the interpreter) is a measurable shortcoming. And the difference in ease of installation (have to ensure that everyone has the interpreter, and have to ensure that the interpreter they have is compatible with the bytecode you have).
The fact that a bunch of people were forced to learn java in college in the early 2000s doesn't make up for these shortcomings. Not by a longshot. Anyone who is a decent coder can easily write safe and efficient code in C as easily as they can in Java. Just because C allows you to do something doesn't mean you have to do it.
The ability to have binary compatible software, for one thing.
The Mac would die the day the CPU would be the same as in a generic PC. Not from a architectural standpoint, I think they could make it happen, but marketingwise.
Now that's an interesting point of view. There probably are some people who buy Macs because of the chip. In my opinion it's not that big of a selling point, but maybe it is.
Chances are if you are running OS X, you probably paid for it at least once every HW upgrade.
And chances are if you are running anything but OS X, you probably paid for Windows at least once every HW upgrade, whether you use it or not.
Getting a PC without any OS on it and then installing a pirated copy is too hard for the average Joe. I think Apple's fears that people doing this will cause a significant decrease in revenues are misplaced.
I can't imagine why Apple would want to move towards x86 hardware
Probably the number one reason would be software compatibility. By moving to x86 hardware, and then customizing Wine, Apple could offer seemless integration of Windows software on OSX.
If apple goes up whenever this rumor comes out, I guess the shareholders think they should actually do it. How long until the shareholders vote in a board of directors who will follow this?
Anyway, I'm enjoying the INTC runup, which makes a lot more sense economically.
It's just Apple trying to get better terms/service from IBM (think Dell's "talks" with AMD)
Total speculation, but maybe.
It will be the death of Apple's hardware division
I don't see how using a different chip could possible kill Apple's hardware division. Could you explain? Note that using a different chip wouldn't mean that Apple would have to allow OEMs to distribute hardware with OSX pre-installed. They could, and given a high enough price for the software (to compensate for the lost profits from hardware), they should, but they wouldn't have to.
Apple will have a hard time supporting the myriad boards, chipsets, and peripherals of PCs
They wouldn't have to support anything they don't already support or that they make themselves. Again, using the Intel chip isn't the same as allowing clones. Those who want to install OSX on non-Apple built or sanctioned hardware do so at their own risk.
Piracy/sharing (pick your preferred new-speak term) will mean a revenue-less expansion of the install base
Maybe. This is the biggest worry. But I think Microsoft has shown that most people aren't going to go through the hassle of installing their own operating system just so they can save a few bucks. Installing an operating system is hard, and most people are going to keep whatever they get with their computer.
If PC users can just buy OS X86 for $99, they might give Mac a try. It wouldn't take that high conversion rate for OS software profits to easily replace hardware profits.
This is the key here. It doesn't matter if Apple is making money off software or hardware, as long as they're making money. It does lead to a different problem, though. Most likely the price point for OSX would have to be set a lot higher than the price for Windows in order for Apple to continue making the same amount of profits. With the prices separated out like that, people are going to realize exactly how much more OSX costs. Of course, Apple could make the switch to Intel chips and only sell the OS bundled with its own hardware. One advantage they'd retain would be in the relative ease of porting an application from one OS to another.
All the hardware interfaces would be 100% compatible with a pc. Even if Apple shipped custom firmware to restrict hardware used with the platform it would be trivial to reflash. Nope, once they go x86 Apple has no more firm a grip on the hardware than any other pc vendor.
I'm curious. Other than the motherboard/operating system/case, what hardware does Apple make serious money on? They definitely don't have a monopoly on hard drives for the Mac, and I'm pretty sure you don't need Apple's permission to make a PCI card. So how would Apple lose, exactly?
Yeah, people could download a CD off the Internet, reflash their BIOS, and install OSX on their PCs. But how many people are actually going to do that? How many people buy computers without OSes and then install a pirated copy of Windows? Not a whole lot, percentage-wise. So I don't think many people would do it for OSX either.
