Well, the original slimp3 didn't even have a "real" CPU, just a microcontroller and a dedicated MP3 decoder chip (actually an FPGA, I believe) that only supported decoding a single MP3 stream at a time. That makes crossfading basically impossible.
The new device is supposed to have a little bit more power, and supports wave file streaming, so at worst you could add a crossfading MP3 decoder on the server and stream wave data to it. I don't know if this is supported yet, but if not I am sure someone will implement it quickly.
One, you control it with a remote control in your living room. Two, if you have $500 (or more!) speakers you can use them with this. Three, it can plug into your home stereo/home theater amplifier and be integraded along side your CD/DVD/PS2/whatever. Four, it is transmitted as digital PCM over wireless ethernet rather than analog FM modulation (which is how many wireless speakers work), so there will be no sound degredation as the link fades until it is no longer able to maintain the bandwidth required.
If you just want continious random play, and don't have a bunch of other AV gear you want to hook it into, this device is probably not for you, and you should go with the wireless speaker route. It will be simpler and cheaper. If you want hi-fi sound to your existing home theater system with a minimum of wires and maximum flexibility, you should order this (or something like it) right now:)
I have a home theater system. I still think it is lame to turn on my TV to listen to music. TV navigation would be a nice feature to have, but it is much more important to have a good front panel display.
Even if my TV is on, I would like to control the music while playing video games on my TV.
I don't like listening to music on my computer, either. It is not too annoying when I am just using it to play background music while I am working on my computer, mostly because it is essentially free, bu even then I would prefer to have a music playing device.
You can switch the text mode to "big" where you only get the bottom line of the display, but the text is twice as big. That is the way I use my slimp3. I can read the small text from across the room, but not easily. It would be nice if the display were about 50% larger, but I suspect that would be considerably more expensive (and involve a custom VFD instead of a standard size)
Many doctors will give experemental treatments to patients who don't respond to standard treatments, even if the patient is not in, or doesn't qualify for a study.
However, the biggest barrier is that most (all?) insurance companies will not cover experemental treatments. For members of test groups, the costs are usually borne by the research grant for the study, but not so for tag-a-longs.
This is indeed a question I have thought about a lot. I don't really know the answer, somehow it seems a little unsavory (if not wrong), but as you say, who gets to decide what makes it happy.
My guess is that it would be fine, but I wouldn't be surprised if said AI decided that despite its natural predelections, it wanted to be a painter. It happens often enough with humans that we conciously override our instinctual behavior for one reason or another, I don't see why that shouldn't happen to machines as well.
My personal feeling is that much of this is false savings. The fact is, from a usability standpoint, web clients totally suck for a lot of applications. Even a well designed web client can substantially imact user productivity compared to a dedicated thin client application (thin in the sense of being "interface only" while allowing the server to do all the work and data storage).
Web clients are going to stick around for a while, but but their days are already numbered.
I don't know what life is (for a human or an AI), but I think I will know it when I see it. When someone develops a sentient program, we damn well better give it the rights to life, liberty, and the pursuit of happiness, to the best of our ability.
I find it strange and worrisome that you seem more comfortable damning an entire new form of sentient entities into potential slavery than with attempting to extend the same basic rights we believe humans deserve by virtue of existence.
these people are moving blindly into an area where they have no business being.
You are the people that hold humanity back, and slow us from recognizing our potential. Our domain is the sun and the stars and all of creation. We will one day make computers that are our mental equals, and it is a mark of sanity that we begin to come to grips with that fact before it happens, so that we may be ready for that day when it comes.
Unfortunately, that isn't always true. A lot of mac software doesn't take advantage of extra buttons simply because they are not present in the default configurations. Also, the Windows (and other) "left button performs the default action, right button gives you a list of actions" is a really powerful framework for allowing people to use computer effectively, and tends to be much more consistent on Windows where it is The Way(tm) than on Macs, where the other buttons are only treated as shortcuts.
Actually, having the same message in several languages is very useful (see the Rosetta stone). Even with only a relatively small number of words it is relativly easy to distinguish different languages, even if you have never seen either of them.
