Since there has been much discussion of the "Linux on the Desktop" issue, I feel that the Kiosk framework will give KDE a real edge!
This is really what I miss when I try putting Linux boxes in an environment with computer illiterate users wanting to poke around. They try fiddling with the settings just as they do on the Windows boxes. Their fiddling around has been great for me as a admin since I've gotten a great argument for upgrading to later (more lockable) windows versions, thus not having to cope with the notoriously unsafe, crashing, generaly sucking Win9x boxes. Now I run Win2k locked down so that they hardly may move the mouse and I long for the day when I can get them to run Linux boxes without letting them fiddle around and come crying about some "lost icons" or something else.
Re:Computer Nertworks - 4th edition
on
The End Of Minix?
·
· Score: 2
The lase quote shows what both AST and MINIX are all about. They are academic tools (I do not say toys). This means that they are written in a very strict fashion (AST stuck with his microkernel structure, even when it didn't make sence), and is easy to modify (due to the microkernel design).
From an academic point of view Linux is not interesting. I'm not a kernel hacker, but I know that much of what is implemented in the Linux kernel has been done before. This does not make Linux a bad OS, it makes it a professional tool.
The difference between an academic tool (at) and a professional tool (pt) is that a pt has to be working all the time (thus we have stable kernel releases) because the average user do not want to fiddle with the kernel, but use the OS to be productive, i.e. run applications. On the other side, an at is meant to be used in education. Thus more work is put into the ease of modifying/understanding/porting the kernel, than the actual work of being able to do something useful with the resulting system. Sure MINIX could be used as a proper OS, but that was never the main intension, just a proof of people wanting to run UNIX on their 286s back then.
"But it does show that redheads need more anethetic on average, that's what correlation is."
Well, that information isn't very useful, unless your an economist trying to asses how much anatheic that will be consumed during the next budget period or something...
This proves that correlation studies can be used to find many results. You can find that statistically, red-heads feel pain easier, but this does not say anything about any single individual.
There is a reason to why one says "lies, damn lies and statistics. Correlation has been used to "prove" many things such as racist ideas (superiority due to colour), intelligence from weight etc. A good correlation between two parameters does *not* prove that they are connected!
No, I'm just saying that the indication of life that they use is not very exact. I'd even do so far as to calling it speculative. I'm not an astronomer nor a biologist, but I know that 'mysterious swirling patches' occur in nature from personal observation.
Technically I agree with you, but it is up to the court to judge if it can be concidered proven that the person has been there. I believe that it is one piece in a large number of circumstances that makes it *very* likely that the person has been at the location at the time.
It seems that half of the comments are from people who has not read the article!
The article talks of a radar system based on the reflected waves from mobile phones. I have a number of problems with this:
* The problem is huge, as each signal emitter is mobile, and thus the signal processing needed to filter out the source of each signal-bounce must be huge.
* As the number of signal emitters are variable in the vicinity of each reciever, this make the signal processing even more complex.
* They claim to being able to put all this in a laptop sized device.
This would not be so controversial if it was a simple cell phone tracing system, as they allready exist. In Sweden, one of the major competators even offer a 'locate' service, allowing other users to locate a phone. This service can be turned on and off from the located phone by sending SMSs. Even when turned off, the phone can still be located, all you block is the ability to get a position on another phone. This can, and has been used by the police to, for example, prove that a certain person has been at a certain location at a certain time.
Get this guy to work with your applications, and suddenly your word processor turns into an adventure game, you surface plots fill with small creatures fighting and, well, your internet browser turns into a pacman clone.
I'd say that this shows that todays game engines are pretty flexible, to say the least. How about hacking Quake into a Pacman clone, imagine your own mirror image: big, yellow and round with a mouth covering 50% of your body. And wouldn't it be nicer to get hunted by yellow, pink and blue ghosts instead of really scary corpses and zombies...
I'd say that there are lots of other, more plaussible, explanations to 'mysterious swirling patches' on a planet surface.
