Modern office buildings generally don't have windows that can be opened.
To steal an idea from Hitchhiker's Guide to the Galaxy, I imagine most office buildings, which act basically as giant greenhouses, would get very unpleasantly hot if the air conditioning failed.
I'll bet this also aggravates adult allergies, as people are placed in a pollen-free environment for eight hours a day.
I agree that the military make lousy police officers -- no one in the military I know is keen on the idea. Where I disagree is the idea that searching for Al Qaeda operatives is strictly a law enforcement activity. By conventions of war they are spies or saboteurs, and we'd be well within our rights to summarily execute them, just as we did with German agents during WWII. (Well, we actually sent them to military tribunals, which is more than is required. Simply put, if you find a spy, you can shoot them.)
See, here's the thing. That sounds great in the abstract, but it's also a rather abusable approach (and has been abused in many countries before). "Shoot this guy, he's a spy...oh wait, maybe he wasn't...".
The reason foreigners don't recieve such protection is that it's expected that their country will look out for their interests (and has, as a if-necessary tool of power, national pressure, allowing embargoes, military force, and so forth). The problem is that if you allow the current government/military leaders power over US civilians without check or oversight, you give them power over the people that are expected to act as a check on out-of-control leaders (by expressing disapproval and voting people out of office). It's easy to create a feedback loop, where powerful people get into powerful positions and use their positions to ensure that they cannot be removed from said positions.
The Fed is asking for powers without oversight or limit with the justification that "you can trust us; we're the government."
Except we clearly *can't* -- take the current terrorist fiasco, slavery laws, or McCarthyism -- stuff that most people today consider pretty awful cases of those in charge massively exploiting people. The idea of those people losing the protections granted them is quite disturbing.
So why, why, why are we giving up our freedom now in the fight against a couple of thousand assholes living in CAVES, for God's sake?
Because Bush realized that there hadn't been a president to stir up and exploit irrational fear in a country for a long, long time, and it was quite a profitable deal to do so.
Except instead of treating this as a law enforcement issue and assisting or coercing countries into imposing legal force, Bush promptly declared a "war on terror" was underway, started a far larger deployment of military hardware and invaded and overthrew two countries, killing and deposing all sorts of people that had nothing to do with "terrorism".
There's too much benefit to the people implementing them, and it's too easy to do and too hard to prevent people from tying databases together.
On the other hand, it's very easy to hold up a guy who turns out to be military intelligence (weren't those the guys who instigated the recent prison abuse problems?) and say "Hey, someone's not following the rules".
There are force-feedback-supporting USB HID adapters for just about every force-feedback console joystick out there (I would assume the XBox controller is included).
Linux supports force feedback on HID devices.
(non) Dangers of using ATA or SATA for Raid
on
SATA vs ATA?
·
· Score: 2, Interesting
Trying to remedy this by turning off write-back caching severly impacts the performance of the drives and some vendors do not certify the recovery of drives that deactivate write-back caching so this may increase failure rates.
I don't buy this argument one bit.
I agree with you that write-back can break journalling FS guarantees.
However, I don't know of any consumer drive vendor that guarantees that their write-back algorithms are in-order. This means that write-back can trash *any* filesystem, and whether it be RAID or not be damned.
Write-back should *never* be on on drives using modern filesystems.
As for an impact on performance, I call foul again. The write-back cache benefits are useful only in the presence of an OS that does poor disk caching. Take a nice Linux box -- it'll use all available free memory as a big fat writeback cache. There is only a single advantage to using a drive's native writeback controller -- the drive knows the true geometry of the disk (not whatever fantasy geometry it's handed off to the host), and furthermore knows the performance characteristics (settle time, seek times, etc) of the drive. That's useful, but it's not comparable to having ten times or more the amount of memory for buffering.