As long as Apple can stop people from reselling computers with OSX pre-installed, they'll still maintain their ability to charge people who use Macs. Sure, you can say Apple makes money off hardware and Microsoft makes money off software, but when most of the people buying Windows are getting it bundled with the hardware anyway, is there really a difference?
Very few people build their own computers. Most buy a box from their local computer store, or order it online from a company like Dell.
OK, but if you could buy the parts and build your own Mac, then Dell could buy the parts and build a Mac for you. Maybe not legally, though. Apple could presumably make it legal to build clones for personal use, but illegal to build them for resale.
Also, no one would run OSX on a standard PC. Just like no one runs BeOS, or ran OS/2.
OSX has a much bigger base of users than BeOS or OS/2. And unlike Be, Apple has a lot of experience delivering complete systems with the OS pre-installed.
An x86 Apple would probably be a proprietary Apple with an x86 in, and no one would care. Just like no one cares that there is a PowerPC in them today.
Most people would care, and use Windows. Some people would care, and use OSX. Some people would care, and use Linux. Some people, including me, would (still) buy the cheapest hardware without regard to what operating system is installed on it. Where Apple would make its biggest gain is with people who want to use OSX, but are forced to use Windows because they are tied to software which is only available on Windows. If Macs ran on X86 chips, then running Windows software on Macs becomes a lot easier - you'd use Wine, and not an emulator.
If they did it right, it'd probably be a win-win situation. Just keep a little bit of the Operating System in ROM, and don't let anyone else sell computers with the OS pre-installed. Sure, you'll get some hard-core geeks who will quickly copy the OS and offer install CDs on some P2P networks, but most people don't install their own OS just to save a few bucks. Meanwhile, you'll make the cost of supporting OSX in addition to Windows come down greatly, and you can probably sell a few OSX CDs to people who already own Intel hardware and are willing to pay to change their OS.
Wine is x86-only, and not all windows programs run on wine.
Cross-platform means it runs on multiple operating systems, not all operating systems. And yes, not all Windows programs run on Wine. Not all run on Java, either.
You are just talking about Sun JDK, but java is more than that.
Well, I was talking about the cross-platform aspect.
It's a language inherently somewhat slower than C++, but much safer, and in some specific cases, even faster than C++.
Anything that can be written in Java can be written as a faster program in C, and hence in C++. It might take more time to develop such a program, but anything you can do in Java you can do in C.
I was just trying to explain that such speed decrease is a packaging, distribution, or even a marketing problem, other than a technical problem.
Unless you're talking about compiled java, in which case you eliminate the only real advantage of java (compile once, run anywhere), the speed decrease in inherent in the very fact that java is an interpreted language.
When competing with native programs in any platform, Java doesn't need to be cross platform, because that's wha native means, non general purpose, but close tot he hardware. In those cases, you can compile with GCJ, and have a theoretically slightly slower binary than with C++, but actually a better binary, because it doesn't have the possibility of buffer overflows, and for all practical purposes doesn't have a speed difference with C++.
There are garbage collection libraries available for C++. One can easily use a memory allocation library which checks for buffer overflows, and prove quite easily that the code doesn't use any memory without going through the library. One generally doesn't do this, because it's relatively easy to code in a way which doesn't allow buffer overflows, but it is possible. In C++ you have the option. With Java, you're forced to do everything the Java way.
When you need cross platform, you use sun JDK, or the native JDK, with the startup delays you talk about, but in this case, it's not slower than anything, because nothing else is cross platform at the binary level, so it's xxx speed against zero speed, so it's still faster, because any positive speed is better than zero, period.
Byte code is not the same as the binary level, and any interpreted language can accomplish the same thing, so I don't see where you're saying that "nothing else" compares.
But realistically, eliminating the need to recompile is a rather narrow market, especially when you consider the memory and storage space issues inherent with java. For anything outside that narrow market, you'd be better off compiling once for each platform.
One market where Java is ahead of the others is in creating a sandboxed environment where you can run untrusted software. This is an example of a market where running compiled software is impossible. This is especially true for Windows. In unix, you can just set up a chrooted sandbox and run the compiled software in that, but Windows AFAIK doesn't support that. So yeah, if I need to run software in a sandboxed environment on Windows, Java would be my number one choice. But other than that one purpose, applets, I don't see the point of Java.
