Regulated means regulated, and the current definition works just as well as the historic definition.
The gap is that you're assuming regulated means regulated by the Federal or state government. It does not, and in fact if you learn about the Bill of Rights, you'll find the intent was to guarantee the protection of rights the founders thought were natural civil rights. The core Constitution and Bill of Rights do not grant rights to the citizens; they recognize those rights as inherent and set limits to the powers of government.
The Federal Government intended to provide for the training and armament of the population. What happened was that after 70+ years of organized militias (the founding years through the Militia Acts), our governments realized the professional soldiers were far more effective in fighting wars, and cut funding to the militias. It's still advocated - see the Civilian Marksmanship Program - but neither mandatory nor comprehensive.
Gun rights activists like to point to Switzerland, and they are right except for one issue: in Switzerland, every conscripted citizen does get a fully-automatic assault rifle, but at the same time, that person is also trained in citizenship, ethics, military discipline, and handling of their weapon. But how do you think Switzerland would be in 150 years, if all other things stay the same, they were to continue providing assault rifles, but stopped educating?
The issue with your approach is that it still allows someone to get close to the school while armed. The solution we found in the military was the concept of remote Entry Control Points. You don't do much other than move the scanner and entrance a few hundred (or thousand) feet away from the school. The military doesn't even typically even have to build fences to keep people from skirting around the ECP - they just shoot anyone who happens to go over a particular painted line. (You wouldn't want to do exactly this at the school, but the point is that it doesn't need to be complicated or expensive). But, even if we did that, we'd stop in 5 years, because with an ECP, you wouldn't catch criminals, you'd deter them, so there'd be no evidence they were working, and they'd be defunded eventually.
There's a good point that if the school administrators had been armed, it's not that the attacker would have been shot down, but that he would have never tried in the first place. The concept is that he went for the softest target, and that if there were no soft targets, he might have killed just his mother and himself. The compromise is to find a way to keep malls and schools from being soft targets without arming everyone.
The Ruger Mini-14 is a typical hunting rifle, semiautomatic action, that uses magazines that are similar to NATO STANAG (they look like AR-15) but lower capacity. The Remington 1100 is a typical hunting shotgun, semiautomatic action, capable of firing 00-buckshot, where each shotshell contains 8+ pellets, each potentially lethal on a human-sized target. The Uberti Revolving Carbine is a fancy black-powder muzzleloader, suitable for hunting, that can fire six.44-caliber shots before reloading, thanks to a revolving breech (like a revolver pistol). And it's not even Federally a firearm, since it doesn't use fixed ammunition.
These three weapons are also specifically identified as hunting rifles in the expired Assault Weapons Ban. They are so far in the realm of intentioned and effective hunting weapons that they were called out.
'The 20-year-old guys provide me more value than the 35-year-olds do.'"
Value=lower salary & willing to give up having a life outside of work.
And that's really it.
Older folks, generally, cost more.
In the US, (I'll make some numbers up, but depending on where you are, the proportions are correct) corporate hiring knows they can hire a rockstar out of college for less than $90k, or an average programmer for less than $70k. (Even as that rockstar is 3-10* as productive as an average employee). Why pay $120-50k for an average 45 year old engineer? They assume the experienced rockstars figured it out, started their own businesses, or otherwise moved into senior non-coder roles, and the aged coders are people who just couldn't cut it doing something else. So your software engineering degree isn't necessarily worth less, but if you expect to be doing the same thing with it at 45 that you did at 21, you have a surprise coming unless you plan very well. There are great ways of doing this - becoming a subject-matter expert in something rare, consulting, moving into a mentoring role, or working for companies that are less bottom-line focused (government/military-industrial complex). But there's a substantial number of software developers for whom there is someone else willing to try to do their job for less $. That's one of the big reasons for both unions and professional licensure, but that's another discussion.
This isn't unique at all to us. Any job enjoys this - "Step Up or Step Out". If you're an aging worker, you've always got to ask yourself what you provide that a college grad doesn't. (And hope you aren't asking what you provide that a HS grad doesn't, like many folks had to during/after the.com bubble). The canonical answer is "experience", but the professions show that isn't really true unless you can directly demonstrate it. More senior doctors in some fields are more prone to mistakes than younger doctors, because the senior doctors trust their "experience" whereas the younger doctors trust research. But the senior doctors also handle more patients, due in part to the same corner-cutting.