But, hey, the sientisist will get a headline or two, and perhaps even a few dollars to spend. I'm just saying that there are reasons to stretch the reality just a bit sometimes. Often these reasons are political or economical. In this case I'd have to go for the latter.
If a RISC is a proper RISC there are probably no more than one or two ways to do an operation. Since all similair instructions takes an equal amount of time there is no need to optimize. That is what I would call, really easy to optimize, i.e. no need too!
You are right that the moder x86 implementations are RISCs with a translation layer around them (except Crusoe which is a VLIW with software translation - much cooler 8P ). Now just imagine if we could get direct access to those highly optimized RISC cores instead of having to code in x86 machine code.
If we're going to stick to the x86 we still do not want to add complexity. I also tried to point out how easy it would be to move to a new architecture.
As you must add complexity I do not think that it would be "quick and easy". It takes huge resources in both time and equipment to verify the timing of a new chip, so these kind of changes (fundamental changes to the way registers are accessed) are expensive and hard since you also need to implement many new hardware solutions and verify the functionality (not only the timing!)
Ok, he realizes that the x86 architecture is flawed. One of the most limiting problems is the lack of general purpose registers (GPR), so he adds more complexity to an allready over-complex solution to solve this problem. All I have to say to this is: when will you see that the solution is as simple as switching architecture!
As most code today is written in higher level languages (C/C++, Java, etc.) all it takes is a recompile and perhaps some patching and adaptations to small peculiarities. The Linux kernel is a proof of this concept, a highly complex piece of code portable to several platforms with a huge part of the code folly portable and shareable. This means that it is not hard to change architecture!
If the main competition and its money would move from the x86 to a RISC architecure (why not Alpha, MIPS, SPARC or PPC) I'm sure that the gap in performance per penny would go away pretty soon. RISCs have several advantages, but the biggest (IMHO) is the simplicity: no akward rules (non-GP registers), no special case instructions, easy to pipeline, easy to understand and easy to optimize code for (since the instruction set is smaller).
And to return to the original article. Please do not introduce more complexity. What we need is simple, beautiful designs, those are the ones that one can make go *really* fast.
Does anyone know anything about the security problems this kind of device can cause? How easy is it to sniff out passwords etc from bluetooth and how easy is it to trick the drive into thinking that someone else is the owner?
I understand it as they have a choice: either they pay (or make another kind of settlement) or they may face the problem of not being prohibited from selling any.
Still, no-one holds a patent to VLIW in it self. It must be a patent of some implementation detail that is now covered in the detail.
EPIC is imply Intel sales lingo for VLIW.
The article fails to mention any technical details concerning the patent, but it sounds as if it has something to do with Intel's EPIC (which basicly is VLIW).
I'd say that Intel will have to pay, because if Intergraph stops the Itanium series of processors, it will mean an even bigger loss of money, and perhaps give AMD time to get a market advantage with the Hammer.
Ok, there is no 'local oil price', since if the oil from the Gulf would be stopped by a war the non-US world would need oil from somewhere else (pehaps even the US) and prices would go up (since the non-US world would be prepared to pay a higher price).
As for you arguing that since the US is not affected it is OK to attack Iraq is *very* short sighted. I wish that Bush (and his followers) would see that their *allies* (within NATO) do not want this (just ask France), and their friends (through Partnership for Piece) do not want it either (just ask Russia). Never in the history of UN, a war, just to be on the safe side, has been sanctioned, and I hope it never will.
The September 11th attack was a cruel terrorist attack on civians, but the US must see why they are picked as the target. Both the current Bagdad regime and the Talibans are (at least partially) the creations of CIA. The Talibans fought the USSR and were supported by the CIA, as was Saddam was supported in the war agains Iran as Iraq was deemed to be a smaller threat. By these kind of operations the US create instability in other regions of the world, and now some fanatics want to bite back. An important note: I do *not* support any terrorist activities, I'm just saying that there is a reason to why people become terrorists (desperation, lack of influense, abuse, etc) and maybe one can try to work in that end instead of bombing everyone not liking you (which leads to more people not liking you).