Bandwidth has nothing to do with the number of towers.
So if there was only one cell phone tower in the world, and everyone in the world had to share the same frequencies on that one cell phone tower, we'd still have the same bandwidth available?
Now consider what happens in the physical case of two towers broadcasting distinct signals on the same band -- we can think of each signal as a channel. Note that this would be required for your scheme to work. For simplicity, assume that they are broadcasting omnidirectionally.
Your simplification greatly reduces the usefulness of having multiple towers.
Fourier analysis (and experimental evidence) assure us that the signals are going to interfere with each other.
Sure, to some very very small extent, which drops off as the square of the distance from the tower. IOW, yes, a cell phone tower in Mexico interferes with a cell phone tower in New York City, but the interference is negligible. Now, reduce the wattage outputted by the tower, and move them closer together, and you get the same neglible interference.
You seem to be making the assumption that we'd increase the number of towers without also decreasing the power those towers are transmitting at. Well yeah, that obviously wouldn't work. But if you increase the number of towers and also decrease the ouput power of each tower, you can serve a lot more users at the same time. The individual bandwidth available to any particular user might not increase, but the total bandwidth certainly does.
If you ever forgot to charge it over-night, it was guaranteed to die around noon the next day.
Even at that extreme (and I think you have to admit the vast majority of phones are not at that extreme), you just have to remember to plug in the phone every night. A slight hassle, I suppose, but I wouldn't put it at the top of my list.
Information theory is going to tell you that the bandwidth through a single channel is limited. But if you have enough towers, then you can dedicate a nearly unlimited number of channels to every single user. Yes, there is still I suppose a theoretical limit, but it's nowhere near anything that we're currently seeing.
Charging is a pain in the ass? I guess I just don't see what's so hard about plugging your phone into the wall.
As for the phone not being mobile when charging, well, it is if it's charging in your car, and most of us have several hours every night when we're not mobile anyway.
I mean, OK, I'm sometimes too lazy to plug in my phone at night, but it's not really *that* big of a deal.
Cellphones are quickly changing into mobile PCs and not just phones
Unfortunately, the cell phone companies are using the Apple model, and not the Intel one. I'm all for cell phones becoming mobile PCs. But I don't want to be tied down to one vendor for the software that runs on those PCs.
I anticipate (article was more early marketing than real info)that users of this service will not be getting a high-resolution image on their cell phone
Well, yeah, almost certainly not. But that's more about the number of pixels in a cell phone screen, not about bandwidth. Verizon's 3G service has "typical speeds of 400-700 kbps". That's more than enough for the kind of resolution you get on a cell phone.
Can we expect ultra-high-resolution TV-style instant video eventually for everyone over a cellphone-style wireless network, or will it become more of a video-on-demand system where you chose ahead what you want to watch, then are notified when your show is available to watch? I wonder what the bandwidth will end up making plausible and simpler to provide.
Well, remember, if you have enough "towers", bandwidth is essentially unlimited. Of course, whether or not you still consider that "cellphone-style" is a matter of semantics. It'd really be more "wifi-style" if we need that many "towers".
Which makes me think - once people get to commonly learn video-on-demand or TIVO-style interfaces, which will be more popular?
I think there's a place for both, actually. Obviously on-demand is more convenient, but there's a place for a television series where everyone is watching at roughly the same time. People don't just watch the show for the content of the show, it's also something that we can talk about at work or school or whatever. And then of course there are live shows, including sporting events and news, but many "reality TV" shows have live events as well.
Slashdot Mirror
Where is 'innocent until proven guilty' these days?
Where it has always been, in the hands of the jury. The doctrine that you are considered innocent until proven guilty beyond a reasonable doubt does not imply that every single piece of evidence introduced into the case must be able to single handedly prove that guilt.
I can guess at why the prosecution would want to get this evidence in the record and why it would be admitted.
