The only reason I read the comments on this story was to figure out what the heck Microsoft could have been doing all this time.
I think maybe the Windows codebase has simply finally reached a level of complexity that renders it unmanagable by mortal humans. To quote an anonymous poster to the linked blog:
Today's announcement is of course no surprise to anyone inside MS. The only surprise is that it was such a short delay announced.Basically we do not believe Vista will make January 2007 or even March 2007. Anyone with any access knows what a frankenstein's monster NT is on the inside. At some point there is a law of diminishing returns trying to do anything to it at all, it seems like that limit is being reached today. The release is pushed back because of bugs but fixing those bugs will create more bugs. It is just godawful to be honest.
Assuming that is true, then probably the only way for Microsoft to move forward and still maintain backwards compatibility with old code is to do what Apple did: Ditch the OS, start fresh with a new one, and provide backwards compatibility with existing Windows applications by shipping the "legacy OS" as an included software application that runs in an emulator. Given the prevalence of VMWare-style technology these days, that should be quite doable; of course getting the new OS up to snuff might take a few years.
Forget waste of space in something as small as a sector. If this is an issue, you use the wrong application - one word file per phone number?
The only reason having "one file per phone number" is considered unreasonable is because (traditionally) filesystems couldn't support that usage pattern efficiently. If you had a file system that could support, say, 20 million 10-byte files in a directory without unacceptable overhead, that would be a perfectly valid design for your app.
Look at it this way: existing filesystems force application developers to reinvent the wheel for every application, by implementing a separate database layer on top of the filesystem. Wouldn't it be nicer if the filesystem itself was powerful enough to act as a basic database on its own?
BeFS and ReiserFS were both moves in this direction, and I think it's a good idea.
at one chop shop, where I was a sysop, we would routinely order the smallest drive within a series and when it arrived, swap the logic from the failed, larger capacity drive, from the same series & viola!... more megs
Kind of reminds me of turning a low-density 3.5" floppy into a "high-density" one, with a hole punch. I hope you weren't storing anything important on those drives!
Meanwhile, back in non-fanboy reality, most major Windows programs were multithreaded for backgroaund tasks 10 years ago
I think you missed the point. The point wasn't that it is possible to write multithreaded code under OS/X -- obviously, that is possible under any modern OS. The point was that for many operations under OS/X you don't have to write multithreaded code to get the benefits of multiprocessing: you just call the regular system libraries and the multithreading goes on "behind the scenes". Your program neither knows nor cares about the number of processors being used, the only difference it notices is that the OSX::SuperRenderImageFunction() system call returns much more quickly when running on a dual-CPU system.
I had a boss who loved to get dual-CPU systems. Why? "Because that way one CPU can run the web server and one CPU can run the database." No matter how often I tried to shake that view from his head it never left. (In point of fact, both were context switching in and out of both CPUs pretty regularly)
Assuming your boss's scheduler was smart enough to load-balance across CPUs, he was probably right. The two processes may have been bouncing around the CPUs a lot, but at any instant of time (assuming these two processes were both CPU bound and no other processes were running) one CPU would indeed be running process A and the other, process B.
Of course, the fact that neither of these tasks was likely to be CPU-bound kind of throws a wrench into the works... your boss probably should have spent the money on more RAM or a RAID instead.
On my personal system, a busy process brings Outlook to its knees. If I had a dual-core system, I wouldn't have a problem.
I've experienced the same sort of thing myself, which is one reason why I've always preferred dual-processor computers in the past.
On the other hand, though, the only reason the 2-CPU system stays responsive when one CPU is pegged is because having the CPUs separated places a natural barrier on how much of the total available CPU time a single process can chew up. That's handy, but there's no reason you can't implement the same function in software: just make the OS's scheduler smart enough to not allow a single runaway process to degrade the system. You could do it crudely, by simply not allowing any process to use more than 50% of the (single) CPU's cycles, or more elegantly, by automatically deprioritizing CPU-hogs. So given modern process schedulers, I don't think this argument should apply anymore.
With dual processors, Norton AV is grabbing more than 100% CPU at times, even with their own throttling settings enabled.
I think Norton AV (and other products like it) are a hopelessly flawed way to try to provide "security". No matter how many CPU cycles they burn trying to detect viruses, there will always be a new virus with a new level of obfuscation that will slip past them. Therefore, dual core CPUs won't be sufficient for this task, because any number of CPUs would not be sufficient -- an unbounded number of CPU cycles (or a solution to the Halting Problem) would be required to do it. The only way to provide reliable security is to carefully design the OS to be secure, so that it doesn't matter if a virus runs -- because the virus won't be able to do any harm anyway.
