One thing that would be interesting would be an "elevator" to orbit. We recently discovered a material that was light enough, but strong enough to do such a thing. However the problem would be finding enough resources to build the damn thing.
I think you are referring to nanotubes, and I'm afraid there are much more serious problems than money. A nanotube is at most a few hundred microns long, and they're terribly slippery buggers. Using them to make a strong macroscopic material is an open problem. And until that is solved, we won't be having any elevators to space.
According to the web page, the processor requires a Pentium-class processor or better. I know that my PPC 604e is definitely much better than a Pentium. Since I'm running linux 2.2, does this mean I can play HOMM3?
BTW, an experienced assembly coder will always beat or at least equal the optimizing compiler, because if nothing else works he can always look at what the compiler produces and see if he can improve on that. Besides, optimizing compilers are good, but not that good, someone has to write them, and when was the last time that you wrote a program that can solve complex creative problems better than you can?
There's a wonderful counterexample to your arguments, which is FFTW (the fastest fourier transform in the west). This is a code that dynamically generates a highly optimized algorithm for doing FFTs. It is written in 100% C, and outperforms (not all the time, but often) hand coded highly optimized FFT routines.
It does this by actually running timing tests, and using the results to optimize the algorithm for your computer, for the size of array you want transformed. It is quite amazing. Here are the benchmarking results from a RS6000 of some sort, comparing (among others) FFTW with ESSL, the IBM hand optimized FFT library.
This is sort of a strange example, in between JIT compilers and normal compilers, and is a case where the compiler actually knows a heck of a lot more than the programmer does. Anyhow, figured I'd point out a cool code that is able to optimize itself.
The LGPL doesn't really do what he wants, since it allows the library to be used in commercial closed source applications (since it is only linked in). GPL is really what he would want in order to retain the freedom to require commercial companies to pay for the use of his library.
(out of interest, how does the fp performance of one node of a proper supercomputer compare to an equivelant x86 chip?)
I have an accquaintance with a beowulf cluster of PIII's or something like that. He says (running the same plane wave density functional software I use) he gets about half as much speed as he gets using an alpha with about the same clock speed. And that is running real world (at least for me) software, not just running some meaningless benchmark. Of course, your milage will vary. And for some reason he is having trouble with the myrinet that makes mpi as slow as to be worthless.:(
Re:The Right Interface for the Job
on
Interface Zen
·
· Score: 1
I agree that there is a place for specialized interfaces, but there is also a place for standardized interfaces. This was one of the greatest things when the mac first came out. It didn't matter what application you used, command-p printed, command-c cut, command-i made something italic. It was wonderfully easy to use any new program. It got harder as programs introduced lots of new features, and they started running out of key combinations. But having some degree of standardization makes it easier to reach a zen state.
For example, as was pointed out using '/' to perform a search is very easy, and works in vi, more and less. I am not a vi user, but because of the similarity between more (on non-linux systems, since linux has a braindead more) and vi allow me to get by pretty well in vi when a collegue is using it, although I still have no idea how to edit.
The problem, of course, is how to decide which set of commands is standard. I would love to have navigator use the emacs commands right now, but it doesn't, so I have to use some other set of commands. But vi users would probably have a different preference, and it seems that Netscape has decided to make it easy for mac or windows users by using commands which are standard on those platforms. Oh well. It looks like I've been rambling.
BTW, Linpack is not a great gauge of a Supercomputers performance. When there a lot of nodes it becomes message bound and does not reflect the true performance of the machine.
Actually, I would say the same thing for the opposite reason. It all depends on what kind of calculations you are interested in doing. I would say that linpack isn't such a good measure because it is not very communication intensive. I run plane wave electronic structure calculations, which are very communication intensive, and would take a small T3E over a large beowulf any day of the week. It all depends on your perspective.
Religiously up-tight mother searches daughter's bedroom to find the password on a Post-It and then discovers the birth control prescription ?
Perhaps it would be helpful to think just a moment. If the mother is searching through her daughter's room, she may as well just find the birth control pills, or the prescription itself. Writing the password down (if you keep it somewhere reasonable) is quite a logical thing to do. You can keep it inside your bottle of pills...
