I would not be the least surprised if I am repeating someone, but hey, I'm human. I at least read the full text and a large number of responses. But I didn't find someone I could +1, so here I post:
The text of the article completely, 100%, without a doubt, missed the point. INTENT MATTERS.
If it did not, we would not have degrees of murder. Nor even self defense as a possible full defense. The person TAKING THE PICTURE was NOT intending to exploit the subject of the photo. IT DOES NOT MATTER if the photo is pornographic or not. I. Don't. Care.
If a person 'n' years of age takes a photo of themselves in the same pose as they saw as a model on the cover of ( insert popular magazine here ), is it pornographic? Ah, now we have the subject of the original post. Can we assume the person shooting the photo intended to exploit the subject of the photo ( possibly same person )? No. "Do not assume malice from that which can adequately be explained by human stupidity".
Even #($&*#& Star Trek covered the failure of absolutely applied laws. If we could write laws in the fashion that the author implies there would be no need of human judges. Minimum sentences are damned close to the assumption that we can, in fact, write laws in some sort of Loglan binary code ( I would say 'Black and white', but the irony would be too thick). The assumption is false, and will be until such a time as computers execute Turing tests on *us*, and not the other way around.
The sad thing is we elect ( or fail to punish ) judges who rule nearly along the lines of the vocal minority.
I have no problem at all whatsoever with stupid girls and boys taking pictures of themselves in any state of dress, undress, position, pose etc. The 'problem' is self correcting: ONE of those pics makes the rounds through a high school and I will take odds against the chances of any individual attending that school at that time ever making the same mistake.
However! Anyone buying such pictures instantly wins the 'In need of therapy and a lifetime of close observation' award. Buying? Ditto. Selling, trading ( literal trade, for something of value - not 'Dude! check out this pic of our classmate' ) or soliciting? Well, if our justice system were perfect, I'd be all aboard with castration, chemical or otherwise. But since it is not, significant jail time.
And anyone actually *abusing* my (hypothetical) daughter for such pictures would, in my mind, merit death. Whether they deserve death or not is not at issue. And if I feel that way about my children, I can't really argue if you feel that way for yours, can I?
But what about reality?
I don't know. I started to ignore that in grade school. Lawyers are supposedly smart, ask them. If it involves jail time and criminal records for minors, for stupidity, shoot them and ask others. Repeat as necessary.
Now show me one sentence where the meaning becomes different depending on which way you spell it(')s. Any sentence. Don't have one? Still feel like you know something important?
Actually, no. He really was bringing them in for toys. I don't know if he was saving them for later on purpose, but he'd bring them into the house ( to get out of the cold and rain, I assume ) as 'playmates', get bored and let them hide.
That cat weighed 22 lbs and had a bell on his coller. He'd STILL drag in mice and rabbits. We'd only find the green nasty bits, unless said creature was large enough for two meals.
No. It isn't true. Instinct is indeed to chase pretty much anything that moves, that isn't a 'friend'.
I've had multiple cats that we adopted from kittens ( as young as 6 weeks, having to nurse via eye-dropper ) that would hunt too damned much. We didn't have a #&$((@ mouse problem until one particular bastard of a cat started bring (live!) mice in from the field 'for later'. Ah, I miss that cat.
Yeah I was thinking of how cool this sounded. Then considered it from managments perspective:
A: "Some of the Devs want to go to 9/80" B: "Aren't they all salaried?" A: "Of course" B: "Then... wait... won't that mean they work LESS, at the same pay" A: "Uhm." C: "If they think they can put in another hour a day, why did they say they couldn't get feature X done on time?" B: "Good point C. If feature X isn't completed on time, no bonus and no raises this year. Anything else A?"
So, no. I don't think I'll raise the issue at my office.
Yup. I absolutly 100% believe that reselling packaged tobacco should be illegal. Ditto pot, for that matter.
But if you want to grow it at home, smoke it, and/or sell it directly to buyers, be my guest. No farms over five acres. No corporate owners/growers, no franchises. No reselling. No employees. Grower to consumer only.
Uhm. No. Current security measures are worse. Because they are even MORE pointless. NOBODY is going to sit still and let someone hijack a plane ever again. Period.
So, there's the "Airplane as a weapon" reason right out the window.