Option 1. Re-implement your legacy application on a modern platform, from legacy source code, or from scratch/reverse-engineering. - You'll pay down your technical debt and possibly have a supportable, and maybe even virtualized, production system going forward. Option 2. Sustain legacy equipment knowing that no modern emulator handles all the details of your particular proprietary hardware. - Double down on your investment and leave the problem for the next guy. Option 3. Hire IBM. (Nobody ever got fired for buying IBM). - Mostly this will end up just like (2).
Oh, you wanted virtualization...but I think that's a solution to a different problem than the one you are facing. Get that app working on something reasonably open and then we can talk about virtualization.
This story shows up in Google's top 10 search results for me for Unisys Unixware right now, which really should emphasize to you the magnitude of the pain you might be facing...
A patent is a set of exclusive rights granted by a state to an inventor or his assignee for a fixed period of time in exchange for a disclosure of an invention.
Patents, from the beginning, were a compromise so that people who would invest in new developments would disclose the work of those developments (for public good) while being able to turn a profit from them in the short term (a motivation for inventing) through an exclusive monopoly.
Vested interests do not write the law, for it is the individual who has the most vested interest in the government. I think you mean to say "sociopathic capitalists write the law".
For patents to benefit society, the term of the monopoly must be greater than that required to recoup investment expenses, but shorter than the portion of an invention's life span where it is valuable to the people. In a government that exists for the benefit of the people, the shortest patent term is the most desirable. That's how our government was set up -- unfortunately, the world is more and more getting exactly what it deserves, as a few have learned that people will sell their freedom for remarkably little.
People can't own their ideas because they were never wholly their ideas. All that we invent is the summation of all that has come before us, perhaps with something new thrown into the mix. Your ideas belong just as much to your teachers, your parents, your peers, and the generations that came before you, as they do to you. In the long term community ownership is the only system that makes sense for such a creation.
In the short term, a man's got to eat. In the long term, society as a whole must reap the rewards for what it has sown. Only a parasite keeps that from society, and like any parasite feasting on a host, society becomes sick when that happens.
This isn't a problem with monolithic-vs-modular software -- Firefox is fairly modular as far as software goes. These problems are design logic problems -- the problem is not the implementation (as is the case with a buffer overrun) but with the logic behind the implementation.
One could argue that modular software is a bad thing because of the unintended or poorly-understood side-effects of modules. For example, take sudo -- there are many commands that can spawn subprocesses that allow a user to use sudo to obtain root access. While sudo itself seems like a nice modular way to allow grant specific, restricted superuser powers to normal users, the complexity of the submodules (any command the user is allowed to run) means it can be intractable to ensure that a user couldn't turn restricted superuser powers into full superuser access.
Rigorous mathematical models can be used to ensure these kinds of problems go away -- but you have to impose severe restrictions on the tools used to develop software to use these models. But these kinds of tools are used in certain industries, like aerospace, where the need for robustness so far outweighs the costs (so far I have not heard of anyone getting killed as the direct result of a firefox crash).
Still, there are ways -- for example if you run Firefox in a virtual machine where the virtual machine has no access to privileged information, then neither does Firefox. The logical integration of that level of security at the application level is the capabilities-based security, and you can prove that applications have no access to privileged information, or that they have access to a minimal, well-defined set of privileged information. Mainframe-style systems, SELinux and Java all can work like this.
Wait, are you implying that people who work for the government might be self-serving parasites?
By making this information available for free, those people who made a living by reproducing it are being put out of a job. And government jobs often attract the kind of people who would are just not adaptable enough to find a new job. So of course they are going to fight this -- because accepting it means they would have to change, and change is hard.
The Toughbook C30s are about as good as you are going to find in a fully rugged portable, but expend to spend about $4000 for a loaded unit.
That said, this guy: http://www.strikingviking.net/ went around the world with a Toshiba Tecra. He apparently had serious problems with a mac and did OK with a vaio.
This would make the risk of a lawsuit -- even a righteous one -- enormous.
Put this in individual terms -- let's say your neighbor's kid broke your window with a baseball, so you want to sue for a new one. But as part of that lawsuit, you would lose your house to the courts if you lose.
You saw your neighbors kid do it, so you go to court, but your neighbor realizes this is his opportunity to really screw you for daring to posit that his kid did something wrong -- so he hires a shark who pays someone off to come up with a fake alibi. And now you're renting a crappy apartment on the edge of town.