2.6 is supposed to be fully preemptable, which should make lots of latencies decrease, leading to better interactive performance on a desktop, even if the overall throughput is lower. What this benchmark shows is that linux kernel 2.6 is a slower uniprocesser server than 2.4. While that is too bad, it doesn't really say much about desktop linux. I just installed 2.6-test5 on my (2 cpu) desktop, but haven't really had time to evaluate its performance relative to 2.4.
I understand the problems involved. I am just saying it annoyes me. While the problem isn't completely solvable, there is a lot that can be done. First, as done in this article, all the potentially slow network services can be started in parallel so they can timeout in parallel. Second, there could be robust dependency system that knew that my home computer doesn't require NIS for login, and doesn't NFS mount critical files.
Most linux systems I have seen are reasonably fast at booting, but could be much faster. Also, many things tend to behave badly when (for instance) the network port is unpluged. Starting things in the background would allow ntpdate, nfs, and all their friends to time out in the background when ifconfig failed. Hell, I wouldn't mind getting up gdm to start up and give a login prompt while still waiting for dhcpcd to time out.
Unfortunately, it only takes a small percentage of idiots to make us all suffer. With email, if 1E-5 people are gullible enough to be taken in by a spam, you can make more money than it takes to harrass the 99.999% of the rest of us.
Thankfully, the phone system is accountable, and therefore amenable to legal means to limit telemarketing, while email is flexible enough to support filtering and other technical (partial) solutions. Neither system will be perfect, but I think it strikes a fair compromise.
Just telling people to "not buy stuff from them and they will go away" is like saying "kids shouldn't be having sex" instead of teaching them about birth control and STDs. It works in principle, but is completely unproductive in practice.
There are a large number of international treaties on spectrum usage. However, I don't know that any of them apply to such high frequency signals. Mostly, they have to do with AM and lower bands (which have the possibility to go hundreds or thousands of miles, and thus become an international concern) or things like the FM bands used for commercial air flight. 2.4 GHz, however, had very little international concern except the interoperability of hardware between different contries. Even among the contries that declare 2.4 GHz as an unlicensed part of the spectrum, there are varying regulations on broadcast power and the like.
Yes. When someone says current flows from A to B, they (usually) mean 'positive current'. And usually, it doesn't matter. Usually (always? I don't remeber that class very well), electrons are the majority carrier in metals, but in semiconductors it can be either way.
>Wether such a description can be used in >gasous/plasma environment I'm not sure.
For sure, in plasma physics, both types of carriers are very important, though in this case it is electrons and ions, rather than holes.
Maybe in legalese "game product" and "including, but not limited to" have precise defenitions, but I really think this was a call to remove all content in anyway related to games. Manuals are "game products" and the list of games is not exclusive. To be safe, this guy really has to take everything down, then track down (not easy -- game companies change hands so often) game owner and see if it is OK to post manuals OR go to court to show that publishing manuals for obsolete games online is OK.
The latter is expensive and not guaranteed to win. The former is a PITA, and the companies involved are likely to say "no", not because it is costing them anything, but because it is the easy, default answer with no risk. Without the DMCA, they would have had to get a court order to force the material to be taken down, by placing that onus on the defender of the copyright, they would likely have found it easier to do nothing.
The DMCA forces the ISP to take the complaint as legitimate and take down the content or risk being included on a future suit, leaving the site owner responsible to prove his innocence before being allowed to put the material back up. With no penalty for unjustly accusing someone of copyright infringment, it make is too easy for lawyers "representing" IP holders to crawl the web looking for names that seem similar products they own and send out automatic threatening emails, with no human ever checking to see if the content is actually what is alleged, that it was not placed there with the owners permission, does not constitue legal use, or that the copyright owner just doesn't care.
In all likelyhood, this warning was sent based on a robot that assumes that the site was hosting ROM images. Redistribution of the technical manuals might (though unfortunately, probably not) constitute fair use, especially if the owners of the games are no longer selling said manuals (since it contains information require to make the machine you own work). Finally, people who actually have the authority to make the decision on whether they want this stopped probably don't care, and if they do, they shouldn't.