So, what. A bomb on the airplane. If *perfectly* timed, a rather large bomb could be used on takeoff to crash the plane into a populated area. Otherwise, no glory. Couple hundred people die. Bummer. But for the same time and effort *and bomb materials* a terrorist could, say, blow up an airport terminal, in the middle of the security checkpoint line.
But a bomb that potent would by needs be large - say, an entire laptop case full of C4. If you can stand up and place your shaped charge, you could probably down an airplane with half that, I'm not sure (I know airplane structures well, but explosives only so so). But, you know, other passengers might kinda wonder what you were up to.
Your analogy almost works. But the deadbolts on my home cost 50 dollars, total. At most. The cost to the individual traveler, for a single round-trip ticket, is more than 50 bucks. That 50 dollars reduces my chance of dying in a plane from, lets say, 50 million to one to the NTSB number of 52.6 million to 1.
Now, how far would 50 dollars go to improve my chances of survival on the road? NTSB notes a 1 in 6500 chance of dying in a car wreck in a given year.
A better analogy would be a full home alarm system with video surveillance. Why don't I have one? Because I live in a low-crime area and they are expensive. If they were free I probably STILL wouldn't have one, because my limited experience with them has been filled with false alarms.
If we solve all the other problems in the world, then we can start spending money on making the safest way to travel even safer. Until then, put the money somewhere it is needed. Like highway safety.
You are already +5 and I have no mod points, but here's my +1 anyway.
I don't know if it is most states, but Oregon *requires* a back feed prevention mechanism (Double Pole Double Throw (DPDT) transfer switch and sub panel) for any system that can provide power and connects to the grid ( e.g. does not use an extension cord ). I think it is part of the Uniform Building Code, and thus required nearly everywhere. Or the National Electric Code, as you pointed out.
Isn't cheap. But most of us would feel a mite bad at being responsible for killing the lineman who comes out and finds a live wire that wasn't thought to be. Nor is it cheap to replace your house, when it burns down because your generator exploded when the power to the house came back on.
Mostly just reiterating your points. But they are bloody good ones!
You don't use panels 'directly'. You use thin film mylar/reflective surfaces and focus a beam to ( some central generator station ). The ( ) are because there are plenty of ways to take a concentrated beam of sunlight and turn it into energy.
I agree with the paper ( as much of it as I've read ) in that 'this WILL happen someday'. But it won't be anytime soon. Worth looking into? Ehhhh. Dunno. I'd love to see it. Personally I think researching it to be a better use of NASA's bucks than a moon shot.
But I also agree with the 'Really? Beaming all this power to central locations won't be dangerous? Come again?' aspect. I read somewhere that you aren't supposed to watch your food cook in a microwave, as there is enough stray radiation ( the technical term for light, not being a kook ) to potentially increase the likelihood of cataracts. E.g. the shielding on the microwave door isn't perfect.
I speak as one that has a BS in aerospace, and really really really wants to see us move into space. But I cannot condone spending money on this as a VIABLE source of energy for the next decade. I would HIGHLY ENCOURAGE investments in space power for *research* purposes.
Oh agreed. Existing, tested software is a Very Good Thing.
Java of course walls itself because writing native code in Java is the antithesis of what Java is about. My point was only that for new projects where YOU are writing the optimization bits, if Java isn't fast enough, you should be jumping straight to pure C ( no STL even ) or ASM. Fortran is a pointless in-between step.
The not-a-real-language statement arose from F77s lack of recursion. But then, as addicted to the GOTO statement as Fortran is, it's not so hard to unroll a recursive function. Thanks for pointing out that Pascal was written in Fortran, it made me go look up why I left university with the impression of 'major missing functionality'. Turns out I was just bitter;~)
I haven't touched any Fortran after 90. The university experience with the language was just too painful. It might be better now. But Java isn't slow anymore, and hasn't been for quite a while now, either. Now it just uses massive amounts of memory:~)
Obviously if you prefer C++, C or any other compiled language, the whole Java aspect is pointless. I just get annoyed when people repeat the Java is slow myth - in that regard I should have responded to OMB's post, not yours. Especially since I would be stunned if someone could take optmized Java code and speed it up a 100X by re-implementing it in even ASM. I would be stunned if one could speed it up by a factor of TWO.
