I used to teach kids programming (at the Computer Ed summer camps in Boston) and I had kids programming in all sorts of crazy languages. I think that it's wonderful what kids can achieve when they're excited about learning.
A few random pleasant memories: - I was teaching a little girl to program in C. She was pretty good, given that we were using pretty primal tools (I think it was Turbo C on my Osborne Executive). The best part was that she was so tiny that she had to reach _way_ up to hold my hand when we crossed the street that ran through the camp. That just blew my mind -- one minute this brilliant kid was coding a sort routine in C, and the next she was a timid little girl holding my hand crossing the street. - I had a whole gang of kids using the Lisp built into the BBC Micro (Acorn?). We had great fun writing an adventure game with a simple parser, so that kids could move around a simple network of rooms, pick stuff up and move it around and drop it. Some of the older kids implemented locking and unlocking doors. Pretty good for a two week, one hour a day course. - A bunch of the older kids learned 6502 assembler on the Apple ][, using a simple assembler and the ROM debugger. Unlike the x86's, the 6502 is so simple to program (very clean design) that by the end of the class some of the kids were reading the binary straight rather than disassembling it. We wrote killer video games -- they had snakes running around the screen, gobbling "apples" and growing longer, until you hit a wall and the game ended. That was two weeks at 2 hours a day, so it was only for the most dedicated little geeks. - Programming Robot Wars -- that was a very simple assembly language that controlled simulated robots. They loved coding their robots and seeing whose robot won. The modern robot simulators are superior in every way (e.g. alphaworks' Robocode, but Robot Wars was nice and simple and fun. - Logo, of course. It's an amazing language. People usually think of it as a simple language for teaching, and it's great for that, but it's actually nearly identical to Lisp, so you can do all of the cool recursion, etc., in Logo. The usual stages of the day were Logo for little kids, then BASIC, then Pascal for the advanced students. I found that kids that went straight from Logo to Pascal did 100% better than the kids who were taught BASIC -- the BASIC kids had so many stupid ideals drilled into them that they were almost incapable of programming. But straight from Logo to Pascal was easy -- though the kids did complain about having to wait for things to compile.:-) - Logo turtles -- the ones that were little robots that ran around on the floor, with a pen and an optical sensor. Those were fun...
Man, that was fun. I've got to get back into teaching.
Well, I think that people fall into three categories. In terms of your analogy to speeding drivers: A) Some drivers want to follow the law, B) Some drivers don't want to get arrested, and C) Some drivers will speed no matter what.
So my theory is that the (A) file sharers will stop as soon as they realize that it's illegal, or the same reason that they stop at stop signs even when there aren't any police around. I think that these people are now "permanently" off of p2p (unless there's some radical development making p2p file sharing of copyrighted music legal).
The (B) file sharers will stop when they're afraid of enforcement, and will start when they aren't. These people will cause p2p usage to bounce up and down based on headlines and court cases.
The (C) file sharers aren't going to stop. They're the ones switching p2p networks in order to avoid detection, signing up for "secure" networks, etc.
So I guess the question is how large A, B and C populations are. I'm guessing that A>B>C, but time will tell...
In theory, this is true. In practice, most people trust the same name brands online that they do offline, and thus type the URL's that are the names of companies that they like and trust. So while there are certainly many new and cool "independent" web sites, etc., morepeople go to CNN.com for news, TVGuide.com for TV listings and reviews, and so on. There are of course some exceptions (Google, Amazon). So while there is an opportunity for people to explore outside of the established brands, the mainstream users stick to brands that they know.
"News flash: *all* IP addresses are associated with a geographical location."
This is technically true, but not in a useful manner, because private networks all map to the location of their internet gateway. For example, all AOL users appear to be in Virginia, even if they're in another country. All employees on a corporate network appear to be wherever the company internet gateway is located. And so on.
There are also plenty of "edge" cases, like people who live in one city or state, but whose network access is mapped to another city or state.
So IP to geographical location mapping is good enough for simple marketing analysis (almost always the right country, usually the right state, often the right city), but nowhere near a precise location for use in legal proceedings.
I'd assume that once they find out people's identifies that they'll refile in the appropriate locations.
"They are going after the right people but it still won't "work" in terms of solving the problem of stopping or significantly reducing infringement in the long term."
