The thing I don't understand is why any printer manufacturer would take such a route. Gutenprint (formerly known as gimp-print) is just a plugin system which is meant to integrate with something like CUPS to provide the actual printing service. Need support for another printer, you just write an appropriate plugin.
The idea (in Unix) is that your program sends something to the printer through the lp command, this is shipped off to the printing system (generally CUPS on Linux - is that also used on Mac OSX?) which handles network support transparently. The data then leaves CUPS and goes into gutenprint for processing before it hits the printer. It's substantially more work to write your own printing system from scratch, because you've got to provide the entire jigsaw rather than just one piece - or, as you have found out, provide enough of the jigsaw to claim support.
Don't know if it helps you now, but one of the side-effects of this is that with a properly configured Linux print server, you can run any printer which has a suitable Linux driver on the Linux box, send straight Postscript to the Linux box from your client and it will be turned into whatever language the printer expects by gutenprint. Of course, if the Linux printer driver doesn't support something which the printer can technically do (such as duplex), then you won't be able to do that from any of the clients either.
As I said in my grandfather post, if gimp-print doesn't have a driver for that printer, you're in trouble.
IMO, installing any printer which doesn't support PostScript and/or PCL is asking for trouble in a networked environment. Sooner or later I guarantee you'll come across a Windows PC which for some reason keeps on crashing when it tries to run something through the fancy 90MB printer-hardware-replacement driver and as likely as not it'll be some dirty great 200-page printout that the managing director's trying to do.
At least with a Postscript printer, you can say "stuff it, I'll run it through a generic postscript driver for now and reimage the PC later".
By "general purpose use", I meant "sticking in a PC, running OS_OF_CHOICE and APPLICATION_OF_CHOICE".
Which is exactly what doesn't happen with cellphones and portable electronics. Generally, the manufacturer and network has quite a bit of say over what runs on them so code can be developed specifically for them.
So, who picked this guy to be the successor to Jack Valenti who once famously said: "If you need a backup copy of a DVD you can go out and buy another one."
In the world occupied by the likes of Valenti, it's no big deal to go out and buy another copy of a DVD every time it gets damaged because compared to what his salary was, the price is fairly nominal.
Clearly nobody every argued back with "But what if it's no longer in production?" or "I have two small children; you expect me to replace every other DVD in my collection every week?!"
"DRM must be made to work without constricting consumers"
Isn't the point of DRM to constrict customers? The only way not to do so is to not have DRM.
Technically correct. But this is the MPAA and they've got an answer to everything.
In this case, their answer is for every fair-use they consider "reasonable", they'll license a product which can do it. Such as a licensed DVD-ripping box which allows you to rip your DVDs but stores the movies in some encrypted form so you can watch them fine but copying them back onto another DVD is made awkward by things like Macrovision (if you try recording through analogue) or encryption (if you try dumping the digital data directly).
The minor issue with this is that fair-use by definition is a use for the media, not the mechanism to enable such use. "Fair Use" which can only be exercised by buying another licensed piece of equipment isn't really fair use at all - it's "fair use provided we don't lose control", which strictly speaking isn't fair use at all. But it's probably close enough that anyone trying the "Your honour, I was exercising my right to fair use" argument can be shot down with "Why didn't you use a licensed product to do so?"
Seriously. It's already been brought up many times that finally, here's someone who's taking a stand and counter-suing the RIAA.
What did you expect the RIAA to do? Roll up in court and say to the judge "Actually, your honour, they're quite right, we're a bunch of misguided nutcases"?
Of course they're going to challenge the counterclaim - and they'll challenge it with everything they can think of because if they lose, suddenly there will be substantially more lawyers prepared to take on defense cases at a very good price (plus a percentage of any winnings in a counterclaim) in their next round of suing people.
Let's face it, nobody buys a mainframe unless they've already got a very specific use planned for it. It's not like an x86 server where it's cheap enough that you might think "We'll use it for X, but even if we don't, we'll use it for Y".
And the Cell isn't really intended for general-purpose use - it's far more appropriate to use it in a system where the code has been written and designed specifically for it.
