We were attempting a stunt with my old 486SX-20. Curious as to whether a computer can operate while submerged in antifreeze, we filled a large oil tub with the stuff (making sure to use the type of antifreeze which contains no water). We took the mobo out of the case, connected a power and video cable, turned it on, and slowly lowered it in.
The instant the corner of the mobo hit the antifreeze, the screen went black.
We decided we should try it again, but there was no good way of getting all the antifreeze off the motherboard. Eventually I decided that we'd just put it in the sink and use the kitchen hose to blast the antifreeze away. This was not even distilled water or anything, just what comes out of the tap.
After a thorough blasting, I took the mobo to the bathroom and blew it dry with a hair dryer, being careful to completely dry the nooks and crannies, including the slots in the PCI connectors, the D-connectors on the back, the power plug, etc. I used the "HOT" setting on the hair dryer.
After all this torture, we took the mobo back to the garage, and hooked it back up. We turned the power on, and the thing actually booted up. Somewhat amazed, we slowly lowered the mobo back into the antifreeze...
So I can personally attest that that particular motherboard was able to withstand two passes through this ridiculous procedure. As a side note, antifreeze does a particularly good job of cleaning the motherboard. The thing looked brand new when we were through with it (which was weird, since the components are all from '94 or sometime around then). It looks like a piece of alien machinery almost.
the cost isn't zero because the processors are making less heat from more electricty than your heater is.
This is wrong. Every last iota of energy that enters your computer eventually leaves it as heat. And it is exactly as much heat as would have been produced by an electric heater with the same power draw.
There is a negligible amount of energy stored in the charged capacitors inside the machine, and in the form of a large electric field inside the CRT, but when these devices are powered off, this energy too eventually turns to heat.
All machines whose purpose is to produce heat are equally efficient, from a thermodynamic standpoint. They are the only machines that have this property.
No, they're small console heaters embedded in the walls. They have dials on them to set the "heat," but no thermostats as such.
By "have no choice" I meant I can't have them install gas heat if I want. I can definitely choose to not run the heaters:-) In fact I'm not sure even how to turn them on since I've never used them nor do I plan to.
Correct. Hence the last sentence of my second paragraph: If you have electric heaters, the costs precisely balance out and you essentially can run your computer for "free."
However, my apartment has electric heaters which I have no choice over. They aren't as uncommon as you think, at least in the area of the world I live in.
Remember, thermodynamics tells us that all energy eventually ends up as heat. By using extra energy in your processor when it is under load, you are ultimately just generating more heat in your living space.
In the winter, you want to heat your apartment. Any heat coming from your computer is less energy you need to use in your heaters. It's 100% "efficient" -- remember that thermodynamic efficiency is measured in terms of how much energy is wasted as heat. But in this case (and what a special case it is!) the very goal is the production of heat! If you have electric heaters, the costs precisely balance out and you essentially can run your computer for "free."
My dual Athlon XP 2400 system (yeah, with the modification to run in dual) produces a ton of heat. Just a few months ago I was unable to run it at full load for more than a few minutes before the overheat alarm went off (I have no AC in my apartment). This doesn't just make my room noticeably warmer; it makes the whole apartment warm. We still haven't run our heaters a single time, and it keeps getting colder outside.
Wherever there is an area you must warm up, the effective cost of running a heat-producing piece of equipment is zero. Think about it.
As a developer you want to have all helper functions for a data type at a single place.
This is known as a "header file." Try opening one up in the editor sometime. You'll find that it contains a list of functions, along with their parameter types. These "header files" are usually fairly organized, with one file (or a set of closely related files) corresponding to a single API. Imagine that.
As the user of an IDE you want to see all available functions when you enter a variable with a given type.
You think it's a good idea to let the IDE drive the way you structure the code? "Yeah, I wrote the code this way because it causes the IDE to pop up a little yellow tooltip box when I type the name of this variable! Isn't that cute?!"
The problem is, you are assuming they are using the term "machine" as a metaphor, which they aren't. The power grid literally is a machine. We are greatly restricted by the laws of physics in how we can construct, interconnect, and run this machine. We can't declare by law that power must flow exclusively in certain ways.
A machine is a device that manipulates and applies force and/or power. This is a very specific definition. The electric grid qualifies, since it delivers power from one location to another in the form of electric fields and flowing charges. Transformers convert between one voltage and another -- this is equivalent to how a lever converts force (total energy expended remains the same, but the force/voltage is converted at the expense of distance/current).