Now, you said copyright law - so -
Imagine if Evilsoft hired some talented programmers to write source code that would compile to similar binary code as something the Good Software Foundation had published under a copyleft. The intent was to entice the Good Software Foundation into sueing Evilsoft, so that Evilsoft could show the source code in court, prove their innocence, and make the Good Software Foundation go away forever.
All distribution ultimately relies on an "upstream" distributor.
Let's say you provide FTP access, but use a colocated server that you don't own -- you have an agreement with an upstream provider to provide bandwidth and disk space to provide your licensees with the source. Why does it matter if it's Rackspace or Ubuntu who serves as the upstream provider? Or if it's Memorex media and the Post Office or a floppy and delivered via sneakernet?
The intent here is clear: the entity distributing binaries to some users must also ensure that said users are able to come into possession of a copy of the source through reasonable means, be it downloading or making at-cost physical copies available. In other words, it seems to me that the distributor of binaries must only take some action for which the receipt of the desired source by the requesting licensee is a result.
It requires you to install a browser plugin, then the songs are played through that plugin -- it's fairly large at 1.1MB downloaded, 2.7mB installed -- it's called nprhapengine.so, which definately incorporates a lot of Helix code, and is also statically linked to some open-source software (OpenSSL, libcURL at least). The player works well enough, but the playlist is buggy -- it doesn't seem to support multiple playlists, or even deletion of individual songs from the playlist. And it didn't import the playlists used by the native Windows client (which are stored on the server). But then again, what's there works for me, and there is a "BETA" icon in the upper left of the browser window.
I got the request to install the plugin as soon as I hit any of the play buttons (the arrows next to tracks) on Rhapsody.com; this is the case on Firefox and Safari.
Quite a bit of the most influential discoveries to date have been done by people under 25. Your brain is constantly developing throughout your entire life if you play it right. And brain cells are constantly dying. While teenagers may not have the experience and sense of causality that older adults do, you can't deny that many of the most focused scholars and athletes did amazing things as teenagers and young adults (and then faded into obscurity). Young people do have a level of focus and creativity that is rarely seen amongst older people with brains atrophied by a blue collar job followed by a night of TV and drinking (not to mention age; as soon as you're born you start dying). Fact is, while you may have more experience at 30, you were probably smarter and more focused at 20.
The brain is never "fully developed", the body never "matures", except in a relativistic sense defined by biologists to make their lives easier. Living is a process, growing old is a process, and there is no destination other than the grave. There's no summit or valley anyone reaches as they age or grow in experience; to think otherwise is a wishful appeal to the authority of age we bow to and desire as children.
Normal diesel engines in their standard configuration can not properly work with vegetable oil. You need modified glow plugs (cheap and easy) and fuel injectors (not cheap and easy) to avoid premature engine failure from improper injection and combustion. Standard fuel filters also won't properly deal with straight vegetable oil (maybe easy, maybe a nightmare), especially at cold temperatures or during extended storage when the oil will crystallize. However, if you're willing to go that route, and install a special tank that preheats (definately not cheap or easy) the vegetable oil before the engine is started, you can run SVO. Even if you don't want to pre-heat, you can mix the oil with a bit of gas or dino diesel to reduce the viscosity, and still get a combustible mixture. Regardless, improper configurations or improper mixtures can permanently damage an engine very quickly (a couple hundred miles).
Your argument is flawed in that you neglect to mention that computer-driven models can generate hypothesis that can be *tested*, both retrospectively, and in the future.
In your example, you'll know the computer model is wrong if in a week you are still alive. On the other hand, if the model can make accurate predictions about historical data, without being based on that data, then you have evidence to the model's accuracy. For example, use a machine learning technique to train a system to predict a 1-year climate trend using data from 1900-1990. Then, see if the system can accurately predict trends from 1990-2005. If so, there's no evidence that it would be wrong when predicting a trend from 2005-2015, when trained from 1990-2005. Other similar tests might include training on even-numbered years, and then predicting climate for odd-numbered years, or training on non-leap years, and predicting for leap years.
When you have sufficient data, you can use rigorous statistical methods to say with a known confidence how accurate your methods are likely to be in making predictions (and I'm not just talking about accuracy and recall; you can validate a hypothesis much more rigorously). You can then make rational, scientific statements.