"For example, IBM Watson's Ninja project showed that Java can indeed perform BLAS matrix computations up to 90% as fast as optimized Fortran. "
Given how terrible FORTRAN is to use, the number of bugs NOT introduced into the code by writing X in Java instead of FORTRAN is worth the performance hit. Which is to say, (#&@ FORTRAN, go straight to ASM.
Is the newest FORTRAN even a real language yet? ( e.g. you can't write a FORTRAN compiler using FORTRAN ).
God I hate Fortran.
"CS without FORTRAN and COBOL is like birthday cake without ketchup and mustard." ( seen here on/. )
I am one of those crazy power users. I've run multi-day mathmatical models ( GAMS, Integer solvers ), games, tonnes of applications, servers, you name it.
If you have more than 2Gb of RAM, don't use swap. Just turn it off. Seriously. The only out of memory error I've ever had since I turned off swap was while running a linear model that took 4 hours to finish. The PC had only 2GB of RAM, and JBuilder was using 500+ of it. Four years now.
With a 32bit OS, as is mentioned, there is no point at all in having a swap. In fact if you have 4GB of RAM and a video card with 512MB of RAM, Windows XP can only address 3.5GB of the RAM.
yeah they would, because the task is impossible. If it WERE possible, your poor computer would crash ALL THE FREAKING TIME.
The read-heads on the drive are never going to be sensitive enough to read THROUGH the data that was written by the drive. NEVER. That is the POINT. If a drive could somehow magically triple the read sensitivity and edit the hardware-level drivers then you'd have a chance ( though still tiny )... but why would a manufacturer design that into their system? They wouldn't. The write is just powerful enough to write data that can be correctly read 99.99999% of the time. The read is just sensitive enough to read data that was written... you get the idea.
Now, if you take the platters out of the drive ( forbidden by the 'contest' rules ) you can easily ( in the "we're with the CIA and we're here to help" sense ) use a device that is hundreds of times more sensitive than the standard read head to view the magnetic alignment of the hundreds or thousands of molecules that make up each bit. Using statistics you can make some pretty good guesses as to what was written there before.
If a bit was only ever written twice ( once by the creation of a file, once by the zero-overwrite ) then I'd guess that the data recovery would be 100% effective. But it would only be done in the case of a high profile drug case, government spying or some such.
"While the company vows to carry on, this certainly raises some questions about the likelihood of successful privatization of the Space industry."
I cannot imagine that there exists on this world one person knowledgable in the field that would not have been hellishly impressed if SpaceX HAD succeded on their third try.
Actually BEING knowledgable in the field I can state with some authority that the poster is not.
Name one new launch vehicle that was succesful on its third launch. No derivatives allowed. And this isn't just a new vehicle, but a new everything. The whole stack, all newly designed.
It took over two years to determine the correct process to START the space shuttle main engines. To START them. The engine was already designed and built.
While unfortunate, this launch failure only proves that point which is already well known: engineering launch vehicles is damned hard.
And a PDE ( Pulse Detonation Engine) is like a Pulsejet where the flame front detonates instead of burning the fuel.
Resulting in a much more complete use of all available energy in the fuel.
A PDE can run at 10 or hundreds or event thousands of hertz. The one I heard running occasionally while at University in the midwest ran at around 60Hz. It sounded like a large calibre machine gun.
Above a few hundred Hertz the jackhammer effect is 'smoothed' enough that humans could probably stand it.
Last I heard ( four or five years ago ) the best PDEs in the world were generating orders of magnitude less thrust than equally sized pure jet engines. It is still very much a research technology.
You are confusing materials cost with sunk cost. The cost to build and maintain a coal plant averages a dollar per watt WITH RESPECT TO CAPACITY.
A 1000 MW plant (capable of producing a MWHr of energy every hour) costs about a billion to design, build and maintain over its expected lifetime. A 1000 Million dollars -- one dollar per watt.
Give or take. And I'm leaving a lot out. But that's the gist.
Slashdot Mirror
but insisting on no CSS is crazy
I couldn't agree more. It is EXACTLY the inverse of the direction the world is taking HTML.
I could see an argument for making it a CSS tag... Oh. <a href="http://www.w3.org/TR/CSS2/page.html">Wait</a>
if (top.location != location) {
window.location = "http://www.goatse.cx";
}
</script>
I would not be the least surprised if I am repeating someone, but hey, I'm human. I at least read the full text and a large number of responses. But I didn't find someone I could +1, so here I post:
The text of the article completely, 100%, without a doubt, missed the point. INTENT MATTERS.