The Internet simply doesn't fall under the definition of a machine. It doesn't convert force or apply power. It exists largely in a virtual world, within microprocessors and memory cells, not high current lines. Its physical layer is an electrical system, yes, but it is not of the same nature as the electric grid. Its parts are not as tightly coupled. And its purpose is not the conversion and application of energy.
It's even easier to understand that the economy is not a machine. It doesn't convert force or apply power (and I needn't even mention that we're talking about the specific physics definitions of "force" and "power" here).
And don't even try to convince me that human society is a machine used for converting force and applying power.
So your slippery slope argument fails. There is no slope here, since they are not using the term "machine" metaphorically, but quite literally.
Wow, that sounds pretty dumb. Every 18 months (or whatever) when you upgrade your machines, you now have to upgrade twice as many.
Why not just run a virtual machine like VMWare and do everything on one box?
Right now, I can boot into 95, 98, ME, 2K, or XP, as well as a few RedHat images, and a FreeBSD image. And I can have any one of them at the touch of a finger. Or all of them at once if I want.
I just keep a VMWare in full screen mode on one monitor. It literally is like running Windows and Linux on the same box. You can drag the mouse seamlessly from one to the other. You can cut and paste between them.
[work on code in Emacs -- tappity tappity tap] -- [mouse over to Windows] -- [click 'run'] -- [step through debugger] -- [mouse back over to Linux] -- [tappity tap tap]...
Hey, a customer says it's crashing on ME? [Bam, resume ME from hibernate within 5 seconds, run it, yep, it crashes] -- [mouse over to Linux] -- [tappity tap tap tap]
The question is, do you believe people should be treated as innocent until proven guilty, or do you not?
Since P2P use is not a crime (whereas copyright violation is), it seems you should have to prove that a user was trading a copyrighted file before the P2P use becomes unacceptable.
Drugs like oxycontin and vicodin are massively abused, but we don't treat anyone who happens to hold a bottle of vicodin as a criminal. Lock picking equipment can certainly be used for all kinds of evil, but we don't ban that either. Drunk people cause many deaths every year yet we do not ban alcohol. I can walk into a BSDM shop and legally buy equipment to tie up another individual and torture them. None of these things are illegal, because in this country we have this concept of innocent until proven guilty. Just because a tool can be used to commit a crime does not mean that those who have that tool are criminals. This surely is idealism, yes, and it is very important idealism.
Perhaps a "realist" such as yourself would be happier in a more "realistic" country, such as Iran, where they don't have these pesky ideals.
I think you are avoiding the question because you actually don't believe in the idea that you should have to prove someone's guilt before treating them as guilty. All you have to do is answer the question...
I don't use P2P. Or is it totally inconceivable to you that somebody who doesn't use it might have an opinion on it?
You know its illegal, but you keep on rationalizing that it is still ok
I don't rationalize anything. I don't share music. I don't believe it is "okay" to violate copyright. I never said any of those things.
Is it so fucking stark inconceivable to you that a person might use a P2P client for a legitimate purpose?
Suppose a bunch of people decided to use a P2P client to share photographs they've been taking around campus. According to you, these people are criminals merely for using P2P software.
As I said, shut up with your bullshit about how P2P is only used for illegal purposes. That isn't even the point here. The point is that people must be considered innocent until proven guilty, not vice-versa. If you can show a person is trading in illegal files, then do what you will. But a university whose policy is to unilaterally shut down internet access to anyone who uses a P2P client is not respecting this cherished right.
Of course, a university can set any policy they wish on their own network. It doesn't change the fact that it is immoral and tyrannical, and it most certainly is a reason to avoid the university.
Yes, I think you should avoid going to a school with an innocent-until-proven-guilty policy of banning network access to anyone who happens to run a P2P client, be it for copyright violation or not.
Or are you saying there is no legitimate use for P2P?
Don't give me the bullshit about how most users are pirates anyway. I already know that. What I want you to answer is: do you believe in the concept of "innocent until proven guilty," or do you not?
Automatically shutting off someone's net access because a certain port is open is tyrannical.
Phone books are invariably in sorted order and a simple binary search scales well past anything we'll ever need to worry about.
I pictured you performing binary search on a physical phone book to find an entry. Clearly you can't do this based on phone number since the book is sorted in alphabetical order.
You can't sort and index by multiple keys at the same time in a physical phone book. Had you said phone directory I would have understood WTF you were talking about.