What better market than one which is composed almost entirely of people with reasonably specific, defined needs?
In Australia we plant non-native pine trees for timber resulting in vast areas of land covered with a pine tree monoculture that is largely devoid of any other lifeforms (even the bugs refuse to live in those forests).
Maybe they don't like the smell of cheap disinfectant.
But very few comments from anyone in the UK. Let me explain, and hopefully redress the balance slightly.
The way the UK education system works is:
At around age 16, you take exams we call GCSE's. These are, in the big scheme of things, fairly straightforward. Most people will take around 10 subjects at GCSE level.
The next year (around age 17), you take AS-levels. Each AS-level is worth half an A-level. Most people will take about 6 subjects.
The year after, you take A-levels. Most people will take 3, though some will take 2 or 4. You don't choose new subjects - you generally carry on doing the things you did well in at AS-level.
Each A-level pass grade (A-E) gets you a certain number of points - obviously, higher grades=more points. AS-levels are worth half the number of points of their equivalent A-levels.
Universities set entrance requirements based primarily on points achieved at A/AS level. They can also demand you do a particular subject for some courses, but that's by no means certain and varies from university to university. Some of the top universities also demand you take another entrance exam.
All of which is well and good. But what I haven't explained yet is the real fuck-up.
There is absolutely no comparison whatsoever between "amount of work required to get a good grade at GCSE" and "amount of work required to get a good grade at A-level". The gap between the two is absolutely huge.
Seriously. Provided you're of reasonable intelligence, you could mess around for 2 years (as I did) at GCSE level and still get reasonably good grades.
Try doing that at A-level and you will almost certainly screw up with a vengeance. This is particularly true in science-y subjects like Maths and Chemistry.
Thing is, a lot of 16 year olds don't take these things seriously. Your teachers can say "You're going to have to pay more attention at A-level" until they're blue in the face, but a lot of people won't really take that on board until it's far too late. So you either drop that A-level in Maths or you fail it.
Now politics comes into play. No government wants to admit that the schoolchildren of the day are failing. But it's a government body which sets the exams. So every year, the exams are a little bit easier than they were the previous year. Not substantially - as a pupil, you probably wouldn't notice unless you were given an exam from 15 years earlier. The unversities notice, though, and they've taken a number of approaches. Some demand an extra entrance exam, others do remedial courses. Such remedial courses have existed for ages - they're called "foundation" courses and are generally a year long. But they are generally only offered for some degree courses, and they seldom get this level of publicity
Plenty of people make that argument, and have been doing so for years.
Yet even today I can point you at a few real business applications which could really benefit from more power. I have no doubt whatsoever that in a few years time, they'll be OK on anything you're likely to run at them, but another troop of applications will have come along which require more power.
However now that chips are going in the direction of multicore rather than ever higher clockspeeds, it means that development methodologies have to shift focus in order to take full advantage of it. Not every app has yet done so, not by a long way.
Case in point: A business application my boss recently bought. Client/server app, with most of the intelligence in the client. They recommended at a minimum a Pentium 4 4GHz. Did such a thing even exist?
With proprietary solutions, when things go wrong with the product or service, there is a firm or product to point a finger to to go fix it and correct it for us, or else we can take our money elsewhere.
The issue they tend not to be aware of is - in much of the world that's dead easy with commodity stuff like electricity or telephone service or even the building the company's leasing.
You ever tried doing that with a piece of software? Even if you've just spend thousands of (insert currency unit here) on it, it's far from uncommon to find vendors who don't really want to support the product, but at the same time if you've just spent so much money on a product, you don't really want to throw it away and start again from scratch.
The management style we're discussing treats more-or-less all computing as a commodity - an item where one product is more or less indistinguishable from any other, therefore it should be easy to get someone to sell it to you, sort out any problems and take your business elsewhere should the vendor not be up to scratch.
That's true and it broadly works with x86 hardware. It certainly isn't the case with software.
But, open with lots of choices wins in the long run every time because it gives ownership of IT to the companies that use it instead of the companies the produce it.