The real goal isn't (IMO) to arrest so many people that people avoid p2p out of fear of arrest -- anyone can do the math and figure out that the odds of getting caught using filesharing is quite low (1,500 divided by 10M is very, very small). The goal (IMO) is to change how people think about file sharing, specifically so that people that thought that file sharing networks were legitimate music services find out that they gave money to scammers who are "selling" someone else's work. And every time people see a headline about someone payuing a settlement, or losing a lawsuit, it reinforces the message that people who share files are criminals. Most people want to be honest, so if the lawsuits make millions of people believe that usnig file sharing networks is wrong, then many of those people leave the networks.
From that perspective, it seems to be working. Perhaps 1,500 or so people got sued, but around 10M people left the p2p networks. Which number do you think is more important?
Man, I'm tired of this "12 year old girl" story. The RIAA didn't sue the 12 year old girl, the RIAA sued the person who paid the ISP, which is the mother. The mother's defense was to blame her daughter for the downloading. That doesn't make as dramatic a headline about the evil RIAA.
"They claim that Open Source threatens "[The US] continued ability to lead the world in technological innovation/[The US] international competitive position in the global software industry"."
Is it better than open source for the US for a US corporation to produce a proprietary product, implemented by outsourced offshore programmers? It might be better for the investors in that company (who make money on the code as it flows through them), but it appears to make no difference for US programmers (who have no jobs either way), and it's worse for the company's customers (who are paying for something that they could have had for free). Come to think of it, for open source software, at least local programmers would get jobs setting up, configuring and customizing the software...
I agree that "Peer-to-peer gained overnight popularity because of the Britneys and Outkasts", but I'd suggest a few corrections:
"As a matter of fact, it is their growth rate that is down, not overall profitability"
Total recorded music sales (in dollars) dropped by 13.5% from its peak in 1999 until the end of 2002 (the last full year numbers on the RIAA site), and dropped 12% for the first half of 2003 vs. first half of 2002. So the total sales numbers are clearly down quite a bit. Given all of the layoffs at music companies over the last year, I'd have to think that the ~20% drop in sales probably affected profitability a bit.
"that growth began to decline well before Napster"
The music industry started shrinking slightly in 2000, and clearly dropping in 2001 and faster in 2002 and 2003. Napster was launched in 2000. That's not to say that Napster caused the drop in sales (there are plenty of other things that also happened, such as the slowing of the LP to CD replacement cycle), but the sequence makes it possible that Napster accellerated the drop in sales.
"Let's also not forget that truly vast quantities of good older music (from which the music outfits have already made their pile) are traded online regularly because there is no other way to get it. Copyright infringement it may be, but it certainly is no loss to the music industry since they don't sell it anymore anyway."
The music companies keep older music available. Look at Rhino, for example. Sure, not everything is kept in print forever, but back catalog sales are a good business.
On a 500 MHz PPC the CPU will be the bottlenech, not the drive(s). So all that RIPping from multiple drives will do is keep the process pipelined a bit better. That is, instead of being able to ignore your computer for twenty minutes as it RIP's one CD, you can ignore if for an hour as it RIP's three CD's.
I don't have multiple CD drives, but perhaps you could test this -- what does iTunes do if you tell it to automatically RIP and eject CD's and you put CD's in multiple drives? I'd guess (since most Apple software is pretty clever) that it would simply work its way through the inserted CD's, in which case you don't need any software -- just load all three of your drives and let iTunes do its work, and stick in new CD's every so often.
I'm so sick of pretentious twits who whine about how much better looking asian women are. First, it's a matter of taste. Second, asian women are tools. I don't know about you, but I don't buy asian women based on how asthetically pleasing they are.
I'm so sick of pretentious twits who whine about how much better looking Biomorph's desks are. First, it's a matter of taste. Second, desks are tools. I don't know about you, but I don't buy desks based on how asthetically pleasing they are.
I'm so sick of pretentious twits who whine about how much better looking BMW's cars are. First, it's a matter of taste. Second, cars are tools. I don't know about you, but I don't buy cars based on how asthetically pleasing they are.
The real problem with Star Trek is that it's controlled by a pair of egotists who:
1) Think of Star Trek as a franchise to be exploited, instead of as an opportunity to tell great stories. This means that everything is derivitive retreads of existing material, because that's the safest tactic. This is whey don't hire real SF writers (e.g. Niven, Ellison, Gerrold) but instead hire TV writers who slap SF gadgetry and doubletalk over generic TV show plots.