Writing passwords down (when it is not a security concern but a privacy concern) is a GOOD THING. It means you can choose passwords that are harder to remember, which are also harder to crack. If someone searches through your personal belongings to find the password, you don't have much privacy anyways.
In contrast, a password that gives access to an account that could be used to crack another (e.g. a UNIX account) IS a security concern, since someone who finds it could use it to break into other people's accounts. In this case writing the password down is not such a good idea.
I've seen many people commenting on how passwords aren't safe as long as people write down their passwords, or choose passwords that people who know them well will be able to guess. This is true, but you have to consider what is gained by the password.
If a password gives you access to a computer account, it makes sense not to write it down. But if it only gives access to your medical records, you may as well keep it with your medical records, or in your home. If someone is willing to break into my home to find out my medical history, I wouldn't consider that a serious security hole. After all, while they are in my home they can do numerous other things to violate my privacy.
I think a little perspective is in order here. Anyone who is physically snooping on you has already violated your privacy, and access to computer records is only a small part of the damage they can do. When it comes to computer security in situations like this, what is important is that bad guys aren't able to use just computers (and no actual snooping around) to find out people's medical records, and the suggestions I've seen here seem to indicate that a well designed password system can prevent such attacks.
how about self modifying worms, that change the subject line of further forwards to any of, oh say, 100 different things, making it pretty impossible to avoid opening the naughty email
Actually, this would be the least of your problems. The easiest and most effective tactic would be to actually forward an actual email that was on the user's machine, chosen at random. Of course, you would have to insert the worm, but that couldn't be too hard. This should be almost trivial to do, and would result in a message that had no distinguishing characteristics, and really looked like the sort of email you might receive from that person.
Here at Cal (aka UC Berkeley) we don't have SSN as our IDs. The IDs are given out sequentially, although I think they increase some significant digit every year. Similar to your situation, the ID # is necesary for just about everything.
I don't see what is wrong with metal detectors. Do you think children should be allowed to bring firearms to school? Or can you think of a better way to keep them from bringing guns to school? Perhaps in your neighborhood this is not an issue, but in many places it is.
Not that is should be an issue, but schools have to do the best they can in the society they are stuck in.
The day KDE 2.0 ships (and that seems to be very soon), talking about "KDE/Linux" systems will make much more sense for a significant proportion of Linux users who will spend almost all of their time using KDE tools.
GNOME is not the reason people say it should be GNU/Linux. It is all the other tools, without which Linux would be useless, such as gcc and make. Without these tools, there would be no KDE on linux. Actually, I suspect there would be no linux.
Having said that, I may as well point out that I call it Linux, not GNU/Linux. It's just easier to say.
I don't think this would be a good idea for humans (even if it doesn't end up causing cancer or strokes or anything). They say that it essentially makes the adult mice have juvenile brains.
Humans also are smarter when they are kids, if you measure intelligence in certain ways. Certainly before the age of six kids are way better at learning languages than adults are, and I suspect there are a lot of other things that kids learn more easily than adults. But I don't think that kids are actually smarter than adults. They just have different ways of learning.
All the measures of intelligence they used were simply measures of memory. How quickly the mice could learn and remember something. In humans, intelligence is more often associated with the ability to reason than with the ability to memorize. Except in elementary school and anatomy classes, in which case the ability to memorize is equated with intelligence.
Basically, I think this is a very interesting development, but to generalize from smart mice to smart people seems a bit silly.
You have no idea what the purpose of patents is. The purpose of the patent system is to ENCOURAGE the sharing of new ideas. Every patented idea is published to the public.
Have you ever heard of trade secrets? Trade secrets are the equivalent of proprietary formats. They don't allow your competitors to make use of your innovative new ideas. Patents are open. Anyone can use a patented method. Patents are the nineteenth century equivalent of open source.
Admittedly, the area of sofware patents is somewhat murky waters, but in general patents are a GOOD THING. Without patents there would be no incentive for inventors and innovators to publish their work, and a strong incentive not to publish (since they would have a monopoly, until someone could reverse engineer their product).
And, I wouldn't mind hearing about some hardware requirements, would it take gobs and gobs of RAM and SCSI drives and massive CPU power to do playback of a track while recording another track to sync to it?
