If I listen to a piece of music a couple of times, I am generally capable of replaying it in my mind. Have I then violated someone's copyright?
For any reasonable definition of copyright, you have not. The purpose of copyright is to strike a useful balance while acknowledging the natural ability and desire of people to copy ideas, works, and music, and share them with others. That's what we call culture. At the same time, we wish to promote the authorship of such content and make it possible to make a living at it without undue hardship. So, copyright is a balance between the author recieving payment for the use of their work and people being actually able to use the work. Hence, such concepts as fair use and limited duration of copyright.
When I see intelligent people discussing copyright while completely missing this point, I am amazed at what a perversion our modern copyright system has become.
In short, listening to music is not a violation of any sane copyright system.
I find this result somewhat odd, because it is contrary to what I have benchamrked on my applications. My case was a heavy compute load, with a substantial number of objects created and then thrown away (basically one per very short inner loop iteration, iirc). Going to a very lightweight factory method is what allowed me to get to the point where the garbage collector did not run in my main loop at all (GC times
I suppose the end result is that, like a large number of optimizations, you have to try it and see. In my case it worked, though my case is probably very different from the business logic applications I'm going to assume he used.
speaking of MD5... I have a couple files that collide in the least significant 76 bits of the hash. It's from a test run of a program that is basically the same as the md5crk.com people's project, but intended to find smaller collisions. It's an entirely academic endeavor, for my computer security class.
Dude, is your C compiler that bad? I like Java a lot, and use it for compute intensive applications, but I think you're either pretty bad witha c compiler or trolling. if you're doing something CPU intensive in C, you need to use gcc -O2 (or -O3, depending), with -march=cputype. This will allow gcc to generate exactly the same code you just described, since it is not limited to 386 instructions. And if you need even more performance, you can just use Intel's C compiler for a lot of things (non-commercial is free as in beer), though it doesn't support some GNU extensions and I think has trouble with some things like the Linux kernel.
Not so slow as that... I have no idea about raytracing, but I've done several compute-intensive applications with Java. Both of them, the Java code has run at most 20-30% slower than the C. The two that I have worked with recently are a program that plays Amazons (a modern board game; fairly nifty) and recently one that does stuff with MD5 (searching for partial collisions). Now, I consider myself competent at Java optimization, but by no means an expert. So, a couple minor pieces of advice for anyone who wants to make their java programs a bit faster, and doesn't really know how:
1. Learn to use java -Xprof. This is a rudimentary profiler, but even the most basic data is useful. Concentrate on the parts that get the most use.
2. If -Xprof says the garbage collector is taking more than about 1-2% of the cpu, it's a problem. If it's at 10%, it's costing you well more than 10% speed -- lots of reasons, like cache misses, thread switching, and the allocations in the first place.
3. Don't delete objects in the main loop. Use factory methods and the like if you have to. This is how you decrease GC times.
4. Some of the standard API pieces are very slow. I've had particular trouble with anything related to the Collections framework, and Strings are even worse. Avoid these.
Now, all this takes work, but it's not particularly harder or easier than doing good optimization of C Code.
In fact, back when the internet was mostly Unix, this very thing happened with the Morris worm.
You mean, the same thing as blaster, nimda, code red, sql slammer, klez, netsky, melissa, and how many others? I think there's a difference in quantity, and that ignoring that is bordering on negligent.
It's those Xenon HID headlights I hate. You know, the ultra-bright, kinda bluish ones that blind you late at night as they come around the curve. Those seem bright enough to be unsafe.
Well, actually, it is scary. Would you think it acceptable if your power failed for a couple minutes every two weeks? How about if your car stalled on the highway? If your stove turned off while you were cooking dinner? There are a lot of things we use every day, that we expect to just work. For me, my computer is one of them. If the software crashes every two weeks, that's about 1.96 times too often for me.