Sure, to prejudice the jury by introducing facts that they aren't going to understand.
if there is other evidence suggesting you have bad files and you have intentionally made those file unreadable, the tools you used to do that are possibly relevant.
OK, but what if there isn't evidence suggesting you intentionally made those files undreadable? "The court didn't say that police had unearthed any encrypted files"
Kind of like pointing out the defendant owned a shredder, there was huge pile of shredded paper by it, and the "smoking gun" documents are no where to be found.
In this case though, they seem to have merely pointed to the existence of the shredder in the first place. Of course, I haven't read the actual ruling, as I can't find it. But from what I read in the article it seems that erred in allowing the evidence. Whether or not that amounts to reversible error I won't even speculate, though.
This wasn't about an insanity defense, which is an affirmative defence which would have had to have been raised at the beginning of the trial. Rather, this is about the mens rea element of the law.
I know you've heard that "ignorance is no excuse" for the law, but that's a gross oversimplification. Rather than explain mens rea to you, I'll point you to the Wikipedia article, which does a better job of it than I can.
Don't exaggerate. This ruling wasn't that anyone with PGP has criminal intent. It was that a judge admitting the fact that the criminal had PGP into evidence wasn't reversible error. If pretty much every Linux distribution had GPG included, and you had such a Linux distribution, it would be quite easy for the defense to put this into evidence.
Locking your doors is a normal and irrelevant thing-- just like encrypting your data before your send it through the internet.
If all he had on his computer was a web browser with HTTPS, that'd be true. But he had PGP, which isn't very common at all. Should we trust a jury to make the right decision when given this evidence (along with presumably a defense showing just how many other reasons he could have had PGP)? I really don't know. I suppose it depends in part on how strongly the defense objected to the evidence before it was presented. Since he wasn't actually using the PGP to encrypt anything I would probably lean toward not admitting the evidence. But is this reversible error sufficient to overturn a ruling? I can't say it is without knowing the details of the case.
The particular reason for encrypting certain documents may be so that your little sister won't read your diary. You don't care if she reads your term paper, so it isn't encrypted.
Anything's possible. It's up to the jury to decide whether that's reasonable. Maybe the guy didn't even have a sister. If he did, and the sister regularly used the computer, then maybe this is something the defense could bring up.
TFA doesn't say that the man used encryption to cover up illegal activity, but that the very existence of PGP itself indicated criminal intent, which is ludicrous.
If that's true, then I'd suggest TFA is wrong. Her's what I read:
Without knowing more about the case, I can't say if I agree or not.
Apple can not do that anymore than anyone else can, the law explicitly prohibits it.
Can't do what, restrict people from distributing their software without permission? Sure they can.
You can't say that windows is as cross plaform as java, because windows programs don't run on java.
Huh?
The point was that wine doesn't provide cross platform for windows binaries, because you are tied to a subset of windows programs, and just x86 processors.
I believe that a program which runs on multiple operating systems is cross platform. It doesn't need to run on multiple architectures. But that's just semantics. Regardless of what you want to call it, what Java provides is not very useful.
You can do whatever you want with C++, but the fact is that you will have a hard time finding people doing good code for C++, and it's much easier to find good java coders.
If that's true it's pretty sad. But I'd rather spend the time and money finding someone who can code well in C++ than release a product in java just because I'm too cheap to build things right.
With Java, safe comes for free.
Safe comes at the cost of slow, bulky, and difficult to install.
You just keep repeating the same preconceptions that make you think java is useless, and I shouldn't try to keep you from doing it.
I'm not sure what it is I'm repeating, but if I'm repeating something it's because you haven't addressed it. Moreover, I can't prove that java is useless. In fact, I admit it has at least one use - applets and other sandboxed software. But as for the rest, I'm just not convinced.
I find Java is a very powerful and versatile laguage, with no measurable shortcomings compared to other commonly used languages, and lots of advantages as a platform.
The difference in speed is a measurable shortcoming. The difference in size of the entire package (including the interpreter) is a measurable shortcoming. And the difference in ease of installation (have to ensure that everyone has the interpreter, and have to ensure that the interpreter they have is compatible with the bytecode you have).