Lots of companies are quite happy with limited ownership of their own IT systems - it means that no matter what goes wrong, there's no politics, no messing about. Just one number to call to raise a fault, and an account manager if things get really out of hand. Were that not the case, there would be no such thing as outsourcing.
I'm not going to get into the merits of how well that works in the real world - myself, I'm not convinced - but nevertheless, plenty of businesses are. At one point, my manager was thinking of outsourcing all our IT - and we're a software development firm.
The government shouldn't tell companies what to charge for their products, but they should ensure that companies can compete freely and make money.
But when you've got a monopolist like Microsoft, it's very difficult for another company to come in and compete freely.
Seeing as about all a company has are products which it exchanges for a certain amount of money, if the government wants to interfere with a company in order to foster competition in a market, about the only thing they CAN do is mess with either the product or its price.
the EU has no god-given-right to use Microsoft Software however they want either.
Technically, everyone in the world has a god-given right to do whatever they damn well please provided it doesn't break the laws laid down in the bible. (Assuming you believe in the bible, but if you don't believe in some sort of religion then there's no such thing as a god-given right.....)
However, thousands of years ago it became apparent that it was far more productive for humanity if there was a set of ground rules saying what people can and can't do, together with some sort of infrastructure for enforcing these rules.
As society has become more complex, so have these ground rules. Today, we call them "laws" and we have an entire segment of society called "government" whose job it is to look at society and keep the "laws" in step.
Break them up into OS and apps and have done with it; we all know that's what's needed.
I disagree.
While the apps business would no longer have the upper hand over other apps companies (they wouldn't necessarily know in advance what the OS business is doing), you'd wind up with two companies, both of which have monopolies in their respective fields (if you consider Office to be probably one of, if not the most important app within MS).
Far more effective would have been two OS companies, both starting with a copy of the source-code for Windows and able to take it where they pleased. Do something similar with apps.
(because 95% of clients are windows and only windows server can serve them)
Ah. Good point.
Samba's fallen badly behind (you can run a Windows domain entirely with Samba, provided you don't mind it being an NT4-style domain, so no group policy stuff) and while they are trying to catch up, it's clearly an uphill battle.
Opening the various bits which still aren't perfectly reverse engineered would have Samba serving AD domains very quickly indeed.
Way back when, there used to be a real benefit to upgrading your 133MHz PC to 200MHz and it was easy to do so just by changing the CPU.
TBH, these days, for general desktop use I don't think that benefit's there any more. If you want to see a real benefit, you're best off replacing the CPU with something drastically faster. This may well involve a new motherboard and possibly new memory.
Alternatively, you upgrade the more sensible way - look at your computer needs, look to see what's causing a bottleneck currently and upgrade that. Much more cost-effective than just replacing a CPU and hoping you see a benefit.
I have seen a lot of this shocking belief : "If it was not secure, computer people would tell us so"
We have been for several years now. However it's rather difficult to get the mainstream media to pay attention, because if voter turnout is low, it follows that interest in the election is low. And most news outlets consider technology stuff to be fairly uninteresting anyway.
So interest in "how secure is the computerised voting system?" (and therefore "how worthwhile is it dedicating some time in a mainstream news broadcast?") is considered by the media to be as near zero as makes no odds.
This is a problem for a company whose entire business model for the last 20 years has been based on double-digit year growth.
TBH, I think that's part of the reason Vista requires so many people to upgrade their hardware. Not because there's any technical need for it to be such a resource hog, but because Microsoft know full well most of their market will be the OEM market so by making people say "Ooh! Shiny! Must buy a new computer!" they won't have to worry about people being intimidated by the thought of installing an OS themselves.
A bit like how people are buying next-gen consoles even though it's far from clear which will have the strongest games catalogue by this time next year.
If you just went and compiled the code I wrote without thinking carefully about what you were doing, you have a serious security issue which you really ought to resolve;)
All joking aside, it's a basic forkbomb. Quite a few unixes and clones (clearly not FreeBSD) will just keep on generating processes until such time as the process table fills up, which takes a fraction of a second. And once the process table is full, no more processes can be started - you can't even log in because even if the logon process is running, once it's authenticated you it will try and execute your shell and fail.