I thought you were making the argument that quantum search isn't necessary because we can already do a fast search on a physical phone book based on any key we want, which isn't true.
It might mean: "Our physical simulations indicate that if we somehow figured out to produce such-and-such a crystallize structure, we would be able to get an efficiency of such-and-such. The only problem is to figure out how to manufacture it."
Theory can predict all kinds of cool stuff, it's actually making the damn thing that is difficult.
I already had this idea. Nobody here seems to take it seriously.
If McBride knew that SCO was irretrievably screwed, maybe he conceived this crazy stunt as a way to go out with a bang while simultaneously boosting Linux.
Apparently this notion is too insane for most Slashdot readers...
Yay, another person who majored in CS because he/she thought it was gonna make them a ton of money.
Hello, welcome to reality. You slaved for four years on material, probably constantly thinking "WTF does this have to do with programming?" You probably hated a lot of it. Now you're finding out that it makes no difference, because CS isn't about software engineering anyway. CS is an academic pursuit.
Your problem is you spent your four college years thinking how to maximize your "specific" skillset. You wanted to acquire particular computer-related skills. Was that really wise? We all know how fucking fast the world changes w.r.t. computers. Why on earth would you assume that Java, Perl, SQL, etc. are going to be worth anything at all in the future?
Instead you might have focused on your general ability to learn, to solve computation-based problems, and to pick up new languages and techniques. You would have expanded your horizon a little bit by learning about the hardware side of things -- some basic physics and circuit theory. These things will make you a valuable person, because you have the general ability to solve problems, not bang out LOCs in some specific language that will be obsolete soon.
I just keep hearing this horror story from every CS grad I talk to lately (I graduated a year ago). It makes me sad to hear people say that they feel their 4 years were wasted, or they studied something they didn't care about. How can Computer Science still be so misunderstood?
I wish you luck in this completely shitty economy.
I've written so many Java programs, web pages, etc in that small program, it's not even funny.
Actually, it's hilarious, since you just admitted that no program you've written has ever been more than 32K in size, since that's all Notepad can handle...:-P
I've seen this site before, and I remember thinking the first time I saw it that it's a very unsatisfying experiment. It depends on having a sticker on the back of the microwave which tells you the frequency, for crying out loud! The whole experiment is pretty pointless since you already know this information, plus, the relationship between the speed of light, frequency, and wavelength is a given.
Why not measure the speed of light without all these givens? Do it the way it was originally done: with a bunch of mirrors, or by looking at the moons of Jupiter, or any other way. These experiments which use high-tech devices with high-precision values already given, seem pretty damn pointless.
It's like saying "Hey, I can use this complex computational algebra system to prove the correctness of the quadratic formula!" Gee, good for you...
Slashdot Mirror
We were attempting a stunt with my old 486SX-20. Curious as to whether a computer can operate while submerged in antifreeze, we filled a large oil tub with the stuff (making sure to use the type of antifreeze which contains no water). We took the mobo out of the case, connected a power and video cable, turned it on, and slowly lowered it in.
The instant the corner of the mobo hit the antifreeze, the screen went black.
We decided we should try it again, but there was no good way of getting all the antifreeze off the motherboard. Eventually I decided that we'd just put it in the sink and use the kitchen hose to blast the antifreeze away. This was not even distilled water or anything, just what comes out of the tap.
After a thorough blasting, I took the mobo to the bathroom and blew it dry with a hair dryer, being careful to completely dry the nooks and crannies, including the slots in the PCI connectors, the D-connectors on the back, the power plug, etc. I used the "HOT" setting on the hair dryer.
After all this torture, we took the mobo back to the garage, and hooked it back up. We turned the power on, and the thing actually booted up. Somewhat amazed, we slowly lowered the mobo back into the antifreeze...
So I can personally attest that that particular motherboard was able to withstand two passes through this ridiculous procedure. As a side note, antifreeze does a particularly good job of cleaning the motherboard. The thing looked brand new when we were through with it (which was weird, since the components are all from '94 or sometime around then). It looks like a piece of alien machinery almost.
It still works, too.
This is wrong. Every last iota of energy that enters your computer eventually leaves it as heat. And it is exactly as much heat as would have been produced by an electric heater with the same power draw.
There is a negligible amount of energy stored in the charged capacitors inside the machine, and in the form of a large electric field inside the CRT, but when these devices are powered off, this energy too eventually turns to heat.
All machines whose purpose is to produce heat are equally efficient, from a thermodynamic standpoint. They are the only machines that have this property.