That said, I think dual core CPUs are great, because it means (or will soon mean) that I can get a 2 processor machine for the price of a 1-processor machine, or a 4-processor machine for the price of a 2-processor. Eventually we'll have things like 64 processors on a chip, which will be great fun to play with -- my own Connection Machine, for cheap!:^)
If the information exists, someone is looking at it.
Tell that to the shoebox of 3.5" AmigaDOS-formatted floppies that is mouldering in the back of my closet. (it would be somewhat difficult these days to even find a floppy drive that could read them, and I'm not sure how reliably readable they would be even then...)
True, but the quadrupling of pork projects during the last 5 years has been quite impressive (especially considering that Republicans, ostensibly "the party of fiscal restraint" have been in control of the government for that time). And of course, Bush has yet to veto anything....
The Kyoto Protocol always was and always will be useless.
Sort of like how Goddard's toy rockets were completely useless as a means of getting man to walk on the moon. I think the main benefit of the Kyoto Protocol was to show that the human race was taking the problem seriously and was capable of co-operating (technically and politically) to try and solve it. Once we had Kyoto as a working prototype/test bed, data could be gathered on what works and what doesn't, and more advanced/effective protocols devised based on the experience.
Of course, some might argue that Kyoto demonstrated mainly the opposite thing, but I suspect that if/when the human consequences of the environmental problems get worse, the world will have more incentive to face up to the problem.
People want to believe that human actions cause global warming
Really? Everything I've seen indicates that people really really want to believe that (a) global warming doesn't exist, or (b) if it does exist, it's not really a problem, or (c) if it is a problem, it's not humanity's fault.
It just doesn't seem logical to think that human actions cause global warming. This article uses data from, what 40 years? And the Earth is 4 billion years old? How can you make any kind of assumption on 0.000001 % of the total data?
If the Earth has been behaving one way for millions/billions of years, and then in the last eye-blink of time (~150 years), both the Industrial Revolution and a sudden change in climatic behaviour occur simultaneously, I'd say that's at least a hell of a coincidence. Note that ice core data goes back much longer than 40 years.
um... those would be RADIO waves... not sound waves. Which sort of follow similar rules (being waves and all), but are definitely different...
I don't know, I think a wireless LAN based on sound waves might be kind of neat. At the very least it could be used to drive the local users (and/or their dogs) crazy...
Re:Strange people are comparing this to video game
on
Playing The Escape
·
· Score: 2, Informative
This could have been done a century ago (different plot of course).
What would you have used a century ago, in place of the RFID player tracking, audio/video playback, and all the other various automation? A horde of employees hiding behind the scenery? Perhaps, but I doubt it would be very profitable.
Any system in which you add more energy will have more force. So when the oceans warm up OF COURSE the hurricanes will be stronger.
Given the above, I wonder about the wisdom of rebuilding New Orleans "back the way it was". Are we going to spend $89 billion+ on reconstruction, only to have another Katrina happen in a year or two (and repeat the process every few years)?
Fortunately I'm a free citizen of a free country (Russia), about the only place in the World where I can respectfully disagree with the US
A bit off topic, but this reminds me of a little joke from the Cold War era... a Russian and an American are debating who has the more open society. The American says "in the USA, we have freedom of speech. If I think Richard Nixon is a bastard, I won't be arrested for saying so." The Russian retorts, "it's just the same in Russia. We too can criticize Nixon whenever we want!"
What American has been incarcerated without trial? That was not, you know, fighting for the Taliban?
The whole point of having a trial is to determine whether a person is guilty or not. Locking people up without due process of law is unconstitutional and illegal. Period.
The act of siding with the enemy is a voluntary surrender of citizenship and the accompanying civil rights.
Can you point to the portion of the Constitution that spells that out? Or are you just making up rules that you wish were so?
just wish the movie was better. I recently grabbed it on DVD, thinking it couldn't be so bad...
My advice is, go back and watch it a second time. When I saw the movie in the theaters, I was also dissappointed, as I suspect every Guide fan was... that was because I was constantly comparing it to the books, and noting where it fell short. A few months later, I watched it again on DVD, and that time I was no longer constantly comparing it to the books, but rather watching it solely as a movie in its own right. The second time I found it quite funny and enjoyable.
Why do they need to go the pub if Ford already has brought beer?
Actually, the big issue many people in the West have is that 1.3 billion people on Earth justify, excuse, aid, abet and harbor those few political extremists.