Since you were given the document willingly by the recipient, it is OK. The disclamer only applies if you receive the document (for instance) on the way to the intended recipient by an erronious fax or mail mishap. Even then, I am not sure if it is actually binding in any sense, but definately if the authorized recipient decides to share it with us all, it is legal.
To define art in terms of what you consider traditional forms, such as painting and sculpture, is to cut its balls of. Art is the combination of skill, intuition, and creativity to show people something, usually about humans, or how the universe relates to humans. Nothing less will do, and nothing more is required. The medium you choose to express yourself must fit the meaning. The reason for much of the dreadful "modern art" hanging in galleries is due to people trying to use traditional artistic media (such as oil on canvas) to express things it is not really suited for. Usually, when I walk through those sections of a museum the message I get is merely the artists frustration at trying to express what is certainly a very real feeling to them onto a 2 dimentional canvas.
On the other hand, non-traditional media are frequenly much better. For instance, the industrial sealed can of excrement is (at least to me) a statement about how we are frequently commercially driven to the point that we would buy almost anything if it were packaged in a nice way and well marketed, even a "turd in a can". The fact that a museum in fact paid a large sum of money for it only makes it more delightful.
Another example of modern art I heard about recently was a goldfish swimming in a blender, the idea being to force the people seeing it to confront the power they have of life or death (ie, they could switch on the blender and kill the fish). The great thing about that is that it engages the observer much more directly than any painting can. In fact, without a person there, it is just a fish tank.
To me, you sound like an old rich curmudgeon who was taught way back when you were a kid that certain things are art and certain things are not, and are unwilling to reconsider. Free your mind.
For a long time, your parents set you upright with a pat on the head everytime you fell while learning to walk, and you had the benefit of millions of years of evolution designing a body and brain nearly hardwired to walk.
I think you dramatically underestimate how much harder avoiding "turtle on back" failure modes is than chasing the ball (or playing soccer, or any other game with a well defined set of rules). To do it well probably requires an array of sensors comprable to that in a human or animal, plus a body of comprable complexity to the human skeletal-muscular system, plus the computing power to process all that sensor data.
That isn't to say this isn't an important problem, but lay off the people who choose to work on something else, and certainly don't call it cheating.
That is only true is chess is somehow "maximally complex" by which I mean that the most efficient algorithm for calculating the optimal move is a lookup table or exhaustive search. This is almost certainly not the case. Whether chess is solvable, then, is an unanswered question -- it may still be complex enough that no winning strategy can be found.
"Solving" chess will require a radical departure from current algorithms, which are basically finite depth exhaustive searches plus heuristics to throw out less promising avenues. Given that computing power looks to completely surpass the best humans in the world within the next decade or so, I don't know how many people will seriously study other approaches.
Go is actually a game I hold a lot of home for computers being able to solve. While the number of configurations is much larger than chess, the mechanics of the game are much simpler. I expect we are merely awaiting a mathematical revolution in go playing that will allow computers to handily beat humans even if doesn't completely solve the game.
Hey, I am as bleeding heart liberal as the come, and me and many of my friends wholeheartedly support after school Bible clubs. I realize that isn't the main point of your post, I just don't like seeing "liberal" equated with "stupid liberal".
I think the best solution for audiophile clubs would be to require an IQ test upon joining and every 6 months thereafter to make sure you aren't stupid enough to fall for the dumbass marketing propaganda and similar myths. Just as it is possible to not believe in God and approve of Bible study clubs, it is possible to seek the ultimate in sound reproduction without buying $400 interconnects and 30 year old tube amplifiers.
Besides, what is heaven if not perfect sound reproduction?
Actually, peak power output with 1% duty cycle is very important for music reproduction, as music tends to have a very high crest factor. The problem is that most manufacturers tend to lie about that as well, which is why it is better just to get average power ratings (RMS power is not useful, BTW).
Also, any well defined amplifier is capable of outputing nearly full power continously. Having a much higher peak output than average output is a sign of weak filter capacitors on the power supply, which will likely lead to poor performance at the rated power and terrible performance on those peaks that it claims to handle. The right thing to do is get an amp that can provide average power (nearly) equal to your peak power needs and run it well below that point.