You can't ping Slashdot because Slashdot doesn't respond to pings. Notice that the web server seems to be working fine? :-P
By "have no choice" I meant I can't have them install gas heat if I want. I can definitely choose to not run the heaters :-) In fact I'm not sure even how to turn them on since I've never used them nor do I plan to.
However, my apartment has electric heaters which I have no choice over. They aren't as uncommon as you think, at least in the area of the world I live in.
In the winter, you want to heat your apartment. Any heat coming from your computer is less energy you need to use in your heaters. It's 100% "efficient" -- remember that thermodynamic efficiency is measured in terms of how much energy is wasted as heat. But in this case (and what a special case it is!) the very goal is the production of heat! If you have electric heaters, the costs precisely balance out and you essentially can run your computer for "free."
My dual Athlon XP 2400 system (yeah, with the modification to run in dual) produces a ton of heat. Just a few months ago I was unable to run it at full load for more than a few minutes before the overheat alarm went off (I have no AC in my apartment). This doesn't just make my room noticeably warmer; it makes the whole apartment warm. We still haven't run our heaters a single time, and it keeps getting colder outside.
Wherever there is an area you must warm up, the effective cost of running a heat-producing piece of equipment is zero. Think about it.
This is the most insightful thing I've read all day, thanks!
This is known as a "header file." Try opening one up in the editor sometime. You'll find that it contains a list of functions, along with their parameter types. These "header files" are usually fairly organized, with one file (or a set of closely related files) corresponding to a single API. Imagine that.
As the user of an IDE you want to see all available functions when you enter a variable with a given type.
You think it's a good idea to let the IDE drive the way you structure the code? "Yeah, I wrote the code this way because it causes the IDE to pop up a little yellow tooltip box when I type the name of this variable! Isn't that cute?!"
What's your problem, it's just a diff! Just run it through patch and you'll be ok.
A machine is a device that manipulates and applies force and/or power. This is a very specific definition. The electric grid qualifies, since it delivers power from one location to another in the form of electric fields and flowing charges. Transformers convert between one voltage and another -- this is equivalent to how a lever converts force (total energy expended remains the same, but the force/voltage is converted at the expense of distance/current).
The Internet simply doesn't fall under the definition of a machine. It doesn't convert force or apply power. It exists largely in a virtual world, within microprocessors and memory cells, not high current lines. Its physical layer is an electrical system, yes, but it is not of the same nature as the electric grid. Its parts are not as tightly coupled. And its purpose is not the conversion and application of energy.
It's even easier to understand that the economy is not a machine. It doesn't convert force or apply power (and I needn't even mention that we're talking about the specific physics definitions of "force" and "power" here).
And don't even try to convince me that human society is a machine used for converting force and applying power.
So your slippery slope argument fails. There is no slope here, since they are not using the term "machine" metaphorically, but quite literally.
Not even within an order of magnitude: (1-0.999868)/(1-0.99999) = 13.2 times more likely to have a failure with 99.9868% vs. 99.999% reliability.
It isn't "five nines."
Why not just run a virtual machine like VMWare and do everything on one box?
Right now, I can boot into 95, 98, ME, 2K, or XP, as well as a few RedHat images, and a FreeBSD image. And I can have any one of them at the touch of a finger. Or all of them at once if I want.
I just keep a VMWare in full screen mode on one monitor. It literally is like running Windows and Linux on the same box. You can drag the mouse seamlessly from one to the other. You can cut and paste between them.
[work on code in Emacs -- tappity tappity tap] -- [mouse over to Windows] -- [click 'run'] -- [step through debugger] -- [mouse back over to Linux] -- [tappity tap tap]...
Hey, a customer says it's crashing on ME? [Bam, resume ME from hibernate within 5 seconds, run it, yep, it crashes] -- [mouse over to Linux] -- [tappity tap tap tap]
Your way sounds really... dumb. And expensive.
Since P2P use is not a crime (whereas copyright violation is), it seems you should have to prove that a user was trading a copyrighted file before the P2P use becomes unacceptable.
Drugs like oxycontin and vicodin are massively abused, but we don't treat anyone who happens to hold a bottle of vicodin as a criminal. Lock picking equipment can certainly be used for all kinds of evil, but we don't ban that either. Drunk people cause many deaths every year yet we do not ban alcohol. I can walk into a BSDM shop and legally buy equipment to tie up another individual and torture them. None of these things are illegal, because in this country we have this concept of innocent until proven guilty. Just because a tool can be used to commit a crime does not mean that those who have that tool are criminals. This surely is idealism, yes, and it is very important idealism.