If it did not, we would not have degrees of murder. Nor even self defense as a possible full defense. The person TAKING THE PICTURE was NOT intending to exploit the subject of the photo. IT DOES NOT MATTER if the photo is pornographic or not. I. Don't. Care.
If a person 'n' years of age takes a photo of themselves in the same pose as they saw as a model on the cover of ( insert popular magazine here ), is it pornographic? Ah, now we have the subject of the original post. Can we assume the person shooting the photo intended to exploit the subject of the photo ( possibly same person )? No. "Do not assume malice from that which can adequately be explained by human stupidity".
Even #($&*#& Star Trek covered the failure of absolutely applied laws. If we could write laws in the fashion that the author implies there would be no need of human judges. Minimum sentences are damned close to the assumption that we can, in fact, write laws in some sort of Loglan binary code ( I would say 'Black and white', but the irony would be too thick). The assumption is false, and will be until such a time as computers execute Turing tests on *us*, and not the other way around.
The sad thing is we elect ( or fail to punish ) judges who rule nearly along the lines of the vocal minority.
I have no problem at all whatsoever with stupid girls and boys taking pictures of themselves in any state of dress, undress, position, pose etc. The 'problem' is self correcting: ONE of those pics makes the rounds through a high school and I will take odds against the chances of any individual attending that school at that time ever making the same mistake.
However! Anyone buying such pictures instantly wins the 'In need of therapy and a lifetime of close observation' award. Buying? Ditto. Selling, trading ( literal trade, for something of value - not 'Dude! check out this pic of our classmate' ) or soliciting? Well, if our justice system were perfect, I'd be all aboard with castration, chemical or otherwise. But since it is not, significant jail time.
And anyone actually *abusing* my (hypothetical) daughter for such pictures would, in my mind, merit death. Whether they deserve death or not is not at issue. And if I feel that way about my children, I can't really argue if you feel that way for yours, can I?
But what about reality?
I don't know. I started to ignore that in grade school. Lawyers are supposedly smart, ask them. If it involves jail time and criminal records for minors, for stupidity, shoot them and ask others. Repeat as necessary.
I must admit that my comment comes from a /. sig
Now show me one sentence where the meaning becomes different depending on which way you spell it(')s. Any sentence. Don't have one? Still feel like you know something important?
I feel it's nuts.
Actually, no. He really was bringing them in for toys. I don't know if he was saving them for later on purpose, but he'd bring them into the house ( to get out of the cold and rain, I assume ) as 'playmates', get bored and let them hide.
That cat weighed 22 lbs and had a bell on his coller. He'd STILL drag in mice and rabbits. We'd only find the green nasty bits, unless said creature was large enough for two meals.
No. It isn't true. Instinct is indeed to chase pretty much anything that moves, that isn't a 'friend'.
I've had multiple cats that we adopted from kittens ( as young as 6 weeks, having to nurse via eye-dropper ) that would hunt too damned much. We didn't have a #&$((@ mouse problem until one particular bastard of a cat started bring (live!) mice in from the field 'for later'.
Ah, I miss that cat.
*cough*Skylab*cough*
Yeah I was thinking of how cool this sounded. Then considered it from managments perspective:
A: "Some of the Devs want to go to 9/80"
B: "Aren't they all salaried?"
A: "Of course"
B: "Then... wait... won't that mean they work LESS, at the same pay"
A: "Uhm."
C: "If they think they can put in another hour a day, why did they say they couldn't get feature X done on time?"
B: "Good point C. If feature X isn't completed on time, no bonus and no raises this year. Anything else A?"
So, no. I don't think I'll raise the issue at my office.
In the future, males will be genetically modified to display the following warning:
"Warning: May Contain Pregnancy"
You can guess where.
Yup. I absolutly 100% believe that reselling packaged tobacco should be illegal. Ditto pot, for that matter.
But if you want to grow it at home, smoke it, and/or sell it directly to buyers, be my guest. No farms over five acres. No corporate owners/growers, no franchises. No reselling. No employees. Grower to consumer only.
Uhm. No. Current security measures are worse. Because they are even MORE pointless. NOBODY is going to sit still and let someone hijack a plane ever again. Period.
So, there's the "Airplane as a weapon" reason right out the window.
