What's the problem with distributing patented technology in source form? I believe this is legal. As an example VTK distributes the marching cube patented method (among others) with no problem.
Unisys never had a problem with any of the LZW implementation in source form. They never asked for them to be pulled out of any site, and neither could they legally. What they asked is if you were using this technology for anything other than research and study (i.e. if you really wanted to compress some file with it for redistribution) *then* you needed a license from them.
The use of patented methods for research and study is legal, this is the whole point of patenting technology. Patenting is a publication process, in exchange for exclusive control of the technology *in applications*. The idea is that other people can study this technology and improve on it.
If you as a user take some source code floating on the net implementing some patented technology, and add it to some application, be the application free or not, you are responsible for obtaining a license from the holder of the patent, but AFAIK the author of the code is in the clear, and so are the distributors.
In a single sentence, you've managed to express the general sentiment of the US population with respect to the culture of the rest of the world, while hinting at what the rest of the world thinks the US think of their own culture, if you catch my drift.
Not to mention the current "might makes right" US philosophy, applicable to a range of domains from oil fields to TV programs. Well done.
Go see the "Barbarian Invasions" film and let's talk about crappy Canadian culture after that.
> Microsoft, nor does any other company, have a > civic duty - their only duty is to make the > shareholders money.
If that were the case then Piracy, Murder and Mayhem would be legal, as means of making money (kill someone, take their money, etc). As would organized crime, etc.
The duties of a company includes respecting the Law and behaving as a responsible civic entity. If they don't they can be sued to oblivion, their assets taken, etc, and this is how is should be.
I stand by my comment, I'm a researcher in image analysis, I do colour processing all the time, I've never used Pantone in my work and never have I seen any other researcher use them, because it's not a published open standard. Pantone solves one problem very well through the sale of proprietary standards and products, but it is not the scientific reference to colour processing. If you want to define a particular colour unambiguously and for all times, the only currently accepted scientific way is to use one of the CIE colour spaces.
Furthermore I have a hard time believing that Pantone it *the* standard for prepress printing. There is such a thing as the ICC, the International Colour Consortium, which would have a better claim to that. There are other systems used in pre-print, such as Focoltone and Trumatch among others. Pantone is a widespread and easy to use system to specify colour (use Pantone colour #285) but it doesn't solve all the problems associated with colours.
The Slashdot article you refer to is very interesting. CmdrTaco made a comment on the submission ("No wireless. Less space than a nomad. Lame."), a mistake which no respectable journalist would make (let the fact speak for themselves).
However the bulk of the article comments, the real mind of Slashdot, most of the +5 insightful comments were saying: "wait a minute, this is *not* lame: firewire, small form factor, cool software, 5GB is plenty, this is gonna fly", which it did.
Don't confuse one editor with the Slashdot collective. I'm always interested by the mixture of inane and extremely insightful comments that Slashdot generates.
Others have replied thoughtfully, now consider what is happening in the proprietary Unix market due to the introduction of Linux and the free BSDs. Sun is hurting, IBM is considering dropping AIX, SCO has become a litigation company, HP and Novell are going full steam ahead with Linux.
In the meantime Microsoft is forging ahead with new versions of Windows that don't seem to add much to what the customer wants. Does XP innovate with respect to win2k? I'm not sure. Will Longhorn be any good when it is released? I don't known.
The point is that if when Longhorn is released ReactOS can already run most win32, and win64 things as well, then it provides competition with Microsoft on its home turf. Microsoft will cease to have a stranglehold on win32 and that can only be good. We can look forward to a sleek, efficient, Free platform that can run most of the commercial and Free software that exists now. How can that be bad?
LISP is a great language that should have used something else than parentheses all the time. Also the exclusive use of recursion for loops gets old quickly.
But it was ahead of its time by decades, and is still alive and well.
Thanks, I'd never heard of this term before. Telecentric lenses seems more complicated than normal lenses, are they only available for high-end camera, or do normal consumer-level lenses for digital cameras also are telecentric?