Perhaps a "realist" such as yourself would be happier in a more "realistic" country, such as Iran, where they don't have these pesky ideals.
I think you are avoiding the question because you actually don't believe in the idea that you should have to prove someone's guilt before treating them as guilty. All you have to do is answer the question...
But I'm tired of arguing with an AC.
I actually meant to type the reverse... I had hoped it was obvious, guess not.
I don't use P2P. Or is it totally inconceivable to you that somebody who doesn't use it might have an opinion on it?
You know its illegal, but you keep on rationalizing that it is still ok
I don't rationalize anything. I don't share music. I don't believe it is "okay" to violate copyright. I never said any of those things.
Is it so fucking stark inconceivable to you that a person might use a P2P client for a legitimate purpose? Suppose a bunch of people decided to use a P2P client to share photographs they've been taking around campus. According to you, these people are criminals merely for using P2P software.
As I said, shut up with your bullshit about how P2P is only used for illegal purposes. That isn't even the point here. The point is that people must be considered innocent until proven guilty, not vice-versa. If you can show a person is trading in illegal files, then do what you will. But a university whose policy is to unilaterally shut down internet access to anyone who uses a P2P client is not respecting this cherished right.
Of course, a university can set any policy they wish on their own network. It doesn't change the fact that it is immoral and tyrannical, and it most certainly is a reason to avoid the university.
Or are you saying there is no legitimate use for P2P?
Don't give me the bullshit about how most users are pirates anyway. I already know that. What I want you to answer is: do you believe in the concept of "innocent until proven guilty," or do you not?
Automatically shutting off someone's net access because a certain port is open is tyrannical.
Phone books are invariably in sorted order and a simple binary search scales well past anything we'll ever need to worry about.
I pictured you performing binary search on a physical phone book to find an entry. Clearly you can't do this based on phone number since the book is sorted in alphabetical order.
You can't sort and index by multiple keys at the same time in a physical phone book. Had you said phone directory I would have understood WTF you were talking about.
I thought you were making the argument that quantum search isn't necessary because we can already do a fast search on a physical phone book based on any key we want, which isn't true.
Now why don't you lay off the asshole throttle?
Theory can predict all kinds of cool stuff, it's actually making the damn thing that is difficult.
I think you're the kind of person who in 1993 would not have "laid money on" something as crazy as a 5 GHz processor, too.
"Gee, the rate of advancement of computer technology has only blown us away 188747121 times in the past. It surely won't do it again."
I'm laughing at you preemptively.
If McBride knew that SCO was irretrievably screwed, maybe he conceived this crazy stunt as a way to go out with a bang while simultaneously boosting Linux.
Apparently this notion is too insane for most Slashdot readers...
Hello, welcome to reality. You slaved for four years on material, probably constantly thinking "WTF does this have to do with programming?" You probably hated a lot of it. Now you're finding out that it makes no difference, because CS isn't about software engineering anyway. CS is an academic pursuit.
Your problem is you spent your four college years thinking how to maximize your "specific" skillset. You wanted to acquire particular computer-related skills. Was that really wise? We all know how fucking fast the world changes w.r.t. computers. Why on earth would you assume that Java, Perl, SQL, etc. are going to be worth anything at all in the future?
Instead you might have focused on your general ability to learn, to solve computation-based problems, and to pick up new languages and techniques. You would have expanded your horizon a little bit by learning about the hardware side of things -- some basic physics and circuit theory. These things will make you a valuable person, because you have the general ability to solve problems, not bang out LOCs in some specific language that will be obsolete soon.
I just keep hearing this horror story from every CS grad I talk to lately (I graduated a year ago). It makes me sad to hear people say that they feel their 4 years were wasted, or they studied something they didn't care about. How can Computer Science still be so misunderstood?
I wish you luck in this completely shitty economy.
Back in my day we spliced a microphone into the phone jack and whistled...
Actually, it's hilarious, since you just admitted that no program you've written has ever been more than 32K in size, since that's all Notepad can handle... :-P
Why not measure the speed of light without all these givens? Do it the way it was originally done: with a bunch of mirrors, or by looking at the moons of Jupiter, or any other way. These experiments which use high-tech devices with high-precision values already given, seem pretty damn pointless. It's like saying "Hey, I can use this complex computational algebra system to prove the correctness of the quadratic formula!" Gee, good for you...
Are you feeling the irony, baby?