The fact that a bunch of people were forced to learn java in college in the early 2000s doesn't make up for these shortcomings. Not by a longshot. Anyone who is a decent coder can easily write safe and efficient code in C as easily as they can in Java. Just because C allows you to do something doesn't mean you have to do it.
Yes, it's already possible, but "if Macs ran on X86 chips, then running Windows software on Macs becomes a lot easier".
Who would buy a Mac when they could buy a Dell.
Well, for starters, anyone who currently buys a Mac when they can buy a Dell.
Does anyone seriously believe Microsoft would release Office for Mac OS X for Intel?
Umm, they already do, don't they?
For the n-th time, what would Apple have to gain?
The ability to have binary compatible software, for one thing.
The Mac would die the day the CPU would be the same as in a generic PC. Not from a architectural standpoint, I think they could make it happen, but marketingwise.
Now that's an interesting point of view. There probably are some people who buy Macs because of the chip. In my opinion it's not that big of a selling point, but maybe it is.
Chances are if you are running OS X, you probably paid for it at least once every HW upgrade.
And chances are if you are running anything but OS X, you probably paid for Windows at least once every HW upgrade, whether you use it or not.
Getting a PC without any OS on it and then installing a pirated copy is too hard for the average Joe. I think Apple's fears that people doing this will cause a significant decrease in revenues are misplaced.
I can't imagine why Apple would want to move towards x86 hardware
Probably the number one reason would be software compatibility. By moving to x86 hardware, and then customizing Wine, Apple could offer seemless integration of Windows software on OSX.
If apple goes up whenever this rumor comes out, I guess the shareholders think they should actually do it. How long until the shareholders vote in a board of directors who will follow this?
Anyway, I'm enjoying the INTC runup, which makes a lot more sense economically.
It's just Apple trying to get better terms/service from IBM (think Dell's "talks" with AMD)
Total speculation, but maybe.
It will be the death of Apple's hardware division
I don't see how using a different chip could possible kill Apple's hardware division. Could you explain? Note that using a different chip wouldn't mean that Apple would have to allow OEMs to distribute hardware with OSX pre-installed. They could, and given a high enough price for the software (to compensate for the lost profits from hardware), they should, but they wouldn't have to.
Apple will have a hard time supporting the myriad boards, chipsets, and peripherals of PCs
They wouldn't have to support anything they don't already support or that they make themselves. Again, using the Intel chip isn't the same as allowing clones. Those who want to install OSX on non-Apple built or sanctioned hardware do so at their own risk.
Piracy/sharing (pick your preferred new-speak term) will mean a revenue-less expansion of the install base
Maybe. This is the biggest worry. But I think Microsoft has shown that most people aren't going to go through the hassle of installing their own operating system just so they can save a few bucks. Installing an operating system is hard, and most people are going to keep whatever they get with their computer.
If PC users can just buy OS X86 for $99, they might give Mac a try. It wouldn't take that high conversion rate for OS software profits to easily replace hardware profits.
This is the key here. It doesn't matter if Apple is making money off software or hardware, as long as they're making money. It does lead to a different problem, though. Most likely the price point for OSX would have to be set a lot higher than the price for Windows in order for Apple to continue making the same amount of profits. With the prices separated out like that, people are going to realize exactly how much more OSX costs. Of course, Apple could make the switch to Intel chips and only sell the OS bundled with its own hardware. One advantage they'd retain would be in the relative ease of porting an application from one OS to another.
All the hardware interfaces would be 100% compatible with a pc. Even if Apple shipped custom firmware to restrict hardware used with the platform it would be trivial to reflash. Nope, once they go x86 Apple has no more firm a grip on the hardware than any other pc vendor.
I'm curious. Other than the motherboard/operating system/case, what hardware does Apple make serious money on? They definitely don't have a monopoly on hard drives for the Mac, and I'm pretty sure you don't need Apple's permission to make a PCI card. So how would Apple lose, exactly?
Yeah, people could download a CD off the Internet, reflash their BIOS, and install OSX on their PCs. But how many people are actually going to do that? How many people buy computers without OSes and then install a pirated copy of Windows? Not a whole lot, percentage-wise. So I don't think many people would do it for OSX either.