2) resent the fact that everyone likes Gene Roddenberry's work better, and keep trying to create an "original vision" instead of executing GR's vision well. This is why they even took the name "Star Trek" off of Enterprise. They don't want to make Star Trek a success, they want to make something "new" a success, only they don't have the guts to actually create anything new, so they're trying to hijack Star Trek. This same issue is why the movie of Dune sucked (the director didn't want to simply film Dune, but had to get his ego involved), but LOTR was wonderful (Jackson told the original story perfectly, no ego BS), only Enterprise gets to suck weekly.
It's a shame, since Star Trek has so many fans, and the actors and effects in Enterprise are first rate. It's just the writing that sucks.
My advice: hire real SF writers and give them real creative control. Or watch Outer Limits instead. Or Farscape, Lexx, or SG-1....
Eliminating the physical costs, you're right, doesn't save much. Eliminating the retailer could save margin, but then fans would have to go to every artist or label's site to buy music -- the retailers provide a service by aggregating the music and presenting it in an organized manner, covering the cost of actually selling the music (running ads for the store, paying credit card processing fees, staffing customer support, etc.). So if you eliminate the margin that they charge, that means that the labels or artists need to do all of the work that retailers do, which (given that the online retailers aren't particularly profitable yet) would cost about as much as the margin that they're charging.
"The music industry can cut prices far more than they have. A download costs them next to nothing when compared to producing and distributing physical media."
If an album sells for $14, and costs $3 to physically produce (packaging, etc.), that means that a purely digital album could sell for $11 and still pay the same amount to the store, distributor, label and artists. That puts you at very roughly $1 a track.
If you view the digital sales as an incremental business on top of CD sales, you could sell music digitally for albums nothing. But if digital sales _replace_ CD sales, that changes the equation. If an album can sell 100K CD's for $14, that generates $1.4M in sales (split between stores, shipping, label, artist, etc.). If you sell 100K copies of the hit single for 5 cents, that's $5K (i.e. you don't bother spending the $100K to record the album). If you sell 100K albums for 75 cents (15 tracks at 5 cents), that's $75K. Would anyone here spend $100K to produce a product, then risk a $1.4M deal to have a shot at a $75K deal?
To make the same money as the 100K CD sales with 5 cent tracks, you'd have to sell 20M tracks. That is, that one album would have to sell nearly as many tracks as the iTunes Store's entire catalog has sold since it launched. Seems unlikely this year...
Now, if you argue for 50 cent tracks, that's not as bad a deal. You'd only need to generate 2.8M track sales. That's far more than anything's sold so far online, but doesn't seem completely impossible as the world grows increasingly digital.
If I had to guess, prices will stick at around $1 a track for a while, at least until the volumes ramp up.
"If the record industry wanted to kill illegal file-swapping instantly, they could do it simply by offering up their entire catalog for online distribution at the price of $0.05 per track, with little or no DRM. Boom. Illegal file-sharing gone."
While I agree that this would eliminate illegal file swapping, it's also a huge roll of the dice, betting that in return for charging 5 cents a track, people would buy 20x as much music. And if they're wrong, the music business dies.
This depends on what the fingerprinting technology is. Audible Magic's technology, for example, can supposedly generate fingerprints for audio files that are independent of specific encoding. I don't know whether they'd end up with a single fingerprint for all versions of a track, or a couple, but in either case the number of fingerprints would be much smaller than the number of hash values.
Also, the system doesn't need to block every single copyrighted song to be effective, because not all songs are equally popular. There are perhaps 1m copyrighted tracks,
There are perhaps 1m copyrighted tracks. That's probably still too many to block. Luckily, tracks are not all equally popular, so if you block the most popular tracks, that would be disproportionately effective. Perhaps the Rolling Stone 500 Greatest Albums of All Time could be a start. That's only, say, 10,000 tracks.
How hard could it be to transmit 10,000 numeric fingerprints, and block sharing of any files that match them? P2P networks are great at transmitting data, so it'd be easy to update the list.:-)
Load a fingerprinted file. Change one bit. It has a new fingerprint."
Right, and all of the swarmed download technologies go right out the window. And the ability to tell which files have multiple sources. And Bitzi.