Adobe is using this to reinforce their stance on the issue; whereas the black hat Skylarov could break the encryption, did so, and released it (illegally according to US law), Adobe takes the white hat stance, saying they *could* break and release the encryption, but do not because it is wrong. This decision merely supports their previous stance -- that violating the DMCA is wrong.
It's not ironic that Adobe refuses to commit the same crime that was committed against them -- it's reaffirming of their principles.
Software RAID *is* very often faster, especially on a modern CPU paired with an older design -- you don't buy HW RAID because it is faster, you buy it for battery backup and offloading of low level operations to conserve CPU time and bus/memory bandwidth for user applications and so that if your OS or CPU/memory/whatever blows up, or you lose power, it won't corrupt the data on your disk array. Hardware RAID dedicated processors are simple, slow, "reliable" units -- not ultra-fast bleeding-edge dedicated units like you see on video cards.
The problem with using CL for heavy computing tasks is that Lisp is a linklist-based language. Linklists are very elegant to the programmer, yet they are extremely time-inefficient. A FORTRAN compiler makes good use of the fact that, in integer_4 array foo, foo(bar) is exactly 4 bytes away from foo(bar+1).
Common Lisp code is list-based, and these lists are compiled into regular assembly code by any good compiler, much like C or Fortran. In addition, Lisp makes using linked lists as data structures very easy -- and modern compilers do an excellent job of optimizing them. However, Common Lisp also supports simple and complex arrays, data structures, and objects -- none of which are generally implemented using linked lists, and all of which tend to be rather efficient. In fact, any good Lisp textbook will tell you that code which relies on non-fix-lengthed lists (fix-lengthed lists can be implemented using arrays) is going to be inefficient, and is hopelessly unoptimal.
Your argument might have been true 20+ years ago, but is no longer valid. Compiler technology has come very far. Note for example, that there exists floating-point example code for cmucl which is faster than equivalent C code, compiled with gcc.
Common Lisp is a very high level language with a tremendous amount of expressiveness, and it is suited towards academia in that in general, functionality is not sacrificed for performance.
Check out http://www.lisp.org, http://cmucl.cons.org/cmucl for a really good implementation (and there are even Debian packages of it).
CL is not known for its parallelization abilities, but if you need a language that lets you describe mathematics, CL is useful.
Lisp is actually based around something called the Lambda Calculus, which is a way of expressing concepts by transforming data into other data using data which is expressed as a "function". Because of this, Lisp has a lot of abilities that other languages lack, such as extremely simple and powerful function composition, even at run-time. CL also has a massive core library with OO facilities, basic mathematic primitives, good FFT suppot in most implementations, windowing system support, and good commercial vendors like Franz. Check it out; it's almost as old as Fortran, but has evolved in a much more elegant manner.
A Sun IPX (or any lunchbox style) system with an AUI port and a modified transceiver is much better. I use one of these as a secure syslog; in particular because you can modify the transceiver so that while it is capable of receiving data, it is incapable of sending at a hardware level. There is no way, short of physical access, to detect the machine. It's great for packet sniffing and logging -- syslog using UDP is connectionless, and works well with read-only network connections. This is also better than modifying the ethernet cable, because these modified cables do not actually work properly (the transceiver with tx pins removed will keep a valid *empty* tx signal, whereas a modified cable usually just pumps the rx'd signal back to tx, confusing the equipment into maintaining a link).
And if you can sneak in once, why not twice? Or better, equip the computer with a cell modem or amateur radio equipment (How many "wartalkers" look for that, eh?) , and dial in. No need for probes which may set off IDS systems, or outgoing packets (like ARP or DNS requests) that alert crackers to a computer's presence.
I think you cut pins 3 and 10 (on the connector to the computer on the transceiver) but that's not certain.
(well, DigitalUNIX/Tru64 customers probably *are* a bit of a pain in the ass, compared to MCSEs).
What, a pain in the ass just because solutions to their problems generally aren't solved by reboots or exchanging the hard drive?
Compaq and HP sell crappy hardware. Period. Unfortunately, businesses as of late seem to have taken to the trend of using laws as a last line of defense for their failing business model.
I wonder if any of the RIAA's "reduced" profit figures are because of boycotts from people they pissed off... maybe it does work.
Slashdot Mirror
Regulated means regulated, and the current definition works just as well as the historic definition.
The gap is that you're assuming regulated means regulated by the Federal or state government. It does not, and in fact if you learn about the Bill of Rights, you'll find the intent was to guarantee the protection of rights the founders thought were natural civil rights. The core Constitution and Bill of Rights do not grant rights to the citizens; they recognize those rights as inherent and set limits to the powers of government.