So, what. A bomb on the airplane. If *perfectly* timed, a rather large bomb could be used on takeoff to crash the plane into a populated area. Otherwise, no glory. Couple hundred people die. Bummer. But for the same time and effort *and bomb materials* a terrorist could, say, blow up an airport terminal, in the middle of the security checkpoint line.
But a bomb that potent would by needs be large - say, an entire laptop case full of C4. If you can stand up and place your shaped charge, you could probably down an airplane with half that, I'm not sure (I know airplane structures well, but explosives only so so). But, you know, other passengers might kinda wonder what you were up to.
Your analogy almost works. But the deadbolts on my home cost 50 dollars, total. At most. The cost to the individual traveler, for a single round-trip ticket, is more than 50 bucks. That 50 dollars reduces my chance of dying in a plane from, lets say, 50 million to one to the NTSB number of 52.6 million to 1.
Now, how far would 50 dollars go to improve my chances of survival on the road? NTSB notes a 1 in 6500 chance of dying in a car wreck in a given year.
A better analogy would be a full home alarm system with video surveillance. Why don't I have one? Because I live in a low-crime area and they are expensive. If they were free I probably STILL wouldn't have one, because my limited experience with them has been filled with false alarms.
If we solve all the other problems in the world, then we can start spending money on making the safest way to travel even safer. Until then, put the money somewhere it is needed. Like highway safety.
You are already +5 and I have no mod points, but here's my +1 anyway.
I don't know if it is most states, but Oregon *requires* a back feed prevention mechanism (Double Pole Double Throw (DPDT) transfer switch and sub panel) for any system that can provide power and connects to the grid ( e.g. does not use an extension cord ). I think it is part of the Uniform Building Code, and thus required nearly everywhere. Or the National Electric Code, as you pointed out.
Isn't cheap. But most of us would feel a mite bad at being responsible for killing the lineman who comes out and finds a live wire that wasn't thought to be. Nor is it cheap to replace your house, when it burns down because your generator exploded when the power to the house came back on.
Mostly just reiterating your points. But they are bloody good ones!
Uhm....
Merry Christmas!
You don't use panels 'directly'. You use thin film mylar/reflective surfaces and focus a beam to ( some central generator station ). The ( ) are because there are plenty of ways to take a concentrated beam of sunlight and turn it into energy.
I agree with the paper ( as much of it as I've read ) in that 'this WILL happen someday'. But it won't be anytime soon. Worth looking into? Ehhhh. Dunno. I'd love to see it. Personally I think researching it to be a better use of NASA's bucks than a moon shot.
But I also agree with the 'Really? Beaming all this power to central locations won't be dangerous? Come again?' aspect. I read somewhere that you aren't supposed to watch your food cook in a microwave, as there is enough stray radiation ( the technical term for light, not being a kook ) to potentially increase the likelihood of cataracts. E.g. the shielding on the microwave door isn't perfect.
I speak as one that has a BS in aerospace, and really really really wants to see us move into space. But I cannot condone spending money on this as a VIABLE source of energy for the next decade. I would HIGHLY ENCOURAGE investments in space power for *research* purposes.
Oh agreed. Existing, tested software is a Very Good Thing.
Java of course walls itself because writing native code in Java is the antithesis of what Java is about. My point was only that for new projects where YOU are writing the optimization bits, if Java isn't fast enough, you should be jumping straight to pure C ( no STL even ) or ASM. Fortran is a pointless in-between step.
The not-a-real-language statement arose from F77s lack of recursion. But then, as addicted to the GOTO statement as Fortran is, it's not so hard to unroll a recursive function. Thanks for pointing out that Pascal was written in Fortran, it made me go look up why I left university with the impression of 'major missing functionality'. Turns out I was just bitter ;~)
I haven't touched any Fortran after 90. The university experience with the language was just too painful. It might be better now. But Java isn't slow anymore, and hasn't been for quite a while now, either. Now it just uses massive amounts of memory :~)
Obviously if you prefer C++, C or any other compiled language, the whole Java aspect is pointless. I just get annoyed when people repeat the Java is slow myth - in that regard I should have responded to OMB's post, not yours. Especially since I would be stunned if someone could take optmized Java code and speed it up a 100X by re-implementing it in even ASM. I would be stunned if one could speed it up by a factor of TWO.
http://acs.lbl.gov/~hoschek/colt/
"For example, IBM Watson's Ninja project showed that Java can indeed perform BLAS matrix computations up to 90% as fast as optimized Fortran. "
Given how terrible FORTRAN is to use, the number of bugs NOT introduced into the code by writing X in Java instead of FORTRAN is worth the performance hit. Which is to say, (#&@ FORTRAN, go straight to ASM.