The hardware requirements for multitrack recording can't be all that big. A year or so ago, Pro Tools released a new version of their multitrack software, and as a promotion gave away free copies of the older version, which runs under 68K emulation on Power Macs. So I got a copy, and have been very impressed (although I would also prefer a free software solution)
On my power mac with a 200 MHz powerpc 604, emulating the 68K and with the built in sound, I have had no problem mixing and playing four tracks of audio while recording in sync a fourth. I think I have also had it running hi/lo pass filters on some of the tracks while doing this. The cpu power needed isn't that great, since most of the time is spent reading from the hard disk. Hope this helps.
I wonder if you could somehow reverse the polarity on the magnetic bottle and use the wind to pull the craft in towards the sun.
Not possible. The reason it is efficient is that it doesn't expend any energy to get the propulsion (apart from dissipative losses in maintaining the magnetic field). The solar wind will only push the craft away from the sun. Too bad. You'll just have to slingshot around some planet if you want to get back.
Slashdot Mirror
I know! I know!
The momentum is the partial derivative of the lagrangian with respect to the time derivative of a coordinate!
Sorry, I'm feeling a bit silly today.
I think you are referring to nanotubes, and I'm afraid there are much more serious problems than money. A nanotube is at most a few hundred microns long, and they're terribly slippery buggers. Using them to make a strong macroscopic material is an open problem. And until that is solved, we won't be having any elevators to space.
...bummer.
There's a wonderful counterexample to your arguments, which is FFTW (the fastest fourier transform in the west). This is a code that dynamically generates a highly optimized algorithm for doing FFTs. It is written in 100% C, and outperforms (not all the time, but often) hand coded highly optimized FFT routines.
It does this by actually running timing tests, and using the results to optimize the algorithm for your computer, for the size of array you want transformed. It is quite amazing. Here are the benchmarking results from a RS6000 of some sort, comparing (among others) FFTW with ESSL, the IBM hand optimized FFT library.
This is sort of a strange example, in between JIT compilers and normal compilers, and is a case where the compiler actually knows a heck of a lot more than the programmer does. Anyhow, figured I'd point out a cool code that is able to optimize itself.
The LGPL doesn't really do what he wants, since it allows the library to be used in commercial closed source applications (since it is only linked in). GPL is really what he would want in order to retain the freedom to require commercial companies to pay for the use of his library.
I have an accquaintance with a beowulf cluster of PIII's or something like that. He says (running the same plane wave density functional software I use) he gets about half as much speed as he gets using an alpha with about the same clock speed. And that is running real world (at least for me) software, not just running some meaningless benchmark. Of course, your milage will vary. And for some reason he is having trouble with the myrinet that makes mpi as slow as to be worthless. :(
For example, as was pointed out using '/' to perform a search is very easy, and works in vi, more and less. I am not a vi user, but because of the similarity between more (on non-linux systems, since linux has a braindead more) and vi allow me to get by pretty well in vi when a collegue is using it, although I still have no idea how to edit.
The problem, of course, is how to decide which set of commands is standard. I would love to have navigator use the emacs commands right now, but it doesn't, so I have to use some other set of commands. But vi users would probably have a different preference, and it seems that Netscape has decided to make it easy for mac or windows users by using commands which are standard on those platforms. Oh well. It looks like I've been rambling.
The d and k keys connect with your middle fingers, which are longest, and most often connect first for that reason.
Actually, I would say the same thing for the opposite reason. It all depends on what kind of calculations you are interested in doing. I would say that linpack isn't such a good measure because it is not very communication intensive. I run plane wave electronic structure calculations, which are very communication intensive, and would take a small T3E over a large beowulf any day of the week. It all depends on your perspective.
Actually, most of the IBM SP systems run either POWER 2 or (if they are newer) POWER 3 processors, which are not used in personal computers.
Perhaps it would be helpful to think just a moment. If the mother is searching through her daughter's room, she may as well just find the birth control pills, or the prescription itself. Writing the password down (if you keep it somewhere reasonable) is quite a logical thing to do. You can keep it inside your bottle of pills...
Writing passwords down (when it is not a security concern but a privacy concern) is a GOOD THING. It means you can choose passwords that are harder to remember, which are also harder to crack. If someone searches through your personal belongings to find the password, you don't have much privacy anyways.