That's nice, except it's not quite true. The whole Pantone schemozle is a great system for colour referencing and colour calibration, but even *with* it there can be no guarantee that the output of the printing process matches what you see on your screen. It depends on the observer and on the lighting conditions, for a start. No such guarantee is possible.
You are correct in saying that in the US design world you won't get anywhere without Pantone but there exists other systems in other countries and other areas, For example the ultimate reference to colour in the scientific world is not Pantone, is CIE.
CIE were the first to conduct scientific colour perception experiments 90 years ago way before the first computer, and now they are the ISO colour standardization body.
I'm not sure how well PS supports the CIE standards but at least the Gimp supports CIE-Lab.
Actually, the nipple is not intuitive at all. I'm sort of tired of seeing this trite sentence all over the place. If you had been a parent, especially a mother, you'd know that breastfeeding is all but easy the first time around. Both mum and baby both must learn how to make it work. It can take a few days, and in some cases longer.
The only things that dont need to be learned in a baby are crying, peeing, pooping and sleeping, and I'm not sure about the last item.
Does Chinese take long to speak? I have a hard time believing this, given that each character is on average 2 syllables and sometimes express complex concepts.
I remember reading the complete rules of western Chess (that came with a set bought in China, on which you could play traditional Chinese Chess and the Western version as well) in about 100 characters.
Every subtitled film in any language prints far less than what is being said. This is because people don't want to spend their time reading subtitles, they want to see what happens too. Some people read fast and would have little problem following everything but many don't read that fast.
In my experience subtitles are usually terse and often wrong too.
There is a post further down the list, which illustrates that even if you harnest the power of a supernova to power the most efficient computer imaginable, you still can't brute-force through an encryption key 256-bit long. The computation time is of no importance, you simply do not have enough energy to do it.
This is simply an illustration of a simple problem that people might actually want to solve, that we will never be able to even with the most efficient computer that we can imagine.
Dyson's famous paper assumed a universe without accelerating expansion. This is precisely the point of the new paper. From some distant supernovae measurement, it seems the rate of expansion of the universe is in fact accelerating.
Not that long. A nanosecond is 30 cm in vacuum. 1GHz mean a period of one nanosecond. 10GHz CPUs are on the drawing board. This means we will be limited to circuits 3cm in diameter, and much less in the future. Transistors have to keep shrinking, this is the only way, and soon we will hit the "a few atoms" limit, this is a dead certainty.
Like you say we can go 3D, but then to sustain Moore's law, the number of layers has to double every 18 months. Like every exponential function, it won't take very many cycles to hit the limit of what can be done. And remember that at 10GHz everything will have to fit in a 3cm ball. Then heat has to be extracted from that ball.
Soon we will hit the real barriers. I know this has been said many times, but this is going to happen eventually and for real, very likely within our lifetime.
Re: RDP, absolutely correct, but you need to be running one of the server editions of Windows (2000 or 2003), otherwise you get logged off at the console in XP. If you are a normal Windows user (not running the server version) then VNC is better. The bandwidth requirement for both the X and Windows versions of VNC are comparable.
Windows is not slower at number crunching but I don't know where you got the information that it was "faster at everything else".
Thanks, this is a good reply, I still have more questions:
1- It is true that even though I've programmed in RT environments, I've never used a RT GC, I'm just relaying the comment that both the C and C++ committee have delayed adoption of a GC standard for those two languages and that a serious issue is real-time performance. The C/C++ standards committees are not a bunch of idiots so I take the view that there must be something true in there. You are correct that for some hard real-time problems even using malloc is a no-go area.
2- Language safety. I'm going back to the comment regarding other resources than memory. If you need a GC for your language to be safe then you need some other automated mechanism to close down open resources for you as well. GCs are not the be-all and end-all of language safety.
What about real-time constraints? GC are generally non-deterministic (they start and finish according to their own rules), which might destroy your maximal response time in a RTOS. This is this very issue that has been the thorn in the side of GC adoption for the C and C++ standards.
How about one of the earlier comments to the effect that mark-and-sweep type algorithms page-faults all the memory used by an application? That has got to be inefficient, and since virtual memory is not under the control of the application by definition there is nothing that can be done, except if the GC is directly under the control of the OS, which doesn't often makes sense (it's not very flexible then).