The Federal Government intended to provide for the training and armament of the population. What happened was that after 70+ years of organized militias (the founding years through the Militia Acts), our governments realized the professional soldiers were far more effective in fighting wars, and cut funding to the militias. It's still advocated - see the Civilian Marksmanship Program - but neither mandatory nor comprehensive.
Gun rights activists like to point to Switzerland, and they are right except for one issue: in Switzerland, every conscripted citizen does get a fully-automatic assault rifle, but at the same time, that person is also trained in citizenship, ethics, military discipline, and handling of their weapon. But how do you think Switzerland would be in 150 years, if all other things stay the same, they were to continue providing assault rifles, but stopped educating?
The issue with your approach is that it still allows someone to get close to the school while armed. The solution we found in the military was the concept of remote Entry Control Points. You don't do much other than move the scanner and entrance a few hundred (or thousand) feet away from the school. The military doesn't even typically even have to build fences to keep people from skirting around the ECP - they just shoot anyone who happens to go over a particular painted line. (You wouldn't want to do exactly this at the school, but the point is that it doesn't need to be complicated or expensive). But, even if we did that, we'd stop in 5 years, because with an ECP, you wouldn't catch criminals, you'd deter them, so there'd be no evidence they were working, and they'd be defunded eventually.
There's a good point that if the school administrators had been armed, it's not that the attacker would have been shot down, but that he would have never tried in the first place. The concept is that he went for the softest target, and that if there were no soft targets, he might have killed just his mother and himself. The compromise is to find a way to keep malls and schools from being soft targets without arming everyone.
Manually reloaded? I'm not sure what you mean.
The Ruger Mini-14 is a typical hunting rifle, semiautomatic action, that uses magazines that are similar to NATO STANAG (they look like AR-15) but lower capacity. .44-caliber shots before reloading, thanks to a revolving breech (like a revolver pistol). And it's not even Federally a firearm, since it doesn't use fixed ammunition.
The Remington 1100 is a typical hunting shotgun, semiautomatic action, capable of firing 00-buckshot, where each shotshell contains 8+ pellets, each potentially lethal on a human-sized target.
The Uberti Revolving Carbine is a fancy black-powder muzzleloader, suitable for hunting, that can fire six
These three weapons are also specifically identified as hunting rifles in the expired Assault Weapons Ban. They are so far in the realm of intentioned and effective hunting weapons that they were called out.
'The 20-year-old guys provide me more value than the 35-year-olds do.'"
Value=lower salary & willing to give up having a life outside of work.
And that's really it.
Older folks, generally, cost more.
In the US, (I'll make some numbers up, but depending on where you are, the proportions are correct) corporate hiring knows they can hire a rockstar out of college for less than $90k, or an average programmer for less than $70k. (Even as that rockstar is 3-10* as productive as an average employee). Why pay $120-50k for an average 45 year old engineer? They assume the experienced rockstars figured it out, started their own businesses, or otherwise moved into senior non-coder roles, and the aged coders are people who just couldn't cut it doing something else. So your software engineering degree isn't necessarily worth less, but if you expect to be doing the same thing with it at 45 that you did at 21, you have a surprise coming unless you plan very well. There are great ways of doing this - becoming a subject-matter expert in something rare, consulting, moving into a mentoring role, or working for companies that are less bottom-line focused (government/military-industrial complex). But there's a substantial number of software developers for whom there is someone else willing to try to do their job for less $. That's one of the big reasons for both unions and professional licensure, but that's another discussion.
This isn't unique at all to us. Any job enjoys this - "Step Up or Step Out". If you're an aging worker, you've always got to ask yourself what you provide that a college grad doesn't. (And hope you aren't asking what you provide that a HS grad doesn't, like many folks had to during/after the .com bubble). The canonical answer is "experience", but the professions show that isn't really true unless you can directly demonstrate it. More senior doctors in some fields are more prone to mistakes than younger doctors, because the senior doctors trust their "experience" whereas the younger doctors trust research. But the senior doctors also handle more patients, due in part to the same corner-cutting.
Option 1. Re-implement your legacy application on a modern platform, from legacy source code, or from scratch/reverse-engineering.
- You'll pay down your technical debt and possibly have a supportable, and maybe even virtualized, production system going forward.