Oh, come on. If every Muslim in the world is guilty of that, then every Christian is guilty of "justifying, excusing, aiding, abetting, and harbouring" Timothy McVeigh and the abortion-clinic bombers. Just because 1.3 billion people seem like a big homogenous hive-mind to you doesn't mean they are, it only means you are ignorant of the real situation.
Ah, but do their teachings specifically state that all 72 virgins will be female? If not, imagine the surprise some of these homicide bombers have in store when they get to Heaven and discover...
... a big crowd of Slashdotters waiting for them. At least heaven will have good LAN parties....
So if they can make fuel cells for ficken *laptops* why can't they do the same thing for cars?
They can, and do. Once the price comes down (or people become willing to pay $500,000 for a car), and the infrastructure issues get worked out, we will have them.
Slashdot Mirror
I think maybe the Windows codebase has simply finally reached a level of complexity that renders it unmanagable by mortal humans. To quote an anonymous poster to the linked blog:
Today's announcement is of course no surprise to anyone inside MS. The only surprise is that it was such a short delay announced.Basically we do not believe Vista will make January 2007 or even March 2007. Anyone with any access knows what a frankenstein's monster NT is on the inside. At some point there is a law of diminishing returns
trying to do anything to it at all, it seems like that limit is being reached today. The release is pushed back because of bugs but fixing those bugs will create more bugs. It is just godawful to be honest.
Assuming that is true, then probably the only way for Microsoft to move forward and still maintain backwards compatibility with old code is to do what Apple did: Ditch the OS, start fresh with a new one, and provide backwards compatibility with existing Windows applications by shipping the "legacy OS" as an included software application that runs in an emulator. Given the prevalence of VMWare-style technology these days, that should be quite doable; of course getting the new OS up to snuff might take a few years.
If this is an issue, you use the wrong application - one word file per phone number?
The only reason having "one file per phone number" is considered unreasonable is because (traditionally) filesystems couldn't support that usage pattern efficiently. If you had a file system that could support, say, 20 million 10-byte files in a directory without unacceptable overhead, that would be a perfectly valid design for your app.
Look at it this way: existing filesystems force application developers to reinvent the wheel for every application, by implementing a separate database layer on top of the filesystem. Wouldn't it be nicer if the filesystem itself was powerful enough to act as a basic database on its own?
BeFS and ReiserFS were both moves in this direction, and I think it's a good idea.
Kind of reminds me of turning a low-density 3.5" floppy into a "high-density" one, with a hole punch. I hope you weren't storing anything important on those drives!
I think you missed the point. The point wasn't that it is possible to write multithreaded code under OS/X -- obviously, that is possible under any modern OS. The point was that for many operations under OS/X you don't have to write multithreaded code to get the benefits of multiprocessing: you just call the regular system libraries and the multithreading goes on "behind the scenes". Your program neither knows nor cares about the number of processors being used, the only difference it notices is that the OSX::SuperRenderImageFunction() system call returns much more quickly when running on a dual-CPU system.
Assuming your boss's scheduler was smart enough to load-balance across CPUs, he was probably right. The two processes may have been bouncing around the CPUs a lot, but at any instant of time (assuming these two processes were both CPU bound and no other processes were running) one CPU would indeed be running process A and the other, process B.
Of course, the fact that neither of these tasks was likely to be CPU-bound kind of throws a wrench into the works... your boss probably should have spent the money on more RAM or a RAID instead.
I've experienced the same sort of thing myself, which is one reason why I've always preferred dual-processor computers in the past.
On the other hand, though, the only reason the 2-CPU system stays responsive when one CPU is pegged is because having the CPUs separated places a natural barrier on how much of the total available CPU time a single process can chew up. That's handy, but there's no reason you can't implement the same function in software: just make the OS's scheduler smart enough to not allow a single runaway process to degrade the system. You could do it crudely, by simply not allowing any process to use more than 50% of the (single) CPU's cycles, or more elegantly, by automatically deprioritizing CPU-hogs. So given modern process schedulers, I don't think this argument should apply anymore.
fork creates a new process, and the new process contains a thread (aka "the main thread"), therefore fork creates a thread.
I think Norton AV (and other products like it) are a hopelessly flawed way to try to provide "security". No matter how many CPU cycles they burn trying to detect viruses, there will always be a new virus with a new level of obfuscation that will slip past them. Therefore, dual core CPUs won't be sufficient for this task, because any number of CPUs would not be sufficient -- an unbounded number of CPU cycles (or a solution to the Halting Problem) would be required to do it. The only way to provide reliable security is to carefully design the OS to be secure, so that it doesn't matter if a virus runs -- because the virus won't be able to do any harm anyway.