Someone upthread executed it with a ulimit - yes, that will prevent it from making your system unusable but that's a piece of userland configuration. The point I was making is that unless userland is appropriately configured (something which is omitted surprisingly often), it's quite easy to render a computer next to useless without crashing it as such, even from userland.
Drivers can, and bugs in the OS can. User-run programs can only (accidentally) trigger one of those... in which case, that's a DoS exploit in the system.
As far as the end user is concerned, what's the practical difference between a BSOD and a resource hog which causes the GUI (but not any other services which aren't immediately visible to the user, eg. printing, networking) to lock up?
The possibility that Quicktime itself is causing the BSOD is infinitesimally next to zero - no userspace application should bring the OS down directly. Quicktime obviously is triggering the BSOD, but there isn't absolutely any "strong possibility that the issue is being caused by Apple software". The issue is most likely bad drivers - and the fact that the device uses a custom tablet isn't of any help. While Apple probably could work around the problem, they are the last party he should ask for it - The first, and most likely to get his problem resolved is Toshiba. Even if they can't to anything about it directly (i.e., the problem is on Vista), they should at least be able to put some pressure on Microsoft to get it done.
Ah, bless.
I bet you that the people on Toshiba's tech support line don't have anything like this in their script. So they'll resort to the closest thing they can find - if the computer's crashing, reinstall Windows. If it still crashes, it's a hardware.... oh, it only crashes when you play back things in Quicktime? Sorry, we don't support Quicktime (click).
I have yet to see the idea that a driver provided on the system could be the cause of such crashes (and therefore merits being escalated to the team maintaining the driver) making it into a tech support script.
Of course, what will happen now is someone will relate a tale of a tech support issue they raised with Toshiba which did get escalated to an appropriate team and an updated driver was made available a week later.
Slashdot Mirror
The thing I don't understand is why any printer manufacturer would take such a route. Gutenprint (formerly known as gimp-print) is just a plugin system which is meant to integrate with something like CUPS to provide the actual printing service. Need support for another printer, you just write an appropriate plugin.
The idea (in Unix) is that your program sends something to the printer through the lp command, this is shipped off to the printing system (generally CUPS on Linux - is that also used on Mac OSX?) which handles network support transparently. The data then leaves CUPS and goes into gutenprint for processing before it hits the printer. It's substantially more work to write your own printing system from scratch, because you've got to provide the entire jigsaw rather than just one piece - or, as you have found out, provide enough of the jigsaw to claim support.
Don't know if it helps you now, but one of the side-effects of this is that with a properly configured Linux print server, you can run any printer which has a suitable Linux driver on the Linux box, send straight Postscript to the Linux box from your client and it will be turned into whatever language the printer expects by gutenprint. Of course, if the Linux printer driver doesn't support something which the printer can technically do (such as duplex), then you won't be able to do that from any of the clients either.
As I said in my grandfather post, if gimp-print doesn't have a driver for that printer, you're in trouble.
IMO, installing any printer which doesn't support PostScript and/or PCL is asking for trouble in a networked environment. Sooner or later I guarantee you'll come across a Windows PC which for some reason keeps on crashing when it tries to run something through the fancy 90MB printer-hardware-replacement driver and as likely as not it'll be some dirty great 200-page printout that the managing director's trying to do.
At least with a Postscript printer, you can say "stuff it, I'll run it through a generic postscript driver for now and reimage the PC later".
By "general purpose use", I meant "sticking in a PC, running OS_OF_CHOICE and APPLICATION_OF_CHOICE".
Which is exactly what doesn't happen with cellphones and portable electronics. Generally, the manufacturer and network has quite a bit of say over what runs on them so code can be developed specifically for them.
So, who picked this guy to be the successor to Jack Valenti who once famously said: "If you need a backup copy of a DVD you can go out and buy another one."