Option 2. Sustain legacy equipment knowing that no modern emulator handles all the details of your particular proprietary hardware.
- Double down on your investment and leave the problem for the next guy.
Option 3. Hire IBM. (Nobody ever got fired for buying IBM).
- Mostly this will end up just like (2).
Oh, you wanted virtualization...but I think that's a solution to a different problem than the one you are facing. Get that app working on something reasonably open and then we can talk about virtualization.
This story shows up in Google's top 10 search results for me for Unisys Unixware right now, which really should emphasize to you the magnitude of the pain you might be facing...
Yeah.
Faster mainframes are about scaling crappy legacy applications.
You know, 'coz no one ever gets fired for choosing IBM.
From http://en.wikipedia.org/wiki/Patent:
A patent is a set of exclusive rights granted by a state to an inventor or his assignee for a fixed period of time in exchange for a disclosure of an invention.
Patents, from the beginning, were a compromise so that people who would invest in new developments would disclose the work of those developments (for public good) while being able to turn a profit from them in the short term (a motivation for inventing) through an exclusive monopoly.
Vested interests do not write the law, for it is the individual who has the most vested interest in the government. I think you mean to say "sociopathic capitalists write the law".
For patents to benefit society, the term of the monopoly must be greater than that required to recoup investment expenses, but shorter than the portion of an invention's life span where it is valuable to the people. In a government that exists for the benefit of the people, the shortest patent term is the most desirable. That's how our government was set up -- unfortunately, the world is more and more getting exactly what it deserves, as a few have learned that people will sell their freedom for remarkably little.
People can't own their ideas because they were never wholly their ideas. All that we invent is the summation of all that has come before us, perhaps with something new thrown into the mix. Your ideas belong just as much to your teachers, your parents, your peers, and the generations that came before you, as they do to you. In the long term community ownership is the only system that makes sense for such a creation.
In the short term, a man's got to eat. In the long term, society as a whole must reap the rewards for what it has sown. Only a parasite keeps that from society, and like any parasite feasting on a host, society becomes sick when that happens.
This isn't a problem with monolithic-vs-modular software -- Firefox is fairly modular as far as software goes. These problems are design logic problems -- the problem is not the implementation (as is the case with a buffer overrun) but with the logic behind the implementation.
One could argue that modular software is a bad thing because of the unintended or poorly-understood side-effects of modules. For example, take sudo -- there are many commands that can spawn subprocesses that allow a user to use sudo to obtain root access. While sudo itself seems like a nice modular way to allow grant specific, restricted superuser powers to normal users, the complexity of the submodules (any command the user is allowed to run) means it can be intractable to ensure that a user couldn't turn restricted superuser powers into full superuser access.
Rigorous mathematical models can be used to ensure these kinds of problems go away -- but you have to impose severe restrictions on the tools used to develop software to use these models. But these kinds of tools are used in certain industries, like aerospace, where the need for robustness so far outweighs the costs (so far I have not heard of anyone getting killed as the direct result of a firefox crash).
Still, there are ways -- for example if you run Firefox in a virtual machine where the virtual machine has no access to privileged information, then neither does Firefox. The logical integration of that level of security at the application level is the capabilities-based security, and you can prove that applications have no access to privileged information, or that they have access to a minimal, well-defined set of privileged information. Mainframe-style systems, SELinux and Java all can work like this.
Wait, are you implying that people who work for the government might be self-serving parasites?
By making this information available for free, those people who made a living by reproducing it are being put out of a job. And government jobs often attract the kind of people who would are just not adaptable enough to find a new job. So of course they are going to fight this -- because accepting it means they would have to change, and change is hard.
The Toughbook C30s are about as good as you are going to find in a fully rugged portable, but expend to spend about $4000 for a loaded unit.
That said, this guy: http://www.strikingviking.net/ went around the world with a Toshiba Tecra. He apparently had serious problems with a mac and did OK with a vaio.
This would make the risk of a lawsuit -- even a righteous one -- enormous.
Put this in individual terms -- let's say your neighbor's kid broke your window with a baseball, so you want to sue for a new one. But as part of that lawsuit, you would lose your house to the courts if you lose.
You saw your neighbors kid do it, so you go to court, but your neighbor realizes this is his opportunity to really screw you for daring to posit that his kid did something wrong -- so he hires a shark who pays someone off to come up with a fake alibi. And now you're renting a crappy apartment on the edge of town.