You are aware that those numbers are usually LinPack numbers, not just theoretical specs, right? the Xbox probably does run at 80 GFlops in some theoretical sense, when you consider the graphics chip. But, that won't translate into LinPack numbers. The Apple supercomputer, on the other hand, actually has a LinPack score worth mentioning.
The general public isn't stupid. Maybe underinformed, or willfully ignorant, or uncaring... but the vast majority are smart enough to understand the ideas in that paper, especially if you take the math out (and the math isn't required for an intuitive understanding anyway). So, tell your friends who don't care as much about why this is a bad idea, in straightforward and probably non-math terms. Pass this paper along to those who would be interested. In general, spread the word in whatever manner is appropriate, and some change will happen. Probably small, but maybe not totally insignificant in the long run. At the very least, you can feel good about yourself for trying to actually do something rather than just bemoaning how the general public doesn't care.
See my other reply in this thread... If you combine it with other random screening, then for any given number of total searches (which is limited by resources), you are taking searches away from the random pool and putting them into the targetted pool, and that decreases security.
Seriously, read the paper. You by definition can't subject everyone marked "green" to the same level of random screening. Security is limited in the number of searches they can realistically perform; increasing that number will almost certainly improve things, but the point of the paper is that weighting who gets searched will tend to *decrease*, not increase, security for any given number of searches.
The point about making it difficult to find people who are willing to be terrorists and not on the lists is very real; however, I think the pool of potential terrorists is large enough that this simply makes it harder, not impossible. I think there is probably a legitimate debate to be had about whether the net result is an increase or decrease in security, and if that debate were actually occuring in public in a real way (ie, such that whether CAPPS II actually gets implemented is impacted), then I would be far, far more comfortable with the whole idea.
That's the basic point of the paper. They add some math, a more detailed explanation, and go into the fact that this was basically used in the WTC attacks. It's remarkably simple once explained, but very few people actually think of it when considering whether CAPPS II is a good thing.
This paper describes how such a system actually makes it more likely that a terrorist cell can carry out a successful attack, when compared with random screening. The basic idea is that it is not hard to determine whether or not you are on the watch list, and then the terrorists can use hijackers who aren't on the watch list. Anyway, I know slashdotters aren't known for reading links, but the paper is actually quite accessible and worth reading at least some of.
Copyright, along with patents, was intended to advance the sciences and useful arts. The way this was accomplished is to allow a period of protection, after which the work was available to the public. Therefore, patents require an explanation of the patented device. Copyrights in the past ensured that the work was available because text is human readable. However, modern copyright is being applied to software in binary form, which is not human readable. After the software enters the public domain, new programmers may still be unable to use it for anything for lack of source code. Therefore, copyright on software should require the disclosure of the source code. The creators are still afforded protection for their work through copyright, but now society gets the benefit that the copyright bargain was supposed to provide -- the later use of the work.
Of course, for this to work, copyright terms need to be returned to something reasonable, but that's a different problem.
Now write "www.wheresgeorge.com" on the bills with a sharpie, and go spend them. People will re-enter at least some of them. Kinda neat, though it doesn't really serve a purpose. People tend to be more likely to enter bills with the web site written on them.
Re:Compressed Gas to blow the dust off?
on
Mars Rovers Update
·
· Score: 4, Informative
My understanding is that a variety of dust removal techniques were considered. The final decision was that an effective dust removal aparatus could be added, at the cost of any one of the instruments. The scientists decided limited life with more instruments was more useful, and so the dust removal system was left off.
It stays completely in the directories you give it, as far as I have seen. Which amounts to an install directory and subdirectories, plus a (potentially) separate download directory for downloading of splitfiles (essentially anything large, ie not the random web page type things).
There is an integrated bandwidth limiter. My understanding is that it is reasonably effective but somewhat fuzzy, ie it will occasionally go over the limit, but makes a "best effort" attempt to stay within while not hurting the network. You could always do OS-level hard limiting without any problems. I'm actually on a fairly fat pipe, so I just set the limit to something vaguely reasonable and left it at that, and haven't had any problems.