So I'd say that yes, common snse went on holidays.:-)
This also misses the point of the discussion. The p2p networks' legal defense has been that they have no knowledge or control over what's going on with the network. Someone just demonstrated that a pretty obvious technical approach can, in fact, control what's moving around the p2p networks by blocking the fingerprints of copyrighted material. So the fact that they don't do this isn't because it's impossible, but because they choose not to do so.
This is important because the p2p networks' defense has been that they're innocent, and that the music company's only recourse is to sue the evil, law-breaking users. If the courts rule that the p2p networks are responsible for not blocking copyrighted materials, the music companies can stop suing individuals...
"So? How did they determine that those IPs should be blocked, and more importantly, that the rest shouldn't be?"
IP blocking wouldn't affect this survey, because the data was collected by (1) running software (with permission) on panel member's computers that watches what they do, and (2) by traditional paper market surveys.
IP blocking would only block external monitoring of the computer. That might effect other surveys, as far as users use IP blocking software...
"Sun is big about "scalability" aspects of it. The problem is that this only applies to problems that can't work on one computer, but *can* work on one computer times some constant that's relatively small from a computer science standpoint."
Why? A well written distributed application can scale to thousands of CPU's. So if you have a big problem to solve (weather modeling, Computational Fluid Dynamics, high volume transaction processing, molecular modeling, rendering movies, etc.), you can either solve it by waiting 15 years for CPU's to get 1,024 times faster, or by clustering 1,024 CPU's. So I agree with your point at as CPU's get faster, more and more problems can be solved on a single CPU, but I think that you're underestimating the number of problems that require lots of computational power to solve.
You don't think that eBay, or a bank, or an airline runs on 1 CPU, do you?
"The only benefits I see to IBM's approach is that companies can theoretically replace their local Windows boxes with thin clients and fire their local IT staff"
This isn't about the kinds of applications that you run on desktops! This is about being able to build virtual mainframes out of commodity "desktop" components.
Well, Thinking Machines got bought out (in effect) when parallel computing made the transition from a "niche" to a "mainstream" technology -- once IBM, Sun, etc., all started selling MPP supercomputers, most of the specialized MPP companies went under or were bought by mainstream computer companies. In Thinking Machine's case, my recollection is that Oracle bought the Data Mining team, and Sun bought the compiler and OS team, and Danny Hillis went on to run a really cool group at Disney and to design a really cool clock. So Thinking Machine's technology and architecture won (virtually all of the world's fastest supercomputers are MPP), the company didn't survive the transition.
Shame -- the people were really cool, and the food was amazing.
"Or the hardest problem of all: how often does an organization need CPU power beyond that of a typical modern desktop machine?"
This isn't a technique that applies to word processing. Think about enterprise class problems such as data mining. When you're American Express, analyzing terabytes of purchasing data to identify customer behavior models (which they do, and make a fortune on), you don't do that on a desktop computer. When you're an oil company, processing seismic survey data, you don't do that on a desktop computer. When you're computing payrolls for 20,000 employees you don't do that on a desktop computer. You get the idea.
So while desktop computers are wonderful and astoundingly fast, there are important business problems that can't be solved in a reasonable period of time on one. That's why people make servers, mainframes, etc. All this is, is an attempt to use clever software to make distributed computing as manageable as mainframes are, while taking advantage of the astounding better price/performance that you get for using commodity components instead of niche hardware.
Slashdot Mirror
I used to teach kids programming (at the Computer Ed summer camps in Boston) and I had kids programming in all sorts of crazy languages. I think that it's wonderful what kids can achieve when they're excited about learning.
:-)
A few random pleasant memories:
- I was teaching a little girl to program in C. She was pretty good, given that we were using pretty primal tools (I think it was Turbo C on my Osborne Executive). The best part was that she was so tiny that she had to reach _way_ up to hold my hand when we crossed the street that ran through the camp. That just blew my mind -- one minute this brilliant kid was coding a sort routine in C, and the next she was a timid little girl holding my hand crossing the street.
- I had a whole gang of kids using the Lisp built into the BBC Micro (Acorn?). We had great fun writing an adventure game with a simple parser, so that kids could move around a simple network of rooms, pick stuff up and move it around and drop it. Some of the older kids implemented locking and unlocking doors. Pretty good for a two week, one hour a day course.
- A bunch of the older kids learned 6502 assembler on the Apple ][, using a simple assembler and the ROM debugger. Unlike the x86's, the 6502 is so simple to program (very clean design) that by the end of the class some of the kids were reading the binary straight rather than disassembling it. We wrote killer video games -- they had snakes running around the screen, gobbling "apples" and growing longer, until you hit a wall and the game ended. That was two weeks at 2 hours a day, so it was only for the most dedicated little geeks.