As long as Apple can stop people from reselling computers with OSX pre-installed, they'll still maintain their ability to charge people who use Macs. Sure, you can say Apple makes money off hardware and Microsoft makes money off software, but when most of the people buying Windows are getting it bundled with the hardware anyway, is there really a difference?
Very few people build their own computers. Most buy a box from their local computer store, or order it online from a company like Dell.
OK, but if you could buy the parts and build your own Mac, then Dell could buy the parts and build a Mac for you. Maybe not legally, though. Apple could presumably make it legal to build clones for personal use, but illegal to build them for resale.
Also, no one would run OSX on a standard PC. Just like no one runs BeOS, or ran OS/2.
OSX has a much bigger base of users than BeOS or OS/2. And unlike Be, Apple has a lot of experience delivering complete systems with the OS pre-installed.
An x86 Apple would probably be a proprietary Apple with an x86 in, and no one would care. Just like no one cares that there is a PowerPC in them today.
Most people would care, and use Windows. Some people would care, and use OSX. Some people would care, and use Linux. Some people, including me, would (still) buy the cheapest hardware without regard to what operating system is installed on it. Where Apple would make its biggest gain is with people who want to use OSX, but are forced to use Windows because they are tied to software which is only available on Windows. If Macs ran on X86 chips, then running Windows software on Macs becomes a lot easier - you'd use Wine, and not an emulator.
If they did it right, it'd probably be a win-win situation. Just keep a little bit of the Operating System in ROM, and don't let anyone else sell computers with the OS pre-installed. Sure, you'll get some hard-core geeks who will quickly copy the OS and offer install CDs on some P2P networks, but most people don't install their own OS just to save a few bucks. Meanwhile, you'll make the cost of supporting OSX in addition to Windows come down greatly, and you can probably sell a few OSX CDs to people who already own Intel hardware and are willing to pay to change their OS.
Wine is x86-only, and not all windows programs run on wine.
Cross-platform means it runs on multiple operating systems, not all operating systems. And yes, not all Windows programs run on Wine. Not all run on Java, either.
You are just talking about Sun JDK, but java is more than that.
Well, I was talking about the cross-platform aspect.
It's a language inherently somewhat slower than C++, but much safer, and in some specific cases, even faster than C++.
Anything that can be written in Java can be written as a faster program in C, and hence in C++. It might take more time to develop such a program, but anything you can do in Java you can do in C.
I was just trying to explain that such speed decrease is a packaging, distribution, or even a marketing problem, other than a technical problem.
Unless you're talking about compiled java, in which case you eliminate the only real advantage of java (compile once, run anywhere), the speed decrease in inherent in the very fact that java is an interpreted language.
When competing with native programs in any platform, Java doesn't need to be cross platform, because that's wha native means, non general purpose, but close tot he hardware. In those cases, you can compile with GCJ, and have a theoretically slightly slower binary than with C++, but actually a better binary, because it doesn't have the possibility of buffer overflows, and for all practical purposes doesn't have a speed difference with C++.
There are garbage collection libraries available for C++. One can easily use a memory allocation library which checks for buffer overflows, and prove quite easily that the code doesn't use any memory without going through the library. One generally doesn't do this, because it's relatively easy to code in a way which doesn't allow buffer overflows, but it is possible. In C++ you have the option. With Java, you're forced to do everything the Java way.
When you need cross platform, you use sun JDK, or the native JDK, with the startup delays you talk about, but in this case, it's not slower than anything, because nothing else is cross platform at the binary level, so it's xxx speed against zero speed, so it's still faster, because any positive speed is better than zero, period.
Byte code is not the same as the binary level, and any interpreted language can accomplish the same thing, so I don't see where you're saying that "nothing else" compares.
But realistically, eliminating the need to recompile is a rather narrow market, especially when you consider the memory and storage space issues inherent with java. For anything outside that narrow market, you'd be better off compiling once for each platform.