It is good that someone tries to chart this problem. At least it makes big corporations aware of the problem with wireless systems and the security issues associated with them.
I like the idea of wireless internet access everywhere, but not though stealing bandwidth of some business with bad security. I feel very bad for the companies being hacked and abused because of the bad security of the wireless solutions they use.
It surprises me that no-one thought of this before the technology was launched.
1) M$ wants to make money from games and selling expensive SDKs.
2) M$ introduces a console called XBox constructed from a PC with some signing stuff in HW.
3) XBox gets modded fairly quickly.
4) XBox can now run Linux.
5) M$ sees a potential threat divided in two parts a. one can copy games, b. one can use the XBox to pull M$'s legg (running Linux on it).
6) M$ sues the hell out off anyone getting to close.
7) M$ ends up with a huge pile of money!
Please excuse me ranting, but I get so tired of their lame attempts to introduce signing. Use a custom CPU with on-chip signing, a motherboard without any standard devices etc. and it will become harder. They could not for a minute have thought that a PC based console would be left unhacked.
As for custom conponents being more expensive - the *big* money can be found in games, not the actual hw. If they were sure to sell games they could just give the thing away!
It is really silly that you need to opt-out to try to get away (and you still get calls even then). If there would be an opt-in it would probably kill off the whole telemarketing business, which I feel would be a good thing.
This problem is just as bad as spam, fax-spamming and all other broad adverticement methods. I wish that one day you would not get more ads than you ask for (i.e. 2-5% of todays load of shit).
For you people living in Sweden I would like to mention the Nix register, it is the Swedish opt-out register.
Slashdot Mirror
Since there has been much discussion of the "Linux on the Desktop" issue, I feel that the Kiosk framework will give KDE a real edge!
This is really what I miss when I try putting Linux boxes in an environment with computer illiterate users wanting to poke around. They try fiddling with the settings just as they do on the Windows boxes. Their fiddling around has been great for me as a admin since I've gotten a great argument for upgrading to later (more lockable) windows versions, thus not having to cope with the notoriously unsafe, crashing, generaly sucking Win9x boxes. Now I run Win2k locked down so that they hardly may move the mouse and I long for the day when I can get them to run Linux boxes without letting them fiddle around and come crying about some "lost icons" or something else.
The lase quote shows what both AST and MINIX are all about. They are academic tools (I do not say toys). This means that they are written in a very strict fashion (AST stuck with his microkernel structure, even when it didn't make sence), and is easy to modify (due to the microkernel design).
From an academic point of view Linux is not interesting. I'm not a kernel hacker, but I know that much of what is implemented in the Linux kernel has been done before. This does not make Linux a bad OS, it makes it a professional tool.
The difference between an academic tool (at) and a professional tool (pt) is that a pt has to be working all the time (thus we have stable kernel releases) because the average user do not want to fiddle with the kernel, but use the OS to be productive, i.e. run applications. On the other side, an at is meant to be used in education. Thus more work is put into the ease of modifying/understanding/porting the kernel, than the actual work of being able to do something useful with the resulting system. Sure MINIX could be used as a proper OS, but that was never the main intension, just a proof of people wanting to run UNIX on their 286s back then.
"But it does show that redheads need more anethetic on average, that's what correlation is."
Well, that information isn't very useful, unless your an economist trying to asses how much anatheic that will be consumed during the next budget period or something...
This proves that correlation studies can be used to find many results. You can find that statistically, red-heads feel pain easier, but this does not say anything about any single individual.
There is a reason to why one says "lies, damn lies and statistics. Correlation has been used to "prove" many things such as racist ideas (superiority due to colour), intelligence from weight etc. A good correlation between two parameters does *not* prove that they are connected!
I'm sorry, my fault!