Slashdot Mirror
Well, the original slimp3 didn't even have a "real" CPU, just a microcontroller and a dedicated MP3 decoder chip (actually an FPGA, I believe) that only supported decoding a single MP3 stream at a time. That makes crossfading basically impossible.
The new device is supposed to have a little bit more power, and supports wave file streaming, so at worst you could add a crossfading MP3 decoder on the server and stream wave data to it. I don't know if this is supported yet, but if not I am sure someone will implement it quickly.
One, you control it with a remote control in your living room. Two, if you have $500 (or more!) speakers you can use them with this. Three, it can plug into your home stereo/home theater amplifier and be integraded along side your CD/DVD/PS2/whatever. Four, it is transmitted as digital PCM over wireless ethernet rather than analog FM modulation (which is how many wireless speakers work), so there will be no sound degredation as the link fades until it is no longer able to maintain the bandwidth required.
:)
If you just want continious random play, and don't have a bunch of other AV gear you want to hook it into, this device is probably not for you, and you should go with the wireless speaker route. It will be simpler and cheaper. If you want hi-fi sound to your existing home theater system with a minimum of wires and maximum flexibility, you should order this (or something like it) right now
I have a home theater system. I still think it is lame to turn on my TV to listen to music. TV navigation would be a nice feature to have, but it is much more important to have a good front panel display.
Even if my TV is on, I would like to control the music while playing video games on my TV.
I don't like listening to music on my computer, either. It is not too annoying when I am just using it to play background music while I am working on my computer, mostly because it is essentially free, bu even then I would prefer to have a music playing device.
You can switch the text mode to "big" where you only get the bottom line of the display, but the text is twice as big. That is the way I use my slimp3. I can read the small text from across the room, but not easily. It would be nice if the display were about 50% larger, but I suspect that would be considerably more expensive (and involve a custom VFD instead of a standard size)
Many doctors will give experemental treatments to patients who don't respond to standard treatments, even if the patient is not in, or doesn't qualify for a study.
However, the biggest barrier is that most (all?) insurance companies will not cover experemental treatments. For members of test groups, the costs are usually borne by the research grant for the study, but not so for tag-a-longs.
This is indeed a question I have thought about a lot. I don't really know the answer, somehow it seems a little unsavory (if not wrong), but as you say, who gets to decide what makes it happy.
My guess is that it would be fine, but I wouldn't be surprised if said AI decided that despite its natural predelections, it wanted to be a painter. It happens often enough with humans that we conciously override our instinctual behavior for one reason or another, I don't see why that shouldn't happen to machines as well.
My personal feeling is that much of this is false savings. The fact is, from a usability standpoint, web clients totally suck for a lot of applications. Even a well designed web client can substantially imact user productivity compared to a dedicated thin client application (thin in the sense of being "interface only" while allowing the server to do all the work and data storage).
Web clients are going to stick around for a while, but but their days are already numbered.
I find it strange and worrisome that you seem more comfortable damning an entire new form of sentient entities into potential slavery than with attempting to extend the same basic rights we believe humans deserve by virtue of existence.
You are the people that hold humanity back, and slow us from recognizing our potential. Our domain is the sun and the stars and all of creation. We will one day make computers that are our mental equals, and it is a mark of sanity that we begin to come to grips with that fact before it happens, so that we may be ready for that day when it comes.
Unfortunately, that isn't always true. A lot of mac software doesn't take advantage of extra buttons simply because they are not present in the default configurations. Also, the Windows (and other) "left button performs the default action, right button gives you a list of actions" is a really powerful framework for allowing people to use computer effectively, and tends to be much more consistent on Windows where it is The Way(tm) than on Macs, where the other buttons are only treated as shortcuts.
Actually, having the same message in several languages is very useful (see the Rosetta stone). Even with only a relatively small number of words it is relativly easy to distinguish different languages, even if you have never seen either of them.