The article itself says that there is no way to make a GC perform as well or better as a finely tuned hand-micro-managed in every case. The article being hugely in favour of GCs I'd take this comment as probably true. The advantage of GC is that it makes memory management easier, not necessarily more efficient.
In languages that don't have GCs you can add one yourself (Bohm's GC works fine for C/C++, and is in fact used for GCJ, the GNU implementation of the Java language), with the benefit that you can turn it off if you don't want it for some reason, something you can't do in Java for example.
Finally Memory is one but many of the critical resources that need to be managed in a program. GCs only manage memory. C++ teaches a nice way of working with all critical resources including memory: the Resource Acquisition is Initialization idiom. Worth learning about, and deterministic.
Then there needs to be some mechanism to transfer data from the PC bus to the RAM on the card, this doesn't come free from my experience (admitedly not up-to-date). Microcontroller usually have ROM too anyway.
I'm a bit worried about the weaponry aspect of your reply, I think self-defence courses should be a pretty good start and would also work in countries and states that don't allow people to bear arms as a matter of course.
> It seems to me that the America's interests > aligned more with peace and freedom than any other > great (former great?) power during the 20th > century.
I will agree that all the former little superpowers of the 20th century in Europe behaved apallingly. I would still argue that America's interests are certainly not aligned with peace or freedom, certainly not now. America also props up horrible dictators, sells arms to butchers (no name needed) and only invoke peace and freedom when it suits them.
This is a feature of all superpowers, not just the USA, nothing personal here.
> there were a lot of US troops in Germany all > through the cold war, and they would certainly > have been killed if a war had broken out. Hence > my point: the US was willing to endanger its own > people, viz, soldiers, to defend a geographical > area outside their homeland.
I don't understand that argument, there were certainly a lot of Russian soldiers in East Germany as well, or for that matter everywhere East of the curtain. Does that implies that the Soviets were somehow good guys?
The alternative is a UN that works. Don't laugh. I mean *that works*. This means everyone giving up a lot of power: no more veto at the SC, majority of 2/3rds in general assembly decides what goes. The UN also needs a stronger army with more on-the-ground capacity for independent intervention. Something along that line will just have to happen eventually. I just hope it won't take another century and another WW.
I don't know why the UN pulled out of Rwanda. Someone was pulling the strings if the advisors were saying otherwise.
Re:I'll tell you exactly why Windows won, it's sim
on
The War Of The Word
·
· Score: 1
I remember the day Windows 3.0 came out. It looked gorgeous compared with the DOS horrors of the day, a hundred times better than Windows 2.1 ; it looked like it was going to be as good as OS/2 which didn't run everywhere (386s and above, or for that matter 286s were still too expensive for a lot of people). It looked like Microsoft was doing the right thing by not being constrained by IBM's decisions which mostly concerned the higher end of the PC market.
People where heartily sick of DOS, that is for sure, the market could have gone with OS/2 if it had been bundled with all new PCs like DOS was, and if IBM had not wanted more control over it than with DOS. Instead Microsoft continued to use its DOS leverare (DOS was cheap), developed Windows 3.x to work on top of DOS to pacify the masses, and the rest is history.
People say that Gates got lucky to get the call in 1981 or so, but even though I don't like Microsoft much these days, you have to admit he did good then, he was able to wrangle a very positive licensing deal with Microsoft which made it possible to sell DOS to the IBM clones that bypassed IBM completely, then he masterfully managed the relationship with IBM such that they asked him again for help with OS/2, and then when he felt strong enough, the backstabbed IBM which was a very gutsy thing to do back then. And then he pushed thing through with Windows and he won.
I think Gates understood better than anyone that forcing one's way with the hardware vendors, particularly the bottom end, to go with their OSes was the key. Any tactic was fine as long as it did the job, and it worked well.
Slashdot Mirror
What's the problem with distributing patented technology in source form? I believe this is legal. As an example VTK distributes the marching cube patented method (among others) with no problem.