You can have a varied number of channels; everything from 1x to 16x is being talked about, and I think the spec discusses as much as 32x. The only real use for a 16x slot for most things is the graphics card. Therefore, you won't be seeing multiple 16x slots on normal motherboards. So you'll still only be able to put a single high end card on a normal motherboard. Now, it's possible that server boards will be different and have multiple wide slots; however, my guess is that those will stop at 4x or maybe 8x slots.
I suppose it might be the case that you can plug a card intended for a 16x slot into a smaller one, but that somehow seems like wishful thinking. If you could, it would sacrifice a little theoretical performance, but I don't think that matters.
Slashdot Mirror
For any reasonable definition of copyright, you have not. The purpose of copyright is to strike a useful balance while acknowledging the natural ability and desire of people to copy ideas, works, and music, and share them with others. That's what we call culture. At the same time, we wish to promote the authorship of such content and make it possible to make a living at it without undue hardship. So, copyright is a balance between the author recieving payment for the use of their work and people being actually able to use the work. Hence, such concepts as fair use and limited duration of copyright.
When I see intelligent people discussing copyright while completely missing this point, I am amazed at what a perversion our modern copyright system has become.
In short, listening to music is not a violation of any sane copyright system.
I find this result somewhat odd, because it is contrary to what I have benchamrked on my applications. My case was a heavy compute load, with a substantial number of objects created and then thrown away (basically one per very short inner loop iteration, iirc). Going to a very lightweight factory method is what allowed me to get to the point where the garbage collector did not run in my main loop at all (GC times I suppose the end result is that, like a large number of optimizations, you have to try it and see. In my case it worked, though my case is probably very different from the business logic applications I'm going to assume he used.
a
b
Dude, is your C compiler that bad? I like Java a lot, and use it for compute intensive applications, but I think you're either pretty bad witha c compiler or trolling. if you're doing something CPU intensive in C, you need to use gcc -O2 (or -O3, depending), with -march=cputype. This will allow gcc to generate exactly the same code you just described, since it is not limited to 386 instructions. And if you need even more performance, you can just use Intel's C compiler for a lot of things (non-commercial is free as in beer), though it doesn't support some GNU extensions and I think has trouble with some things like the Linux kernel.
1. Learn to use java -Xprof. This is a rudimentary profiler, but even the most basic data is useful. Concentrate on the parts that get the most use.
2. If -Xprof says the garbage collector is taking more than about 1-2% of the cpu, it's a problem. If it's at 10%, it's costing you well more than 10% speed -- lots of reasons, like cache misses, thread switching, and the allocations in the first place.
3. Don't delete objects in the main loop. Use factory methods and the like if you have to. This is how you decrease GC times.
4. Some of the standard API pieces are very slow. I've had particular trouble with anything related to the Collections framework, and Strings are even worse. Avoid these.
Now, all this takes work, but it's not particularly harder or easier than doing good optimization of C Code.
You mean, the same thing as blaster, nimda, code red, sql slammer, klez, netsky, melissa, and how many others? I think there's a difference in quantity, and that ignoring that is bordering on negligent.
Well, once Kazaa gets taken down, the spam campaign of "D0wnl*pd Fre*E mUz1c he`re" will start.
Is there a way to make eveything default to Firefox instead of Konq? or at least the random web links I can click various places?
It's those Xenon HID headlights I hate. You know, the ultra-bright, kinda bluish ones that blind you late at night as they come around the curve. Those seem bright enough to be unsafe.
Well, if you have 1-2 urinals per capita, it seems unlikely that they'd get used 68.5 times per day each. Just a thought.
Well, actually, it is scary. Would you think it acceptable if your power failed for a couple minutes every two weeks? How about if your car stalled on the highway? If your stove turned off while you were cooking dinner? There are a lot of things we use every day, that we expect to just work. For me, my computer is one of them. If the software crashes every two weeks, that's about 1.96 times too often for me.