In contrast, a password that gives access to an account that could be used to crack another (e.g. a UNIX account) IS a security concern, since someone who finds it could use it to break into other people's accounts. In this case writing the password down is not such a good idea.
If a password gives you access to a computer account, it makes sense not to write it down. But if it only gives access to your medical records, you may as well keep it with your medical records, or in your home. If someone is willing to break into my home to find out my medical history, I wouldn't consider that a serious security hole. After all, while they are in my home they can do numerous other things to violate my privacy.
I think a little perspective is in order here. Anyone who is physically snooping on you has already violated your privacy, and access to computer records is only a small part of the damage they can do. When it comes to computer security in situations like this, what is important is that bad guys aren't able to use just computers (and no actual snooping around) to find out people's medical records, and the suggestions I've seen here seem to indicate that a well designed password system can prevent such attacks.
Actually, this would be the least of your problems. The easiest and most effective tactic would be to actually forward an actual email that was on the user's machine, chosen at random. Of course, you would have to insert the worm, but that couldn't be too hard. This should be almost trivial to do, and would result in a message that had no distinguishing characteristics, and really looked like the sort of email you might receive from that person.
Why, a lump of coal, of course! :)
Here at Cal (aka UC Berkeley) we don't have SSN as our IDs. The IDs are given out sequentially, although I think they increase some significant digit every year. Similar to your situation, the ID # is necesary for just about everything.
Not that is should be an issue, but schools have to do the best they can in the society they are stuck in.
GNOME is not the reason people say it should be GNU/Linux. It is all the other tools, without which Linux would be useless, such as gcc and make. Without these tools, there would be no KDE on linux. Actually, I suspect there would be no linux.
Having said that, I may as well point out that I call it Linux, not GNU/Linux. It's just easier to say.
It's nice to see lilypond is included. Lilypond is the main program I run on my home computer (which happens to be running linuxppc).
Humans also are smarter when they are kids, if you measure intelligence in certain ways. Certainly before the age of six kids are way better at learning languages than adults are, and I suspect there are a lot of other things that kids learn more easily than adults. But I don't think that kids are actually smarter than adults. They just have different ways of learning.
All the measures of intelligence they used were simply measures of memory. How quickly the mice could learn and remember something. In humans, intelligence is more often associated with the ability to reason than with the ability to memorize. Except in elementary school and anatomy classes, in which case the ability to memorize is equated with intelligence.
Basically, I think this is a very interesting development, but to generalize from smart mice to smart people seems a bit silly.
Yeah, I remember that story. It was good. :)
Have you ever heard of trade secrets? Trade secrets are the equivalent of proprietary formats. They don't allow your competitors to make use of your innovative new ideas. Patents are open. Anyone can use a patented method. Patents are the nineteenth century equivalent of open source.
Admittedly, the area of sofware patents is somewhat murky waters, but in general patents are a GOOD THING. Without patents there would be no incentive for inventors and innovators to publish their work, and a strong incentive not to publish (since they would have a monopoly, until someone could reverse engineer their product).
The hardware requirements for multitrack recording can't be all that big. A year or so ago, Pro Tools released a new version of their multitrack software, and as a promotion gave away free copies of the older version, which runs under 68K emulation on Power Macs. So I got a copy, and have been very impressed (although I would also prefer a free software solution)
On my power mac with a 200 MHz powerpc 604, emulating the 68K and with the built in sound, I have had no problem mixing and playing four tracks of audio while recording in sync a fourth. I think I have also had it running hi/lo pass filters on some of the tracks while doing this. The cpu power needed isn't that great, since most of the time is spent reading from the hard disk. Hope this helps.
http://lienhard.desy.de/mackag/homepages/jan/kooBa se/home.html
Unfortunately it is still a work in progress. For example, they still haven't implemented hard disk recording.
You can find kooBase here. Unfortunately it is still a work in progress. For example, they still haven't implemented hard disk recording.
Not possible. The reason it is efficient is that it doesn't expend any energy to get the propulsion (apart from dissipative losses in maintaining the magnetic field). The solar wind will only push the craft away from the sun. Too bad. You'll just have to slingshot around some planet if you want to get back.