2.6 is supposed to be fully preemptable, which should make lots of latencies decrease, leading to better interactive performance on a desktop, even if the overall throughput is lower. What this benchmark shows is that linux kernel 2.6 is a slower uniprocesser server than 2.4. While that is too bad, it doesn't really say much about desktop linux. I just installed 2.6-test5 on my (2 cpu) desktop, but haven't really had time to evaluate its performance relative to 2.4.
I understand the problems involved. I am just saying it annoyes me. While the problem isn't completely solvable, there is a lot that can be done. First, as done in this article, all the potentially slow network services can be started in parallel so they can timeout in parallel. Second, there could be robust dependency system that knew that my home computer doesn't require NIS for login, and doesn't NFS mount critical files.
Most linux systems I have seen are reasonably fast at booting, but could be much faster. Also, many things tend to behave badly when (for instance) the network port is unpluged. Starting things in the background would allow ntpdate, nfs, and all their friends to time out in the background when ifconfig failed. Hell, I wouldn't mind getting up gdm to start up and give a login prompt while still waiting for dhcpcd to time out.
Unfortunately, it only takes a small percentage of idiots to make us all suffer. With email, if 1E-5 people are gullible enough to be taken in by a spam, you can make more money than it takes to harrass the 99.999% of the rest of us.
Thankfully, the phone system is accountable, and therefore amenable to legal means to limit telemarketing, while email is flexible enough to support filtering and other technical (partial) solutions. Neither system will be perfect, but I think it strikes a fair compromise.
Just telling people to "not buy stuff from them and they will go away" is like saying "kids shouldn't be having sex" instead of teaching them about birth control and STDs. It works in principle, but is completely unproductive in practice.
There are a large number of international treaties on spectrum usage. However, I don't know that any of them apply to such high frequency signals. Mostly, they have to do with AM and lower bands (which have the possibility to go hundreds or thousands of miles, and thus become an international concern) or things like the FM bands used for commercial air flight. 2.4 GHz, however, had very little international concern except the interoperability of hardware between different contries. Even among the contries that declare 2.4 GHz as an unlicensed part of the spectrum, there are varying regulations on broadcast power and the like.
Yes. When someone says current flows from A to B, they (usually) mean 'positive current'. And usually, it doesn't matter. Usually (always? I don't remeber that class very well), electrons are the majority carrier in metals, but in semiconductors it can be either way.
>Wether such a description can be used in >gasous/plasma environment I'm not sure.
For sure, in plasma physics, both types of carriers are very important, though in this case it is electrons and ions, rather than holes.
Maybe in legalese "game product" and "including, but not limited to" have precise defenitions, but I really think this was a call to remove all content in anyway related to games. Manuals are "game products" and the list of games is not exclusive. To be safe, this guy really has to take everything down, then track down (not easy -- game companies change hands so often) game owner and see if it is OK to post manuals OR go to court to show that publishing manuals for obsolete games online is OK.
The latter is expensive and not guaranteed to win. The former is a PITA, and the companies involved are likely to say "no", not because it is costing them anything, but because it is the easy, default answer with no risk. Without the DMCA, they would have had to get a court order to force the material to be taken down, by placing that onus on the defender of the copyright, they would likely have found it easier to do nothing.
In this case, both parties effectively knew the communicaton was recorded. Such laws do not apply to written communication.
The DMCA forces the ISP to take the complaint as legitimate and take down the content or risk being included on a future suit, leaving the site owner responsible to prove his innocence before being allowed to put the material back up. With no penalty for unjustly accusing someone of copyright infringment, it make is too easy for lawyers "representing" IP holders to crawl the web looking for names that seem similar products they own and send out automatic threatening emails, with no human ever checking to see if the content is actually what is alleged, that it was not placed there with the owners permission, does not constitue legal use, or that the copyright owner just doesn't care.
In all likelyhood, this warning was sent based on a robot that assumes that the site was hosting ROM images. Redistribution of the technical manuals might (though unfortunately, probably not) constitute fair use, especially if the owners of the games are no longer selling said manuals (since it contains information require to make the machine you own work). Finally, people who actually have the authority to make the decision on whether they want this stopped probably don't care, and if they do, they shouldn't.