Now, you said copyright law - so -
Imagine if Evilsoft hired some talented programmers to write source code that would compile to similar binary code as something the Good Software Foundation had published under a copyleft. The intent was to entice the Good Software Foundation into sueing Evilsoft, so that Evilsoft could show the source code in court, prove their innocence, and make the Good Software Foundation go away forever.
From TFA:
"In the human trials, due to start this summer, the scientists will use levels of infra-red that occur naturally in sunlight."
All code SHOULD be signed, with l33t ASCII art!
This is silly. The GPL is not that nefarious.
All distribution ultimately relies on an "upstream" distributor.
Let's say you provide FTP access, but use a colocated server that you don't own -- you have an agreement with an upstream provider to provide bandwidth and disk space to provide your licensees with the source. Why does it matter if it's Rackspace or Ubuntu who serves as the upstream provider? Or if it's Memorex media and the Post Office or a floppy and delivered via sneakernet?
The intent here is clear: the entity distributing binaries to some users must also ensure that said users are able to come into possession of a copy of the source through reasonable means, be it downloading or making at-cost physical copies available. In other words, it seems to me that the distributor of binaries must only take some action for which the receipt of the desired source by the requesting licensee is a result.
It requires you to install a browser plugin, then the songs are played through that plugin -- it's fairly large at 1.1MB downloaded, 2.7mB installed -- it's called nprhapengine.so, which definately incorporates a lot of Helix code, and is also statically linked to some open-source software (OpenSSL, libcURL at least). The player works well enough, but the playlist is buggy -- it doesn't seem to support multiple playlists, or even deletion of individual songs from the playlist. And it didn't import the playlists used by the native Windows client (which are stored on the server). But then again, what's there works for me, and there is a "BETA" icon in the upper left of the browser window.
I got the request to install the plugin as soon as I hit any of the play buttons (the arrows next to tracks) on Rhapsody.com; this is the case on Firefox and Safari.
Finished developing yet, eh...
Quite a bit of the most influential discoveries to date have been done by people under 25. Your brain is constantly developing throughout your entire life if you play it right. And brain cells are constantly dying. While teenagers may not have the experience and sense of causality that older adults do, you can't deny that many of the most focused scholars and athletes did amazing things as teenagers and young adults (and then faded into obscurity). Young people do have a level of focus and creativity that is rarely seen amongst older people with brains atrophied by a blue collar job followed by a night of TV and drinking (not to mention age; as soon as you're born you start dying). Fact is, while you may have more experience at 30, you were probably smarter and more focused at 20.
The brain is never "fully developed", the body never "matures", except in a relativistic sense defined by biologists to make their lives easier. Living is a process, growing old is a process, and there is no destination other than the grave. There's no summit or valley anyone reaches as they age or grow in experience; to think otherwise is a wishful appeal to the authority of age we bow to and desire as children.
Normal diesel engines in their standard configuration can not properly work with vegetable oil. You need modified glow plugs (cheap and easy) and fuel injectors (not cheap and easy) to avoid premature engine failure from improper injection and combustion. Standard fuel filters also won't properly deal with straight vegetable oil (maybe easy, maybe a nightmare), especially at cold temperatures or during extended storage when the oil will crystallize. However, if you're willing to go that route, and install a special tank that preheats (definately not cheap or easy) the vegetable oil before the engine is started, you can run SVO. Even if you don't want to pre-heat, you can mix the oil with a bit of gas or dino diesel to reduce the viscosity, and still get a combustible mixture. Regardless, improper configurations or improper mixtures can permanently damage an engine very quickly (a couple hundred miles).
Here's a nice link:
http://journeytoforever.org/biodiesel_svo.html
Your argument is flawed in that you neglect to mention that computer-driven models can generate hypothesis that can be *tested*, both retrospectively, and in the future.
In your example, you'll know the computer model is wrong if in a week you are still alive. On the other hand, if the model can make accurate predictions about historical data, without being based on that data, then you have evidence to the model's accuracy. For example, use a machine learning technique to train a system to predict a 1-year climate trend using data from 1900-1990. Then, see if the system can accurately predict trends from 1990-2005. If so, there's no evidence that it would be wrong when predicting a trend from 2005-2015, when trained from 1990-2005. Other similar tests might include training on even-numbered years, and then predicting climate for odd-numbered years, or training on non-leap years, and predicting for leap years.