Hard drive vendors would be *much* better off from a performance standpoint exporting a profile of their drive's performance characteristics to the host -- "settle time on the drive can be determined by this function, seek time can be determined by this function, this is the real geometry", etc. Then, the much more powerful (in both memory, CPU, and code size) host could do whatever scheduling it wanted to try out.
They have the gold master, which can be copied and sold as pirate software before the game even gets to the shelves, and with the delay even earlier before it gets to the shelves.
Perhaps, but honestly, of how much value is that?
As someone else pointed out, the golden master is nothing more or less than what will appear in the retail boxes. It doesn't have the source or anything on it.
Sure, it gives a *slight* jump (maybe three weeks, based on a quick Google on game GM-to-shelf time) over having a guy at the store send Razor 911 a CD image to work with, but this isn't make-or-break material.
Now, a leaked beta or devel copy (as has happened to many pieces of software before) is much more likely to be damaging, as it may be represented as the final game and give people a bad impression of the software (which spreads via word of mouth), plus a much longer delay, perhaps on the order of several months, before the game can hit shelves.
For example, Bungie has had people steal laptops containing demo versions of Marathon. I know people that got ahold of Oni well before release, so I assume that that somehow leaked as well. Despite that, Bungie is doing quite well for themselves (well, aside from having been assimilated by the Borg, but doing so is generally considered a business success story).
Yes, but the thing is -- once you use GNU/Linux, you don't *want* to use other stuff -- but that's okay, because you can download all the GNU/Linux you want for free and will always be able to do so.
Microsoft tries very hard to get product lock-in at a customer, then extracts more money than the initial purchase appears to be.
To be fair, the tactics they are using against employees of their customers that judge their product harshly are also unbearable.
As someone else pointed out, Ballmer said that "Linux is a cancer" quite recently -- this is hardly worse than "Microsoft uses the business plans of drug dealers".
I think that he was just commenting that Microsoft used the same approach that drug dealers do -- to give away cheap or free product to produce a dependence, and then to take advantage of that dependence.
It's hardly unreasonable or untrue (though it might well be damaging) and would be entirely legal under US law.
They are just like the linux guys who threw so much FUD at Microsoft that they've retaliated so hard against us. From the beginning, Linux guys have made their goal the destruction of Microsoft. That attitude is what I think created or exacerbated the problem we have today.
That dislike of Microsoft is a product of putting up with years of Microsoft's abuse of their customers, not because everyone suddenly decided that "OS vendors should not have names beginning with M, and everyone in such a class should be destroyed".
If the problem derives from a driver, you'd boot into VGA mode. You'd be able to get administrative access to the OS without ever using the card's driver.
I agree with other people that checking cables is underrated. I've spent quite some time considering all sorts of bizarre causes when the problem is just a bad/loose cable.
And you know they will, because the people behind these outfits are the same types behind SCO and Enron.
John Carmack has written some of the most groundbreaking entertainment software out there.
He has donated his old engines to the world, GPLing them.
Id has stayed small and privately-owned deliberately, and avoided the problems with "shareholder short-term return issues" that so many people complain about.
He has spent extensive amounts of time and effort doing volunteer code on 3d drivers for XFree86, allowing Linux and BSD folks to enjoy 3d games. He used his influence to help get Linux and BSD folks the games that they have today.
He has pushed hard for technological improvements in the GPU arena, and has done consumer education on GPU features. He is famously open about what he is working on and his thoughts (the Carmack.plan is regular reading material for many) -- one of the main things that OSS adherents want to see in the software world.
And you call him one of the same types of people behind SCO and Enron? That is not only absurd, it's an attack on one of the better men in the software world. I question whether *you* have done as much for society.
Re:Can I have an infinite budget to write the code
on
Java Faster Than C++?
·
· Score: 1
Unix GUIs are far easier to implement with Java than with C++. Athena/Xlib are a huge fargin' hassle and no-one ever gets the widgets right. Motif is too expensive to license.