Since you were given the document willingly by the recipient, it is OK. The disclamer only applies if you receive the document (for instance) on the way to the intended recipient by an erronious fax or mail mishap. Even then, I am not sure if it is actually binding in any sense, but definately if the authorized recipient decides to share it with us all, it is legal.
To define art in terms of what you consider traditional forms, such as painting and sculpture, is to cut its balls of. Art is the combination of skill, intuition, and creativity to show people something, usually about humans, or how the universe relates to humans. Nothing less will do, and nothing more is required. The medium you choose to express yourself must fit the meaning. The reason for much of the dreadful "modern art" hanging in galleries is due to people trying to use traditional artistic media (such as oil on canvas) to express things it is not really suited for. Usually, when I walk through those sections of a museum the message I get is merely the artists frustration at trying to express what is certainly a very real feeling to them onto a 2 dimentional canvas.
On the other hand, non-traditional media are frequenly much better. For instance, the industrial sealed can of excrement is (at least to me) a statement about how we are frequently commercially driven to the point that we would buy almost anything if it were packaged in a nice way and well marketed, even a "turd in a can". The fact that a museum in fact paid a large sum of money for it only makes it more delightful.
Another example of modern art I heard about recently was a goldfish swimming in a blender, the idea being to force the people seeing it to confront the power they have of life or death (ie, they could switch on the blender and kill the fish). The great thing about that is that it engages the observer much more directly than any painting can. In fact, without a person there, it is just a fish tank.
To me, you sound like an old rich curmudgeon who was taught way back when you were a kid that certain things are art and certain things are not, and are unwilling to reconsider. Free your mind.
Who said you get to write the rules?
For a long time, your parents set you upright with a pat on the head everytime you fell while learning to walk, and you had the benefit of millions of years of evolution designing a body and brain nearly hardwired to walk.
I think you dramatically underestimate how much harder avoiding "turtle on back" failure modes is than chasing the ball (or playing soccer, or any other game with a well defined set of rules). To do it well probably requires an array of sensors comprable to that in a human or animal, plus a body of comprable complexity to the human skeletal-muscular system, plus the computing power to process all that sensor data.
That isn't to say this isn't an important problem, but lay off the people who choose to work on something else, and certainly don't call it cheating.
That is only true is chess is somehow "maximally complex" by which I mean that the most efficient algorithm for calculating the optimal move is a lookup table or exhaustive search. This is almost certainly not the case. Whether chess is solvable, then, is an unanswered question -- it may still be complex enough that no winning strategy can be found.
"Solving" chess will require a radical departure from current algorithms, which are basically finite depth exhaustive searches plus heuristics to throw out less promising avenues. Given that computing power looks to completely surpass the best humans in the world within the next decade or so, I don't know how many people will seriously study other approaches.
Go is actually a game I hold a lot of home for computers being able to solve. While the number of configurations is much larger than chess, the mechanics of the game are much simpler. I expect we are merely awaiting a mathematical revolution in go playing that will allow computers to handily beat humans even if doesn't completely solve the game.
Hey, I am as bleeding heart liberal as the come, and me and many of my friends wholeheartedly support after school Bible clubs. I realize that isn't the main point of your post, I just don't like seeing "liberal" equated with "stupid liberal".
I think the best solution for audiophile clubs would be to require an IQ test upon joining and every 6 months thereafter to make sure you aren't stupid enough to fall for the dumbass marketing propaganda and similar myths. Just as it is possible to not believe in God and approve of Bible study clubs, it is possible to seek the ultimate in sound reproduction without buying $400 interconnects and 30 year old tube amplifiers.
Besides, what is heaven if not perfect sound reproduction?
Actually, peak power output with 1% duty cycle is very important for music reproduction, as music tends to have a very high crest factor. The problem is that most manufacturers tend to lie about that as well, which is why it is better just to get average power ratings (RMS power is not useful, BTW).
Also, any well defined amplifier is capable of outputing nearly full power continously. Having a much higher peak output than average output is a sign of weak filter capacitors on the power supply, which will likely lead to poor performance at the rated power and terrible performance on those peaks that it claims to handle. The right thing to do is get an amp that can provide average power (nearly) equal to your peak power needs and run it well below that point.