In the world occupied by the likes of Valenti, it's no big deal to go out and buy another copy of a DVD every time it gets damaged because compared to what his salary was, the price is fairly nominal.
Clearly nobody every argued back with "But what if it's no longer in production?" or "I have two small children; you expect me to replace every other DVD in my collection every week?!"
"DRM must be made to work without constricting consumers"
Isn't the point of DRM to constrict customers? The only way not to do so is to not have DRM.
Technically correct. But this is the MPAA and they've got an answer to everything.
In this case, their answer is for every fair-use they consider "reasonable", they'll license a product which can do it. Such as a licensed DVD-ripping box which allows you to rip your DVDs but stores the movies in some encrypted form so you can watch them fine but copying them back onto another DVD is made awkward by things like Macrovision (if you try recording through analogue) or encryption (if you try dumping the digital data directly).
The minor issue with this is that fair-use by definition is a use for the media, not the mechanism to enable such use. "Fair Use" which can only be exercised by buying another licensed piece of equipment isn't really fair use at all - it's "fair use provided we don't lose control", which strictly speaking isn't fair use at all. But it's probably close enough that anyone trying the "Your honour, I was exercising my right to fair use" argument can be shot down with "Why didn't you use a licensed product to do so?"
Seriously. It's already been brought up many times that finally, here's someone who's taking a stand and counter-suing the RIAA.
What did you expect the RIAA to do? Roll up in court and say to the judge "Actually, your honour, they're quite right, we're a bunch of misguided nutcases"?
Of course they're going to challenge the counterclaim - and they'll challenge it with everything they can think of because if they lose, suddenly there will be substantially more lawyers prepared to take on defense cases at a very good price (plus a percentage of any winnings in a counterclaim) in their next round of suing people.
Let's face it, nobody buys a mainframe unless they've already got a very specific use planned for it. It's not like an x86 server where it's cheap enough that you might think "We'll use it for X, but even if we don't, we'll use it for Y".
And the Cell isn't really intended for general-purpose use - it's far more appropriate to use it in a system where the code has been written and designed specifically for it.
What better market than one which is composed almost entirely of people with reasonably specific, defined needs?
In Australia we plant non-native pine trees for timber resulting in vast areas of land covered with a pine tree monoculture that is largely devoid of any other lifeforms (even the bugs refuse to live in those forests).
Maybe they don't like the smell of cheap disinfectant.
But very few comments from anyone in the UK. Let me explain, and hopefully redress the balance slightly.
The way the UK education system works is:
At around age 16, you take exams we call GCSE's. These are, in the big scheme of things, fairly straightforward. Most people will take around 10 subjects at GCSE level.
The next year (around age 17), you take AS-levels. Each AS-level is worth half an A-level. Most people will take about 6 subjects.
The year after, you take A-levels. Most people will take 3, though some will take 2 or 4. You don't choose new subjects - you generally carry on doing the things you did well in at AS-level.
Each A-level pass grade (A-E) gets you a certain number of points - obviously, higher grades=more points. AS-levels are worth half the number of points of their equivalent A-levels.
Universities set entrance requirements based primarily on points achieved at A/AS level. They can also demand you do a particular subject for some courses, but that's by no means certain and varies from university to university. Some of the top universities also demand you take another entrance exam.
All of which is well and good. But what I haven't explained yet is the real fuck-up.
There is absolutely no comparison whatsoever between "amount of work required to get a good grade at GCSE" and "amount of work required to get a good grade at A-level". The gap between the two is absolutely huge.
Seriously. Provided you're of reasonable intelligence, you could mess around for 2 years (as I did) at GCSE level and still get reasonably good grades.
Try doing that at A-level and you will almost certainly screw up with a vengeance. This is particularly true in science-y subjects like Maths and Chemistry.
Thing is, a lot of 16 year olds don't take these things seriously. Your teachers can say "You're going to have to pay more attention at A-level" until they're blue in the face, but a lot of people won't really take that on board until it's far too late. So you either drop that A-level in Maths or you fail it.