Whoah, there. This was right...a number of years ago. GTK and Qt have neatly reversed this. It's much more of a pain in the ass to do, say, a Win32 GUI than it is a GTK or Qt GUI.
there is not necessarily any "extra layer" created by the JVM. the whole idea is that the JVM is actually a run-time compiler, and when it's done compiling the.class to native code you really have native code directly executing.
No overhead added by the JVM? I'll be impressed to hear about how you implement, say, polymorphism and reflection without some overhead.
the JVM runtime compiler can perform optimizations that are not available to a C++ compiler. for example, if the JVM realizes that there is only one instance loaded of an abstract/virtual class it can compile all the code that accesses it statically against that single code, as if there were no inheritence at all, saving a pointer reference. a C++ compiler can never do that because it does not know what you will link against.
And the C++ compiler can perform optimizations that the Java one cannot because the Java one has to have certain runtime performance guarantees.
Many of you are simply going to find ways to criticize Java because you are religious. You used to criticize performance, and if that is taken away from you, you will say it was never really important and criticize something else. You need to think whether or not you are being objective and rational, or simply theocratic.
Have you considered the opposite, that you might be fighting a religious war yourself? There are basically no horizontal-market Java apps despite years of Sun marketing, the few attempts at doing so have failed because of massive resource (CPU and memory) demands, promises of "faster performance with this *new* technology, really, this time" have been made over and over for years. I use the IBM JDK, currently the fastest JVM out for Linux, to run G2 GUI, a front end for mldonkey, and freenet. Both of these two programs run quite slowly and use a significant chunk of memory -- much more than my C daemons and front ends do.
I agree that Java has the potential for certain performance improvements that C/C++ does not, but in the real world, Java really does significantly slower for reasonable tests.
I've attended lectures in which I've heard distinguished language people break down performance issues with Java, and spoken with language people that have focused heavily on Java, and I'm reasonably comfortable with a claim that Java will not approach C/C++ performance for real-world tasks. Performance just was not a focus during Java's design. You can clearly see, looking at the featureset in C++, that almost no features that cause overhead were chosen, and those that were cause relatively minimal overhead. Java has a number of fundamental design decisons that cause performance hits (lack of generics (according to another post, which I have not yet verified, Java's 1.5 generic support is syntactic sugar and cannot be used for optimization), bounds-checking, heavy reliance on dynamic typing, a design that encourages frequent object creation and deletion, and so forth). Java was made to have a large number of whizz-bang features (which it does), have a strong set of functionality in the base language (which it does), be reasonably syntax-compatible with C++ to allow C++ coders to relatively easily use it (which it does). The strict observance of only adding features that induce compile-time overhead was not made (read a book on C++ or an ML variant and think to yourself how the compiler writer actually implements each feature -- there is a very strong emphasis on features that impose only such overhead).
This does not mean that I think that Java is useless. I have written software in Java (by choice, where I had complete control over language choice) because it has a number of good points. I do *not* think that it is a C/C++ replacement (as many people have tried to bill it, and as Apple is trying to get people to use it). It just does not have an acceptable level of performance -- even if you have a fast computer, the user might as well be using the computer before his
As was pointed out in the many comments concerning Java's speed when compared to C++, that hash code for C++ is seriously screwed up. It leaks memory like crazy, and one of the examples doesn't even do the same thing as the Java example because the code uses hash_map::operator[]() to test for element existence, instead of hash_map::count().
Funny that you should say that, because hash tables are one of the easiest ways to "leak" memory in Java -- to have stray references floating around.
Slashdot Mirror
Modern office buildings generally don't have windows that can be opened.
To steal an idea from Hitchhiker's Guide to the Galaxy, I imagine most office buildings, which act basically as giant greenhouses, would get very unpleasantly hot if the air conditioning failed.
I'll bet this also aggravates adult allergies, as people are placed in a pollen-free environment for eight hours a day.