Unisys never had a problem with any of the LZW implementation in source form. They never asked for them to be pulled out of any site, and neither could they legally. What they asked is if you were using this technology for anything other than research and study (i.e. if you really wanted to compress some file with it for redistribution) *then* you needed a license from them.
The use of patented methods for research and study is legal, this is the whole point of patenting technology. Patenting is a publication process, in exchange for exclusive control of the technology *in applications*. The idea is that other people can study this technology and improve on it.
If you as a user take some source code floating on the net implementing some patented technology, and add it to some application, be the application free or not, you are responsible for obtaining a license from the holder of the patent, but AFAIK the author of the code is in the clear, and so are the distributors.
Congratulations,
In a single sentence, you've managed to express the general sentiment of the US population with respect to the culture of the rest of the world, while hinting at what the rest of the world thinks the US think of their own culture, if you catch my drift.
Not to mention the current "might makes right" US philosophy, applicable to a range of domains from oil fields to TV programs. Well done.
Go see the "Barbarian Invasions" film and let's talk about crappy Canadian culture after that.
This is not true:
> Microsoft, nor does any other company, have a
> civic duty - their only duty is to make the
> shareholders money.
If that were the case then Piracy, Murder and Mayhem would be legal, as means of making money (kill someone, take their money, etc). As would organized crime, etc.
The duties of a company includes respecting the Law and behaving as a responsible civic entity. If they don't they can be sued to oblivion, their assets taken, etc, and this is how is should be.
I stand by my comment, I'm a researcher in image analysis, I do colour processing all the time, I've never used Pantone in my work and never have I seen any other researcher use them, because it's not a published open standard. Pantone solves one problem very well through the sale of proprietary standards and products, but it is not the scientific reference to colour processing. If you want to define a particular colour unambiguously and for all times, the only currently accepted scientific way is to use one of the CIE colour spaces.
Furthermore I have a hard time believing that Pantone it *the* standard for prepress printing. There is such a thing as the ICC, the International Colour Consortium, which would have a better claim to that. There are other systems used in pre-print, such as Focoltone and Trumatch among others. Pantone is a widespread and easy to use system to specify colour (use Pantone colour #285) but it doesn't solve all the problems associated with colours.
The Slashdot article you refer to is very interesting. CmdrTaco made a comment on the submission ("No wireless. Less space than a nomad. Lame."), a mistake which no respectable journalist would make (let the fact speak for themselves).
However the bulk of the article comments, the real mind of Slashdot, most of the +5 insightful comments were saying: "wait a minute, this is *not* lame: firewire, small form factor, cool software, 5GB is plenty, this is gonna fly", which it did.
Don't confuse one editor with the Slashdot collective. I'm always interested by the mixture of inane and extremely insightful comments that Slashdot generates.
Others have replied thoughtfully, now consider what is happening in the proprietary Unix market due to the introduction of Linux and the free BSDs. Sun is hurting, IBM is considering dropping AIX, SCO has become a litigation company, HP and Novell are going full steam ahead with Linux.
In the meantime Microsoft is forging ahead with new versions of Windows that don't seem to add much to what the customer wants. Does XP innovate with respect to win2k? I'm not sure. Will Longhorn be any good when it is released? I don't known.
The point is that if when Longhorn is released ReactOS can already run most win32, and win64 things as well, then it provides competition with Microsoft on its home turf. Microsoft will cease to have a stranglehold on win32 and that can only be good. We can look forward to a sleek, efficient, Free platform that can run most of the commercial and Free software that exists now. How can that be bad?
Sorry to be pedandic,
find / -name "*base*" -exec chown -R us {} \;
works better.
LISP is a great language that should have used something else than parentheses all the time. Also the exclusive use of recursion for loops gets old quickly.
But it was ahead of its time by decades, and is still alive and well.
Thanks, I'd never heard of this term before. Telecentric lenses seems more complicated than normal lenses, are they only available for high-end camera, or do normal consumer-level lenses for digital cameras also are telecentric?