Now politics comes into play. No government wants to admit that the schoolchildren of the day are failing. But it's a government body which sets the exams. So every year, the exams are a little bit easier than they were the previous year. Not substantially - as a pupil, you probably wouldn't notice unless you were given an exam from 15 years earlier. The unversities notice, though, and they've taken a number of approaches. Some demand an extra entrance exam, others do remedial courses. Such remedial courses have existed for ages - they're called "foundation" courses and are generally a year long. But they are generally only offered for some degree courses, and they seldom get this level of publicity
Plenty of people make that argument, and have been doing so for years.
Yet even today I can point you at a few real business applications which could really benefit from more power. I have no doubt whatsoever that in a few years time, they'll be OK on anything you're likely to run at them, but another troop of applications will have come along which require more power.
However now that chips are going in the direction of multicore rather than ever higher clockspeeds, it means that development methodologies have to shift focus in order to take full advantage of it. Not every app has yet done so, not by a long way.
Case in point: A business application my boss recently bought. Client/server app, with most of the intelligence in the client. They recommended at a minimum a Pentium 4 4GHz. Did such a thing even exist?
With proprietary solutions, when things go wrong with the product or service, there is a firm or product to point a finger to to go fix it and correct it for us, or else we can take our money elsewhere.
The issue they tend not to be aware of is - in much of the world that's dead easy with commodity stuff like electricity or telephone service or even the building the company's leasing.
You ever tried doing that with a piece of software? Even if you've just spend thousands of (insert currency unit here) on it, it's far from uncommon to find vendors who don't really want to support the product, but at the same time if you've just spent so much money on a product, you don't really want to throw it away and start again from scratch.
The management style we're discussing treats more-or-less all computing as a commodity - an item where one product is more or less indistinguishable from any other, therefore it should be easy to get someone to sell it to you, sort out any problems and take your business elsewhere should the vendor not be up to scratch.
That's true and it broadly works with x86 hardware. It certainly isn't the case with software.
But, open with lots of choices wins in the long run every time because it gives ownership of IT to the companies that use it instead of the companies the produce it.
Lots of companies are quite happy with limited ownership of their own IT systems - it means that no matter what goes wrong, there's no politics, no messing about. Just one number to call to raise a fault, and an account manager if things get really out of hand. Were that not the case, there would be no such thing as outsourcing.
I'm not going to get into the merits of how well that works in the real world - myself, I'm not convinced - but nevertheless, plenty of businesses are. At one point, my manager was thinking of outsourcing all our IT - and we're a software development firm.
The president of the RIAA has a monocle and stokes a cat all day.
Really? How do you stoke a cat, anyway?
Imagine a beowulf cluster of these!
The government shouldn't tell companies what to charge for their products, but they should ensure that companies can compete freely and make money.
But when you've got a monopolist like Microsoft, it's very difficult for another company to come in and compete freely.
Seeing as about all a company has are products which it exchanges for a certain amount of money, if the government wants to interfere with a company in order to foster competition in a market, about the only thing they CAN do is mess with either the product or its price.
the EU has no god-given-right to use Microsoft Software however they want either.
Technically, everyone in the world has a god-given right to do whatever they damn well please provided it doesn't break the laws laid down in the bible. (Assuming you believe in the bible, but if you don't believe in some sort of religion then there's no such thing as a god-given right.....)
However, thousands of years ago it became apparent that it was far more productive for humanity if there was a set of ground rules saying what people can and can't do, together with some sort of infrastructure for enforcing these rules.
As society has become more complex, so have these ground rules. Today, we call them "laws" and we have an entire segment of society called "government" whose job it is to look at society and keep the "laws" in step.
Break them up into OS and apps and have done with it; we all know that's what's needed.
I disagree.
While the apps business would no longer have the upper hand over other apps companies (they wouldn't necessarily know in advance what the OS business is doing), you'd wind up with two companies, both of which have monopolies in their respective fields (if you consider Office to be probably one of, if not the most important app within MS).
Far more effective would have been two OS companies, both starting with a copy of the source-code for Windows and able to take it where they pleased. Do something similar with apps.
(because 95% of clients are windows and only windows server can serve them)
Ah. Good point.