I hope that I haven't caused to much inconvinience. But you must admit, it does look good!
No, I'm just saying that the indication of life that they use is not very exact. I'd even do so far as to calling it speculative. I'm not an astronomer nor a biologist, but I know that 'mysterious swirling patches' occur in nature from personal observation.
Technically I agree with you, but it is up to the court to judge if it can be concidered proven that the person has been there. I believe that it is one piece in a large number of circumstances that makes it *very* likely that the person has been at the location at the time.
It seems that half of the comments are from people who has not read the article!
The article talks of a radar system based on the reflected waves from mobile phones. I have a number of problems with this:
* The problem is huge, as each signal emitter is mobile, and thus the signal processing needed to filter out the source of each signal-bounce must be huge.
* As the number of signal emitters are variable in the vicinity of each reciever, this make the signal processing even more complex.
* They claim to being able to put all this in a laptop sized device.
This would not be so controversial if it was a simple cell phone tracing system, as they allready exist. In Sweden, one of the major competators even offer a 'locate' service, allowing other users to locate a phone. This service can be turned on and off from the located phone by sending SMSs. Even when turned off, the phone can still be located, all you block is the ability to get a position on another phone. This can, and has been used by the police to, for example, prove that a certain person has been at a certain location at a certain time.
Get this guy to work with your applications, and suddenly your word processor turns into an adventure game, you surface plots fill with small creatures fighting and, well, your internet browser turns into a pacman clone.
I'd say that this shows that todays game engines are pretty flexible, to say the least. How about hacking Quake into a Pacman clone, imagine your own mirror image: big, yellow and round with a mouth covering 50% of your body. And wouldn't it be nicer to get hunted by yellow, pink and blue ghosts instead of really scary corpses and zombies...
I'd say that there are lots of other, more plaussible, explanations to 'mysterious swirling patches' on a planet surface.
But, hey, the sientisist will get a headline or two, and perhaps even a few dollars to spend. I'm just saying that there are reasons to stretch the reality just a bit sometimes. Often these reasons are political or economical. In this case I'd have to go for the latter.
I must recommend Qt. You'll get neat code, portable to Win32, X-windows, BeOS, etc. Get it from Trolltech!
It is easier to optimize for x86 than RISC.
If a RISC is a proper RISC there are probably no more than one or two ways to do an operation. Since all similair instructions takes an equal amount of time there is no need to optimize. That is what I would call, really easy to optimize, i.e. no need too!
You are right that the moder x86 implementations are RISCs with a translation layer around them (except Crusoe which is a VLIW with software translation - much cooler 8P ). Now just imagine if we could get direct access to those highly optimized RISC cores instead of having to code in x86 machine code.
If we're going to stick to the x86 we still do not want to add complexity. I also tried to point out how easy it would be to move to a new architecture.
As you must add complexity I do not think that it would be "quick and easy". It takes huge resources in both time and equipment to verify the timing of a new chip, so these kind of changes (fundamental changes to the way registers are accessed) are expensive and hard since you also need to implement many new hardware solutions and verify the functionality (not only the timing!)
Ok, he realizes that the x86 architecture is flawed. One of the most limiting problems is the lack of general purpose registers (GPR), so he adds more complexity to an allready over-complex solution to solve this problem. All I have to say to this is: when will you see that the solution is as simple as switching architecture!
As most code today is written in higher level languages (C/C++, Java, etc.) all it takes is a recompile and perhaps some patching and adaptations to small peculiarities. The Linux kernel is a proof of this concept, a highly complex piece of code portable to several platforms with a huge part of the code folly portable and shareable. This means that it is not hard to change architecture!
If the main competition and its money would move from the x86 to a RISC architecure (why not Alpha, MIPS, SPARC or PPC) I'm sure that the gap in performance per penny would go away pretty soon. RISCs have several advantages, but the biggest (IMHO) is the simplicity: no akward rules (non-GP registers), no special case instructions, easy to pipeline, easy to understand and easy to optimize code for (since the instruction set is smaller).