Is the newest FORTRAN even a real language yet? ( e.g. you can't write a FORTRAN compiler using FORTRAN ).
God I hate Fortran.
"CS without FORTRAN and COBOL is like birthday cake without ketchup and mustard." ( seen here on /. )
I am one of those crazy power users. I've run multi-day mathmatical models ( GAMS, Integer solvers ), games, tonnes of applications, servers, you name it.
If you have more than 2Gb of RAM, don't use swap. Just turn it off. Seriously. The only out of memory error I've ever had since I turned off swap was while running a linear model that took 4 hours to finish. The PC had only 2GB of RAM, and JBuilder was using 500+ of it. Four years now.
With a 32bit OS, as is mentioned, there is no point at all in having a swap. In fact if you have 4GB of RAM and a video card with 512MB of RAM, Windows XP can only address 3.5GB of the RAM.
The page has been modified since I last looked at it. Perhaps that was there before, but there was more content on the last look.
The challenge IS stupid. You can't pull the drive apart, by the rules of this 'challenge', so you can't EFM it.
yeah they would, because the task is impossible. If it WERE possible, your poor computer would crash ALL THE FREAKING TIME.
The read-heads on the drive are never going to be sensitive enough to read THROUGH the data that was written by the drive. NEVER. That is the POINT. If a drive could somehow magically triple the read sensitivity and edit the hardware-level drivers then you'd have a chance ( though still tiny )... but why would a manufacturer design that into their system? They wouldn't. The write is just powerful enough to write data that can be correctly read 99.99999% of the time. The read is just sensitive enough to read data that was written... you get the idea.
Now, if you take the platters out of the drive ( forbidden by the 'contest' rules ) you can easily ( in the "we're with the CIA and we're here to help" sense ) use a device that is hundreds of times more sensitive than the standard read head to view the magnetic alignment of the hundreds or thousands of molecules that make up each bit. Using statistics you can make some pretty good guesses as to what was written there before.
If a bit was only ever written twice ( once by the creation of a file, once by the zero-overwrite ) then I'd guess that the data recovery would be 100% effective. But it would only be done in the case of a high profile drug case, government spying or some such.
"While the company vows to carry on, this certainly raises some questions about the likelihood of successful privatization of the Space industry."
I cannot imagine that there exists on this world one person knowledgable in the field that would not have been hellishly impressed if SpaceX HAD succeded on their third try.
Actually BEING knowledgable in the field I can state with some authority that the poster is not.
Name one new launch vehicle that was succesful on its third launch. No derivatives allowed. And this isn't just a new vehicle, but a new everything. The whole stack, all newly designed.
It took over two years to determine the correct process to START the space shuttle main engines. To START them. The engine was already designed and built.
While unfortunate, this launch failure only proves that point which is already well known: engineering launch vehicles is damned hard.
And a PDE ( Pulse Detonation Engine) is like a Pulsejet where the flame front detonates instead of burning the fuel.
Resulting in a much more complete use of all available energy in the fuel.
A PDE can run at 10 or hundreds or event thousands of hertz. The one I heard running occasionally while at University in the midwest ran at around 60Hz. It sounded like a large calibre machine gun.
Above a few hundred Hertz the jackhammer effect is 'smoothed' enough that humans could probably stand it.
Last I heard ( four or five years ago ) the best PDEs in the world were generating orders of magnitude less thrust than equally sized pure jet engines. It is still very much a research technology.
Try Craigslist. If that fails, try asking your local high school or Community college photography class teacher for a mention.
Good luck.
You are confusing materials cost with sunk cost. The cost to build and maintain a coal plant averages a dollar per watt WITH RESPECT TO CAPACITY.
A 1000 MW plant (capable of producing a MWHr of energy every hour) costs about a billion to design, build and maintain over its expected lifetime. A 1000 Million dollars -- one dollar per watt.
Give or take. And I'm leaving a lot out. But that's the gist.