Samba's fallen badly behind (you can run a Windows domain entirely with Samba, provided you don't mind it being an NT4-style domain, so no group policy stuff) and while they are trying to catch up, it's clearly an uphill battle.
Opening the various bits which still aren't perfectly reverse engineered would have Samba serving AD domains very quickly indeed.
Way back when, there used to be a real benefit to upgrading your 133MHz PC to 200MHz and it was easy to do so just by changing the CPU.
TBH, these days, for general desktop use I don't think that benefit's there any more. If you want to see a real benefit, you're best off replacing the CPU with something drastically faster. This may well involve a new motherboard and possibly new memory.
Alternatively, you upgrade the more sensible way - look at your computer needs, look to see what's causing a bottleneck currently and upgrade that. Much more cost-effective than just replacing a CPU and hoping you see a benefit.
I have seen a lot of this shocking belief : "If it was not secure, computer people would tell us so"
We have been for several years now. However it's rather difficult to get the mainstream media to pay attention, because if voter turnout is low, it follows that interest in the election is low. And most news outlets consider technology stuff to be fairly uninteresting anyway.
So interest in "how secure is the computerised voting system?" (and therefore "how worthwhile is it dedicating some time in a mainstream news broadcast?") is considered by the media to be as near zero as makes no odds.
This is a problem for a company whose entire business model for the last 20 years has been based on double-digit year growth.
TBH, I think that's part of the reason Vista requires so many people to upgrade their hardware. Not because there's any technical need for it to be such a resource hog, but because Microsoft know full well most of their market will be the OEM market so by making people say "Ooh! Shiny! Must buy a new computer!" they won't have to worry about people being intimidated by the thought of installing an OS themselves.
A bit like how people are buying next-gen consoles even though it's far from clear which will have the strongest games catalogue by this time next year.
If you just went and compiled the code I wrote without thinking carefully about what you were doing, you have a serious security issue which you really ought to resolve ;)
All joking aside, it's a basic forkbomb. Quite a few unixes and clones (clearly not FreeBSD) will just keep on generating processes until such time as the process table fills up, which takes a fraction of a second. And once the process table is full, no more processes can be started - you can't even log in because even if the logon process is running, once it's authenticated you it will try and execute your shell and fail.
Someone upthread executed it with a ulimit - yes, that will prevent it from making your system unusable but that's a piece of userland configuration. The point I was making is that unless userland is appropriately configured (something which is omitted surprisingly often), it's quite easy to render a computer next to useless without crashing it as such, even from userland.
Drivers can, and bugs in the OS can. User-run programs can only (accidentally) trigger one of those... in which case, that's a DoS exploit in the system.
As far as the end user is concerned, what's the practical difference between a BSOD and a resource hog which causes the GUI (but not any other services which aren't immediately visible to the user, eg. printing, networking) to lock up?
The possibility that Quicktime itself is causing the BSOD is infinitesimally next to zero - no userspace application should bring the OS down directly. Quicktime obviously is triggering the BSOD, but there isn't absolutely any "strong possibility that the issue is being caused by Apple software". The issue is most likely bad drivers - and the fact that the device uses a custom tablet isn't of any help. While Apple probably could work around the problem, they are the last party he should ask for it - The first, and most likely to get his problem resolved is Toshiba. Even if they can't to anything about it directly (i.e., the problem is on Vista), they should at least be able to put some pressure on Microsoft to get it done.
Ah, bless.
I bet you that the people on Toshiba's tech support line don't have anything like this in their script. So they'll resort to the closest thing they can find - if the computer's crashing, reinstall Windows. If it still crashes, it's a hardware.... oh, it only crashes when you play back things in Quicktime? Sorry, we don't support Quicktime (click).
I have yet to see the idea that a driver provided on the system could be the cause of such crashes (and therefore merits being escalated to the team maintaining the driver) making it into a tech support script.
Of course, what will happen now is someone will relate a tale of a tech support issue they raised with Toshiba which did get escalated to an appropriate team and an updated driver was made available a week later.