That said, I think dual core CPUs are great, because it means (or will soon mean) that I can get a 2 processor machine for the price of a 1-processor machine, or a 4-processor machine for the price of a 2-processor. Eventually we'll have things like 64 processors on a chip, which will be great fun to play with -- my own Connection Machine, for cheap!
Tell that to the shoebox of 3.5" AmigaDOS-formatted floppies that is mouldering in the back of my closet. (it would be somewhat difficult these days to even find a floppy drive that could read them, and I'm not sure how reliably readable they would be even then...)
True, but I suspect the growth in monetary amount is similar.
Further, the definition of "pork" can be fuzzy, so the results may depend on who is doing the classification.
In this case, it was the Club for Growth, a Republican organization.
True, but the quadrupling of pork projects during the last 5 years has been quite impressive (especially considering that Republicans, ostensibly "the party of fiscal restraint" have been in control of the government for that time). And of course, Bush has yet to veto anything....
Sort of like how Goddard's toy rockets were completely useless as a means of getting man to walk on the moon. I think the main benefit of the Kyoto Protocol was to show that the human race was taking the problem seriously and was capable of co-operating (technically and politically) to try and solve it. Once we had Kyoto as a working prototype/test bed, data could be gathered on what works and what doesn't, and more advanced/effective protocols devised based on the experience.
Of course, some might argue that Kyoto demonstrated mainly the opposite thing, but I suspect that if/when the human consequences of the environmental problems get worse, the world will have more incentive to face up to the problem.
Really? Everything I've seen indicates that people really really want to believe that (a) global warming doesn't exist, or (b) if it does exist, it's not really a problem, or (c) if it is a problem, it's not humanity's fault.
It just doesn't seem logical to think that human actions cause global warming. This article uses data from, what 40 years? And the Earth is 4 billion years old? How can you make any kind of assumption on 0.000001 % of the total data?
If the Earth has been behaving one way for millions/billions of years, and then in the last eye-blink of time (~150 years), both the Industrial Revolution and a sudden change in climatic behaviour occur simultaneously, I'd say that's at least a hell of a coincidence. Note that ice core data goes back much longer than 40 years.
I don't know, I think a wireless LAN based on sound waves might be kind of neat. At the very least it could be used to drive the local users (and/or their dogs) crazy...
What would you have used a century ago, in place of the RFID player tracking, audio/video playback, and all the other various automation? A horde of employees hiding behind the scenery? Perhaps, but I doubt it would be very profitable.
You think Mars is bad, check out the greenhouse problems they have on Venus...
Given the above, I wonder about the wisdom of rebuilding New Orleans "back the way it was". Are we going to spend $89 billion+ on reconstruction, only to have another Katrina happen in a year or two (and repeat the process every few years)?
A bit off topic, but this reminds me of a little joke from the Cold War era... a Russian and an American are debating who has the more open society. The American says "in the USA, we have freedom of speech. If I think Richard Nixon is a bastard, I won't be arrested for saying so." The Russian retorts, "it's just the same in Russia. We too can criticize Nixon whenever we want!"
The whole point of having a trial is to determine whether a person is guilty or not. Locking people up without due process of law is unconstitutional and illegal. Period.
The act of siding with the enemy is a voluntary surrender of citizenship and the accompanying civil rights.
Can you point to the portion of the Constitution that spells that out? Or are you just making up rules that you wish were so?
My advice is, go back and watch it a second time. When I saw the movie in the theaters, I was also dissappointed, as I suspect every Guide fan was... that was because I was constantly comparing it to the books, and noting where it fell short. A few months later, I watched it again on DVD, and that time I was no longer constantly comparing it to the books, but rather watching it solely as a movie in its own right. The second time I found it quite funny and enjoyable.
Why do they need to go the pub if Ford already has brought beer?
Hm... maybe they really needed the peanuts also.
Oh, come on. If every Muslim in the world is guilty of that, then every Christian is guilty of "justifying, excusing, aiding, abetting, and harbouring" Timothy McVeigh and the abortion-clinic bombers. Just because 1.3 billion people seem like a big homogenous hive-mind to you doesn't mean they are, it only means you are ignorant of the real situation.
To be fair, that was true at the time.
What makes you think a bunch of space scientists are even capable of fixing the world we live in?
For that matter, why do you spend so much time working in (your career of choice), when you ought to be out fixing the world you live in?
They can, and do. Once the price comes down (or people become willing to pay $500,000 for a car), and the infrastructure issues get worked out, we will have them.