That's nice, except it's not quite true. The whole Pantone schemozle is a great system for colour referencing and colour calibration, but even *with* it there can be no guarantee that the output of the printing process matches what you see on your screen. It depends on the observer and on the lighting conditions, for a start. No such guarantee is possible.
You are correct in saying that in the US design world you won't get anywhere without Pantone but there exists other systems in other countries and other areas, For example the ultimate reference to colour in the scientific world is not Pantone, is CIE.
CIE were the first to conduct scientific colour perception experiments 90 years ago way before the first computer, and now they are the ISO colour standardization body.
I'm not sure how well PS supports the CIE standards but at least the Gimp supports CIE-Lab.
Actually, the nipple is not intuitive at all. I'm sort of tired of seeing this trite sentence all over the place. If you had been a parent, especially a mother, you'd know that breastfeeding is all but easy the first time around. Both mum and baby both must learn how to make it work. It can take a few days, and in some cases longer.
The only things that dont need to be learned in a baby are crying, peeing, pooping and sleeping, and I'm not sure about the last item.
No mod points for ACs, sorry.
Re: your sig
> According to slashdot users I'm funny,
> insightful, and interesting! So why arn't girls
> all over me?
Because you are only funny, insightful and interesting to other geeks.
Does Chinese take long to speak? I have a hard time believing this, given that each character is on average 2 syllables and sometimes express complex concepts.
I remember reading the complete rules of western Chess (that came with a set bought in China, on which you could play traditional Chinese Chess and the Western version as well) in about 100 characters.
Every subtitled film in any language prints far less than what is being said. This is because people don't want to spend their time reading subtitles, they want to see what happens too. Some people read fast and would have little problem following everything but many don't read that fast.
In my experience subtitles are usually terse and often wrong too.
There is a post further down the list, which illustrates that even if you harnest the power of a supernova to power the most efficient computer imaginable, you still can't brute-force through an encryption key 256-bit long. The computation time is of no importance, you simply do not have enough energy to do it.
This is simply an illustration of a simple problem that people might actually want to solve, that we will never be able to even with the most efficient computer that we can imagine.
Dyson's famous paper assumed a universe without accelerating expansion. This is precisely the point of the new paper. From some distant supernovae measurement, it seems the rate of expansion of the universe is in fact accelerating.
Not that long. A nanosecond is 30 cm in vacuum. 1GHz mean a period of one nanosecond. 10GHz CPUs are on the drawing board. This means we will be limited to circuits 3cm in diameter, and much less in the future. Transistors have to keep shrinking, this is the only way, and soon we will hit the "a few atoms" limit, this is a dead certainty.
Like you say we can go 3D, but then to sustain Moore's law, the number of layers has to double every 18 months. Like every exponential function, it won't take very many cycles to hit the limit of what can be done. And remember that at 10GHz everything will have to fit in a 3cm ball. Then heat has to be extracted from that ball.
Soon we will hit the real barriers. I know this has been said many times, but this is going to happen eventually and for real, very likely within our lifetime.
Re: RDP, absolutely correct, but you need to be running one of the server editions of Windows (2000 or 2003), otherwise you get logged off at the console in XP. If you are a normal Windows user (not running the server version) then VNC is better. The bandwidth requirement for both the X and Windows versions of VNC are comparable.
Windows is not slower at number crunching but I don't know where you got the information that it was "faster at everything else".
Thanks, this is a good reply, I still have more questions:
1- It is true that even though I've programmed in RT environments, I've never used a RT GC, I'm just relaying the comment that both the C and C++ committee have delayed adoption of a GC standard for those two languages and that a serious issue is real-time performance. The C/C++ standards committees are not a bunch of idiots so I take the view that there must be something true in there.
You are correct that for some hard real-time problems even using malloc is a no-go area.
2- Language safety. I'm going back to the comment regarding other resources than memory. If you need a GC for your language to be safe then you need some other automated mechanism to close down open resources for you as well. GCs are not the be-all and end-all of language safety.
What about real-time constraints? GC are generally non-deterministic (they start and finish according to their own rules), which might destroy your maximal response time in a RTOS. This is this very issue that has been the thorn in the side of GC adoption for the C and C++ standards.