I agree that the military make lousy police officers -- no one in the military I know is keen on the idea. Where I disagree is the idea that searching for Al Qaeda operatives is strictly a law enforcement activity. By conventions of war they are spies or saboteurs, and we'd be well within our rights to summarily execute them, just as we did with German agents during WWII. (Well, we actually sent them to military tribunals, which is more than is required. Simply put, if you find a spy, you can shoot them.)
See, here's the thing. That sounds great in the abstract, but it's also a rather abusable approach (and has been abused in many countries before). "Shoot this guy, he's a spy...oh wait, maybe he wasn't...".
The reason foreigners don't recieve such protection is that it's expected that their country will look out for their interests (and has, as a if-necessary tool of power, national pressure, allowing embargoes, military force, and so forth). The problem is that if you allow the current government/military leaders power over US civilians without check or oversight, you give them power over the people that are expected to act as a check on out-of-control leaders (by expressing disapproval and voting people out of office). It's easy to create a feedback loop, where powerful people get into powerful positions and use their positions to ensure that they cannot be removed from said positions.
This really isn't hyperbole.
The Fed is asking for powers without oversight or limit with the justification that "you can trust us; we're the government."
Except we clearly *can't* -- take the current terrorist fiasco, slavery laws, or McCarthyism -- stuff that most people today consider pretty awful cases of those in charge massively exploiting people. The idea of those people losing the protections granted them is quite disturbing.
"regulation" and "law" are synonyms in common English.
Not sure what the legal meaning of this is, though.
So why, why, why are we giving up our freedom now in the fight against a couple of thousand assholes living in CAVES, for God's sake?
Because Bush realized that there hadn't been a president to stir up and exploit irrational fear in a country for a long, long time, and it was quite a profitable deal to do so.
Except instead of treating this as a law enforcement issue and assisting or coercing countries into imposing legal force, Bush promptly declared a "war on terror" was underway, started a far larger deployment of military hardware and invaded and overthrew two countries, killing and deposing all sorts of people that had nothing to do with "terrorism".
Ah, but see, this isn't law enforcement. It's a war on terror, and war is clearly the department of military control.
And military stuff isn't subject to the degree of oversight that civilian law enforcement, like the FBI, is.
Cute.
TIA-style programs are probably unstoppable.
There's too much benefit to the people implementing them, and it's too easy to do and too hard to prevent people from tying databases together.
On the other hand, it's very easy to hold up a guy who turns out to be military intelligence (weren't those the guys who instigated the recent prison abuse problems?) and say "Hey, someone's not following the rules".
It should already do so.
There are force-feedback-supporting USB HID adapters for just about every force-feedback console joystick out there (I would assume the XBox controller is included).
Linux supports force feedback on HID devices.
Trying to remedy this by turning off write-back caching severly impacts the performance of the drives and some vendors do not certify the recovery of drives that deactivate write-back caching so this may increase failure rates.
I don't buy this argument one bit.
I agree with you that write-back can break journalling FS guarantees.
However, I don't know of any consumer drive vendor that guarantees that their write-back algorithms are in-order. This means that write-back can trash *any* filesystem, and whether it be RAID or not be damned.
Write-back should *never* be on on drives using modern filesystems.
As for an impact on performance, I call foul again. The write-back cache benefits are useful only in the presence of an OS that does poor disk caching. Take a nice Linux box -- it'll use all available free memory as a big fat writeback cache. There is only a single advantage to using a drive's native writeback controller -- the drive knows the true geometry of the disk (not whatever fantasy geometry it's handed off to the host), and furthermore knows the performance characteristics (settle time, seek times, etc) of the drive. That's useful, but it's not comparable to having ten times or more the amount of memory for buffering.
Hard drive vendors would be *much* better off from a performance standpoint exporting a profile of their drive's performance characteristics to the host -- "settle time on the drive can be determined by this function, seek time can be determined by this function, this is the real geometry", etc. Then, the much more powerful (in both memory, CPU, and code size) host could do whatever scheduling it wanted to try out.