When you have sufficient data, you can use rigorous statistical methods to say with a known confidence how accurate your methods are likely to be in making predictions (and I'm not just talking about accuracy and recall; you can validate a hypothesis much more rigorously). You can then make rational, scientific statements.
Think about it from another perspective:
Adobe is using this to reinforce their stance on the issue; whereas the black hat Skylarov could break the encryption, did so, and released it (illegally according to US law), Adobe takes the white hat stance, saying they *could* break and release the encryption, but do not because it is wrong. This decision merely supports their previous stance -- that violating the DMCA is wrong.
It's not ironic that Adobe refuses to commit the same crime that was committed against them -- it's reaffirming of their principles.
Software RAID *is* very often faster, especially on a modern CPU paired with an older design -- you don't buy HW RAID because it is faster, you buy it for battery backup and offloading of low level operations to conserve CPU time and bus/memory bandwidth for user applications and so that if your OS or CPU/memory/whatever blows up, or you lose power, it won't corrupt the data on your disk array. Hardware RAID dedicated processors are simple, slow, "reliable" units -- not ultra-fast bleeding-edge dedicated units like you see on video cards.
The problem with using CL for heavy computing tasks is that Lisp is a linklist-based language. Linklists are very elegant to the programmer, yet they are extremely time-inefficient. A FORTRAN compiler makes good use of the fact that, in integer_4 array foo, foo(bar) is exactly 4 bytes away from foo(bar+1).
Common Lisp code is list-based, and these lists are compiled into regular assembly code by any good compiler, much like C or Fortran. In addition, Lisp makes using linked lists as data structures very easy -- and modern compilers do an excellent job of optimizing them. However, Common Lisp also supports simple and complex arrays, data structures, and objects -- none of which are generally implemented using linked lists, and all of which tend to be rather efficient. In fact, any good Lisp textbook will tell you that code which relies on non-fix-lengthed lists (fix-lengthed lists can be implemented using arrays) is going to be inefficient, and is hopelessly unoptimal.
Your argument might have been true 20+ years ago, but is no longer valid. Compiler technology has come very far. Note for example, that there exists floating-point example code for cmucl which is faster than equivalent C code, compiled with gcc.
Common Lisp is a very high level language with a tremendous amount of expressiveness, and it is suited towards academia in that in general, functionality is not sacrificed for performance.
Check out http://www.lisp.org, http://cmucl.cons.org/cmucl for a really good implementation (and there are even Debian packages of it).
CL is not known for its parallelization abilities, but if you need a language that lets you describe mathematics, CL is useful.
Lisp is actually based around something called the Lambda Calculus, which is a way of expressing concepts by transforming data into other data using data which is expressed as a "function". Because of this, Lisp has a lot of abilities that other languages lack, such as extremely simple and powerful function composition, even at run-time. CL also has a massive core library with OO facilities, basic mathematic primitives, good FFT suppot in most implementations, windowing system support, and good commercial vendors like Franz. Check it out; it's almost as old as Fortran, but has evolved in a much more elegant manner.
... And we all saw what happened to OS/2, which ran many MS Windows 3.1 applications better than Windows 3.1 did.
But Linux is not about gaming; gamers end up selling their souls one way or another, so why worry?
A Sun IPX (or any lunchbox style) system with an AUI port and a modified transceiver is much better. I use one of these as a secure syslog; in particular because you can modify the transceiver so that while it is capable of receiving data, it is incapable of sending at a hardware level. There is no way, short of physical access, to detect the machine. It's great for packet sniffing and logging -- syslog using UDP is connectionless, and works well with read-only network connections. This is also better than modifying the ethernet cable, because these modified cables do not actually work properly (the transceiver with tx pins removed will keep a valid *empty* tx signal, whereas a modified cable usually just pumps the rx'd signal back to tx, confusing the equipment into maintaining a link).
And if you can sneak in once, why not twice? Or better, equip the computer with a cell modem or amateur radio equipment (How many "wartalkers" look for that, eh?) , and dial in. No need for probes which may set off IDS systems, or outgoing packets (like ARP or DNS requests) that alert crackers to a computer's presence.
I think you cut pins 3 and 10 (on the connector to the computer on the transceiver) but that's not certain.
Compaq and HP sell crappy hardware. Period. Unfortunately, businesses as of late seem to have taken to the trend of using laws as a last line of defense for their failing business model.
I wonder if any of the RIAA's "reduced" profit figures are because of boycotts from people they pissed off... maybe it does work.