How about one of the earlier comments to the effect that mark-and-sweep type algorithms page-faults all the memory used by an application? That has got to be inefficient, and since virtual memory is not under the control of the application by definition there is nothing that can be done, except if the GC is directly under the control of the OS, which doesn't often makes sense (it's not very flexible then).
The article itself says that there is no way to make a GC perform as well or better as a finely tuned hand-micro-managed in every case. The article being hugely in favour of GCs I'd take this comment as probably true. The advantage of GC is that it makes memory management easier, not necessarily more efficient.
In languages that don't have GCs you can add one yourself (Bohm's GC works fine for C/C++, and is in fact used for GCJ, the GNU implementation of the Java language), with the benefit that you can turn it off if you don't want it for some reason, something you can't do in Java for example.
Finally Memory is one but many of the critical resources that need to be managed in a program. GCs only manage memory. C++ teaches a nice way of working with all critical resources including memory: the Resource Acquisition is Initialization idiom. Worth learning about, and deterministic.
Then there needs to be some mechanism to transfer data from the PC bus to the RAM on the card, this doesn't come free from my experience (admitedly not up-to-date). Microcontroller usually have ROM too anyway.
I'm a bit worried about the weaponry aspect of your reply, I think self-defence courses should be a pretty good start and would also work in countries and states that don't allow people to bear arms as a matter of course.
Otherwise thanks for the insight.
> It seems to me that the America's interests
> aligned more with peace and freedom than any other
> great (former great?) power during the 20th
> century.
I will agree that all the former little superpowers of the 20th century in Europe behaved apallingly. I would still argue that America's interests are certainly not aligned with peace or freedom, certainly not now. America also props up horrible dictators, sells arms to butchers (no name needed) and only invoke peace and freedom when it suits them.
This is a feature of all superpowers, not just the USA, nothing personal here.
> there were a lot of US troops in Germany all
> through the cold war, and they would certainly
> have been killed if a war had broken out. Hence
> my point: the US was willing to endanger its own
> people, viz, soldiers, to defend a geographical
> area outside their homeland.
I don't understand that argument, there were certainly a lot of Russian soldiers in East Germany as well, or for that matter everywhere East of the curtain. Does that implies that the Soviets were somehow good guys?
The alternative is a UN that works. Don't laugh. I mean *that works*. This means everyone giving up a lot of power: no more veto at the SC, majority of 2/3rds in general assembly decides what goes. The UN also needs a stronger army with more on-the-ground capacity for independent intervention. Something along that line will just have to happen eventually. I just hope it won't take another century and another WW.
I don't know why the UN pulled out of Rwanda. Someone was pulling the strings if the advisors were saying otherwise.
I remember the day Windows 3.0 came out. It looked gorgeous compared with the DOS horrors of the day, a hundred times better than Windows 2.1 ; it looked like it was going to be as good as OS/2 which didn't run everywhere (386s and above, or for that matter 286s were still too expensive for a lot of people). It looked like Microsoft was doing the right thing by not being constrained by IBM's decisions which mostly concerned the higher end of the PC market.
People where heartily sick of DOS, that is for sure, the market could have gone with OS/2 if it had been bundled with all new PCs like DOS was, and if IBM had not wanted more control over it than with DOS. Instead Microsoft continued to use its DOS leverare (DOS was cheap), developed Windows 3.x to work on top of DOS to pacify the masses, and the rest is history.
People say that Gates got lucky to get the call in 1981 or so, but even though I don't like Microsoft much these days, you have to admit he did good then, he was able to wrangle a very positive licensing deal with Microsoft which made it possible to sell DOS to the IBM clones that bypassed IBM completely, then he masterfully managed the relationship with IBM such that they asked him again for help with OS/2, and then when he felt strong enough, the backstabbed IBM which was a very gutsy thing to do back then. And then he pushed thing through with Windows and he won.
I think Gates understood better than anyone that forcing one's way with the hardware vendors, particularly the bottom end, to go with their OSes was the key. Any tactic was fine as long as it did the job, and it worked well.