Hmm, my music collection is already in (mostly) sub-10MB chunks. A few scripts and a few accounts, and it sounds really useful as a backup device.
Sure. If you'll excuse me a minute, I need to go check the couch cushions.
You are aware that those numbers are usually LinPack numbers, not just theoretical specs, right? the Xbox probably does run at 80 GFlops in some theoretical sense, when you consider the graphics chip. But, that won't translate into LinPack numbers. The Apple supercomputer, on the other hand, actually has a LinPack score worth mentioning.
The general public isn't stupid. Maybe underinformed, or willfully ignorant, or uncaring... but the vast majority are smart enough to understand the ideas in that paper, especially if you take the math out (and the math isn't required for an intuitive understanding anyway). So, tell your friends who don't care as much about why this is a bad idea, in straightforward and probably non-math terms. Pass this paper along to those who would be interested. In general, spread the word in whatever manner is appropriate, and some change will happen. Probably small, but maybe not totally insignificant in the long run. At the very least, you can feel good about yourself for trying to actually do something rather than just bemoaning how the general public doesn't care.
See my other reply in this thread... If you combine it with other random screening, then for any given number of total searches (which is limited by resources), you are taking searches away from the random pool and putting them into the targetted pool, and that decreases security.
The point about making it difficult to find people who are willing to be terrorists and not on the lists is very real; however, I think the pool of potential terrorists is large enough that this simply makes it harder, not impossible. I think there is probably a legitimate debate to be had about whether the net result is an increase or decrease in security, and if that debate were actually occuring in public in a real way (ie, such that whether CAPPS II actually gets implemented is impacted), then I would be far, far more comfortable with the whole idea.
That's the basic point of the paper. They add some math, a more detailed explanation, and go into the fact that this was basically used in the WTC attacks. It's remarkably simple once explained, but very few people actually think of it when considering whether CAPPS II is a good thing.
This paper describes how such a system actually makes it more likely that a terrorist cell can carry out a successful attack, when compared with random screening. The basic idea is that it is not hard to determine whether or not you are on the watch list, and then the terrorists can use hijackers who aren't on the watch list. Anyway, I know slashdotters aren't known for reading links, but the paper is actually quite accessible and worth reading at least some of.
Of course, for this to work, copyright terms need to be returned to something reasonable, but that's a different problem.
Now write "www.wheresgeorge.com" on the bills with a sharpie, and go spend them. People will re-enter at least some of them. Kinda neat, though it doesn't really serve a purpose. People tend to be more likely to enter bills with the web site written on them.
My understanding is that a variety of dust removal techniques were considered. The final decision was that an effective dust removal aparatus could be added, at the cost of any one of the instruments. The scientists decided limited life with more instruments was more useful, and so the dust removal system was left off.
Whoa. Tell me how you plan to do that! I have a test coming up I could use some advance info on...
There is an integrated bandwidth limiter. My understanding is that it is reasonably effective but somewhat fuzzy, ie it will occasionally go over the limit, but makes a "best effort" attempt to stay within while not hurting the network. You could always do OS-level hard limiting without any problems. I'm actually on a fairly fat pipe, so I just set the limit to something vaguely reasonable and left it at that, and haven't had any problems.
You can have a varied number of channels; everything from 1x to 16x is being talked about, and I think the spec discusses as much as 32x. The only real use for a 16x slot for most things is the graphics card. Therefore, you won't be seeing multiple 16x slots on normal motherboards. So you'll still only be able to put a single high end card on a normal motherboard. Now, it's possible that server boards will be different and have multiple wide slots; however, my guess is that those will stop at 4x or maybe 8x slots.
I suppose it might be the case that you can plug a card intended for a 16x slot into a smaller one, but that somehow seems like wishful thinking. If you could, it would sacrifice a little theoretical performance, but I don't think that matters.