They have the gold master, which can be copied and sold as pirate software before the game even gets to the shelves, and with the delay even earlier before it gets to the shelves.
Perhaps, but honestly, of how much value is that?
As someone else pointed out, the golden master is nothing more or less than what will appear in the retail boxes. It doesn't have the source or anything on it.
Sure, it gives a *slight* jump (maybe three weeks, based on a quick Google on game GM-to-shelf time) over having a guy at the store send Razor 911 a CD image to work with, but this isn't make-or-break material.
Now, a leaked beta or devel copy (as has happened to many pieces of software before) is much more likely to be damaging, as it may be represented as the final game and give people a bad impression of the software (which spreads via word of mouth), plus a much longer delay, perhaps on the order of several months, before the game can hit shelves.
For example, Bungie has had people steal laptops containing demo versions of Marathon. I know people that got ahold of Oni well before release, so I assume that that somehow leaked as well. Despite that, Bungie is doing quite well for themselves (well, aside from having been assimilated by the Borg, but doing so is generally considered a business success story).
Yes, but the thing is -- once you use GNU/Linux, you don't *want* to use other stuff -- but that's okay, because you can download all the GNU/Linux you want for free and will always be able to do so.
Microsoft tries very hard to get product lock-in at a customer, then extracts more money than the initial purchase appears to be.
This is true and sheer genius. Dunno about Brazilian law, but this might even be usable in a US court.
To be fair, the tactics they are using against employees of their customers that judge their product harshly are also unbearable.
As someone else pointed out, Ballmer said that "Linux is a cancer" quite recently -- this is hardly worse than "Microsoft uses the business plans of drug dealers".
I think that he was just commenting that Microsoft used the same approach that drug dealers do -- to give away cheap or free product to produce a dependence, and then to take advantage of that dependence.
It's hardly unreasonable or untrue (though it might well be damaging) and would be entirely legal under US law.
Are you basing this on your experiences with marijuana or with heroin?
They are just like the linux guys who threw so much FUD at Microsoft that they've retaliated so hard against us. From the beginning, Linux guys have made their goal the destruction of Microsoft. That attitude is what I think created or exacerbated the problem we have today.
That dislike of Microsoft is a product of putting up with years of Microsoft's abuse of their customers, not because everyone suddenly decided that "OS vendors should not have names beginning with M, and everyone in such a class should be destroyed".
They should rename it "Linus/GPL", since, after all, it's getting the most press coverage from Linux. ;-)
If the problem derives from a driver, you'd boot into VGA mode. You'd be able to get administrative access to the OS without ever using the card's driver.
I agree with other people that checking cables is underrated. I've spent quite some time considering all sorts of bizarre causes when the problem is just a bad/loose cable.
Only a rare few of them have the human skills to effectively communicate and lead.
It's a shame that they don't wind up as managers -- not enough backstabbing skills.
My wife is hiring for a low-level Python programming position. She can't seem to find anyone who knows Python who is willing to work for under $100K.
Really and honestly, I think that if you provided a link to said job opening in your Slashdot post, she might get a couple of applicants.
It'd help them, help your wife, everyone wins.
And you know they will, because the people behind
.plan is regular reading material for many) -- one of the main things that OSS adherents want to see in the software world.
these outfits are the same types behind SCO and
Enron.
John Carmack has written some of the most groundbreaking entertainment software out there.
He has donated his old engines to the world, GPLing them.
Id has stayed small and privately-owned deliberately, and avoided the problems with "shareholder short-term return issues" that so many people complain about.
He has spent extensive amounts of time and effort doing volunteer code on 3d drivers for XFree86, allowing Linux and BSD folks to enjoy 3d games. He used his influence to help get Linux and BSD folks the games that they have today.