One market where Java is ahead of the others is in creating a sandboxed environment where you can run untrusted software. This is an example of a market where running compiled software is impossible. This is especially true for Windows. In unix, you can just set up a chrooted sandbox and run the compiled software in that, but Windows AFAIK doesn't support that. So yeah, if I need to run software in a sandboxed environment on Windows, Java would be my number one choice. But other than that one purpose, applets, I don't see the point of Java.
Maybe you should get a phone with a wall plug. That'd be a useful feature.
Bandwidth has nothing to do with the number of towers.
So if there was only one cell phone tower in the world, and everyone in the world had to share the same frequencies on that one cell phone tower, we'd still have the same bandwidth available?
Now consider what happens in the physical case of two towers broadcasting distinct signals on the same band -- we can think of each signal as a channel. Note that this would be required for your scheme to work. For simplicity, assume that they are broadcasting omnidirectionally.
Your simplification greatly reduces the usefulness of having multiple towers.
Fourier analysis (and experimental evidence) assure us that the signals are going to interfere with each other.
Sure, to some very very small extent, which drops off as the square of the distance from the tower. IOW, yes, a cell phone tower in Mexico interferes with a cell phone tower in New York City, but the interference is negligible. Now, reduce the wattage outputted by the tower, and move them closer together, and you get the same neglible interference.
You seem to be making the assumption that we'd increase the number of towers without also decreasing the power those towers are transmitting at. Well yeah, that obviously wouldn't work. But if you increase the number of towers and also decrease the ouput power of each tower, you can serve a lot more users at the same time. The individual bandwidth available to any particular user might not increase, but the total bandwidth certainly does.
If you ever forgot to charge it over-night, it was guaranteed to die around noon the next day.
Even at that extreme (and I think you have to admit the vast majority of phones are not at that extreme), you just have to remember to plug in the phone every night. A slight hassle, I suppose, but I wouldn't put it at the top of my list.
Information theory is going to tell you that the bandwidth through a single channel is limited. But if you have enough towers, then you can dedicate a nearly unlimited number of channels to every single user. Yes, there is still I suppose a theoretical limit, but it's nowhere near anything that we're currently seeing.
Charging is a pain in the ass? I guess I just don't see what's so hard about plugging your phone into the wall.
As for the phone not being mobile when charging, well, it is if it's charging in your car, and most of us have several hours every night when we're not mobile anyway.
I mean, OK, I'm sometimes too lazy to plug in my phone at night, but it's not really *that* big of a deal.
The climactic emotional moment, I swear to God, is a rip-off of Homer Simpson.
OK, I saw the movie, but I don't get it. Anyone care to explain?
Cellphones are quickly changing into mobile PCs and not just phones
Unfortunately, the cell phone companies are using the Apple model, and not the Intel one. I'm all for cell phones becoming mobile PCs. But I don't want to be tied down to one vendor for the software that runs on those PCs.
I anticipate (article was more early marketing than real info)that users of this service will not be getting a high-resolution image on their cell phone
Well, yeah, almost certainly not. But that's more about the number of pixels in a cell phone screen, not about bandwidth. Verizon's 3G service has "typical speeds of 400-700 kbps". That's more than enough for the kind of resolution you get on a cell phone.
Can we expect ultra-high-resolution TV-style instant video eventually for everyone over a cellphone-style wireless network, or will it become more of a video-on-demand system where you chose ahead what you want to watch, then are notified when your show is available to watch? I wonder what the bandwidth will end up making plausible and simpler to provide.
Well, remember, if you have enough "towers", bandwidth is essentially unlimited. Of course, whether or not you still consider that "cellphone-style" is a matter of semantics. It'd really be more "wifi-style" if we need that many "towers".
Which makes me think - once people get to commonly learn video-on-demand or TIVO-style interfaces, which will be more popular?
I think there's a place for both, actually. Obviously on-demand is more convenient, but there's a place for a television series where everyone is watching at roughly the same time. People don't just watch the show for the content of the show, it's also something that we can talk about at work or school or whatever. And then of course there are live shows, including sporting events and news, but many "reality TV" shows have live events as well.