And to return to the original article. Please do not introduce more complexity. What we need is simple, beautiful designs, those are the ones that one can make go *really* fast.
Does anyone know anything about the security problems this kind of device can cause? How easy is it to sniff out passwords etc from bluetooth and how easy is it to trick the drive into thinking that someone else is the owner?
I understand it as they have a choice: either they pay (or make another kind of settlement) or they may face the problem of not being prohibited from selling any.
Still, no-one holds a patent to VLIW in it self. It must be a patent of some implementation detail that is now covered in the detail.
EPIC is imply Intel sales lingo for VLIW.
The article fails to mention any technical details concerning the patent, but it sounds as if it has something to do with Intel's EPIC (which basicly is VLIW).
I'd say that Intel will have to pay, because if Intergraph stops the Itanium series of processors, it will mean an even bigger loss of money, and perhaps give AMD time to get a market advantage with the Hammer.
Ok, there is no 'local oil price', since if the oil from the Gulf would be stopped by a war the non-US world would need oil from somewhere else (pehaps even the US) and prices would go up (since the non-US world would be prepared to pay a higher price).
As for you arguing that since the US is not affected it is OK to attack Iraq is *very* short sighted. I wish that Bush (and his followers) would see that their *allies* (within NATO) do not want this (just ask France), and their friends (through Partnership for Piece) do not want it either (just ask Russia). Never in the history of UN, a war, just to be on the safe side, has been sanctioned, and I hope it never will.
The September 11th attack was a cruel terrorist attack on civians, but the US must see why they are picked as the target. Both the current Bagdad regime and the Talibans are (at least partially) the creations of CIA. The Talibans fought the USSR and were supported by the CIA, as was Saddam was supported in the war agains Iran as Iraq was deemed to be a smaller threat. By these kind of operations the US create instability in other regions of the world, and now some fanatics want to bite back.
An important note: I do *not* support any terrorist activities, I'm just saying that there is a reason to why people become terrorists (desperation, lack of influense, abuse, etc) and maybe one can try to work in that end instead of bombing everyone not liking you (which leads to more people not liking you).
It is good that someone tries to chart this problem. At least it makes big corporations aware of the problem with wireless systems and the security issues associated with them.
I like the idea of wireless internet access everywhere, but not though stealing bandwidth of some business with bad security. I feel very bad for the companies being hacked and abused because of the bad security of the wireless solutions they use.
It surprises me that no-one thought of this before the technology was launched.
As always, money talks!
1) M$ wants to make money from games and selling expensive SDKs.
2) M$ introduces a console called XBox constructed from a PC with some signing stuff in HW.
3) XBox gets modded fairly quickly.
4) XBox can now run Linux.
5) M$ sees a potential threat divided in two parts a. one can copy games, b. one can use the XBox to pull M$'s legg (running Linux on it).
6) M$ sues the hell out off anyone getting to close.
7) M$ ends up with a huge pile of money!
Please excuse me ranting, but I get so tired of their lame attempts to introduce signing. Use a custom CPU with on-chip signing, a motherboard without any standard devices etc. and it will become harder. They could not for a minute have thought that a PC based console would be left unhacked.
As for custom conponents being more expensive - the *big* money can be found in games, not the actual hw. If they were sure to sell games they could just give the thing away!
It was ironc to see how the Swedish newspaper 'Aftonbladet' put this article on the same page as the ads for the latest games...
I've heard a saying that 'fighting leads to love'... It seems like it applies to companies too ;-)
It is really silly that you need to opt-out to try to get away (and you still get calls even then). If there would be an opt-in it would probably kill off the whole telemarketing business, which I feel would be a good thing.
This problem is just as bad as spam, fax-spamming and all other broad adverticement methods. I wish that one day you would not get more ads than you ask for (i.e. 2-5% of todays load of shit).
For you people living in Sweden I would like to mention the Nix register, it is the Swedish opt-out register.