He has pushed hard for technological improvements in the GPU arena, and has done consumer education on GPU features. He is famously open about what he is working on and his thoughts (the Carmack
And you call him one of the same types of people behind SCO and Enron? That is not only absurd, it's an attack on one of the better men in the software world. I question whether *you* have done as much for society.
Unix GUIs are far easier to implement with Java than with C++. Athena/Xlib are a huge fargin' hassle and no-one ever gets the widgets right. Motif is too expensive to license.
Whoah, there. This was right...a number of years ago. GTK and Qt have neatly reversed this. It's much more of a pain in the ass to do, say, a Win32 GUI than it is a GTK or Qt GUI.
there is not necessarily any "extra layer" created by the JVM. the whole idea is that the JVM is actually a run-time compiler, and when it's done compiling the .class to native code you really have native code directly executing.
No overhead added by the JVM? I'll be impressed to hear about how you implement, say, polymorphism and reflection without some overhead.
the JVM runtime compiler can perform optimizations that are not available to a C++ compiler. for example, if the JVM realizes that there is only one instance loaded of an abstract/virtual class it can compile all the code that accesses it statically against that single code, as if there were no inheritence at all, saving a pointer reference. a C++ compiler can never do that because it does not know what you will link against.
And the C++ compiler can perform optimizations that the Java one cannot because the Java one has to have certain runtime performance guarantees.
Many of you are simply going to find ways to criticize Java because you are religious. You used to criticize performance, and if that is taken away from you, you will say it was never really important and criticize something else. You need to think whether or not you are being objective and rational, or simply theocratic.
Have you considered the opposite, that you might be fighting a religious war yourself? There are basically no horizontal-market Java apps despite years of Sun marketing, the few attempts at doing so have failed because of massive resource (CPU and memory) demands, promises of "faster performance with this *new* technology, really, this time" have been made over and over for years. I use the IBM JDK, currently the fastest JVM out for Linux, to run G2 GUI, a front end for mldonkey, and freenet. Both of these two programs run quite slowly and use a significant chunk of memory -- much more than my C daemons and front ends do.
I agree that Java has the potential for certain performance improvements that C/C++ does not, but in the real world, Java really does significantly slower for reasonable tests.
I've attended lectures in which I've heard distinguished language people break down performance issues with Java, and spoken with language people that have focused heavily on Java, and I'm reasonably comfortable with a claim that Java will not approach C/C++ performance for real-world tasks. Performance just was not a focus during Java's design. You can clearly see, looking at the featureset in C++, that almost no features that cause overhead were chosen, and those that were cause relatively minimal overhead. Java has a number of fundamental design decisons that cause performance hits (lack of generics (according to another post, which I have not yet verified, Java's 1.5 generic support is syntactic sugar and cannot be used for optimization), bounds-checking, heavy reliance on dynamic typing, a design that encourages frequent object creation and deletion, and so forth). Java was made to have a large number of whizz-bang features (which it does), have a strong set of functionality in the base language (which it does), be reasonably syntax-compatible with C++ to allow C++ coders to relatively easily use it (which it does). The strict observance of only adding features that induce compile-time overhead was not made (read a book on C++ or an ML variant and think to yourself how the compiler writer actually implements each feature -- there is a very strong emphasis on features that impose only such overhead).
This does not mean that I think that Java is useless. I have written software in Java (by choice, where I had complete control over language choice) because it has a number of good points. I do *not* think that it is a C/C++ replacement (as many people have tried to bill it, and as Apple is trying to get people to use it). It just does not have an acceptable level of performance -- even if you have a fast computer, the user might as well be using the computer before his
As was pointed out in the many comments concerning Java's speed when compared to C++, that hash code for C++ is seriously screwed up. It leaks memory like crazy, and one of the examples doesn't even do the same thing as the Java example because the code uses hash_map::operator[]() to test for element existence, instead of hash_map::count().
Funny that you should say that, because hash tables are one of the easiest ways to "leak" memory in Java -- to have stray references floating around.