- Programming Robot Wars -- that was a very simple assembly language that controlled simulated robots. They loved coding their robots and seeing whose robot won. The modern robot simulators are superior in every way (e.g. alphaworks' Robocode, but Robot Wars was nice and simple and fun.
- Logo, of course. It's an amazing language. People usually think of it as a simple language for teaching, and it's great for that, but it's actually nearly identical to Lisp, so you can do all of the cool recursion, etc., in Logo. The usual stages of the day were Logo for little kids, then BASIC, then Pascal for the advanced students. I found that kids that went straight from Logo to Pascal did 100% better than the kids who were taught BASIC -- the BASIC kids had so many stupid ideals drilled into them that they were almost incapable of programming. But straight from Logo to Pascal was easy -- though the kids did complain about having to wait for things to compile.
- Logo turtles -- the ones that were little robots that ran around on the floor, with a pen and an optical sensor. Those were fun...
Man, that was fun. I've got to get back into teaching.
I wish I had moderator points to give you!
I agree completely. And, for the record, I think that having all of these new options is a wonderful thing!
"I just don't think it's going to work."
Well, I think that people fall into three categories. In terms of your analogy to speeding drivers:
A) Some drivers want to follow the law,
B) Some drivers don't want to get arrested, and
C) Some drivers will speed no matter what.
So my theory is that the (A) file sharers will stop as soon as they realize that it's illegal, or the same reason that they stop at stop signs even when there aren't any police around. I think that these people are now "permanently" off of p2p (unless there's some radical development making p2p file sharing of copyrighted music legal).
The (B) file sharers will stop when they're afraid of enforcement, and will start when they aren't. These people will cause p2p usage to bounce up and down based on headlines and court cases.
The (C) file sharers aren't going to stop. They're the ones switching p2p networks in order to avoid detection, signing up for "secure" networks, etc.
So I guess the question is how large A, B and C populations are. I'm guessing that A>B>C, but time will tell...
"On the Internet, name brand means nothing."
In theory, this is true. In practice, most people trust the same name brands online that they do offline, and thus type the URL's that are the names of companies that they like and trust. So while there are certainly many new and cool "independent" web sites, etc., morepeople go to CNN.com for news, TVGuide.com for TV listings and reviews, and so on. There are of course some exceptions (Google, Amazon). So while there is an opportunity for people to explore outside of the established brands, the mainstream users stick to brands that they know.
"News flash: *all* IP addresses are associated with a geographical location."
This is technically true, but not in a useful manner, because private networks all map to the location of their internet gateway. For example, all AOL users appear to be in Virginia, even if they're in another country. All employees on a corporate network appear to be wherever the company internet gateway is located. And so on.
There are also plenty of "edge" cases, like people who live in one city or state, but whose network access is mapped to another city or state.
So IP to geographical location mapping is good enough for simple marketing analysis (almost always the right country, usually the right state, often the right city), but nowhere near a precise location for use in legal proceedings.
I'd assume that once they find out people's identifies that they'll refile in the appropriate locations.
"They are going after the right people but it still won't "work" in terms of solving the problem of stopping or significantly reducing infringement in the long term."
The real goal isn't (IMO) to arrest so many people that people avoid p2p out of fear of arrest -- anyone can do the math and figure out that the odds of getting caught using filesharing is quite low (1,500 divided by 10M is very, very small). The goal (IMO) is to change how people think about file sharing, specifically so that people that thought that file sharing networks were legitimate music services find out that they gave money to scammers who are "selling" someone else's work. And every time people see a headline about someone payuing a settlement, or losing a lawsuit, it reinforces the message that people who share files are criminals. Most people want to be honest, so if the lawsuits make millions of people believe that usnig file sharing networks is wrong, then many of those people leave the networks.
From that perspective, it seems to be working. Perhaps 1,500 or so people got sued, but around 10M people left the p2p networks. Which number do you think is more important?
Man, I'm tired of this "12 year old girl" story. The RIAA didn't sue the 12 year old girl, the RIAA sued the person who paid the ISP, which is the mother. The mother's defense was to blame her daughter for the downloading. That doesn't make as dramatic a headline about the evil RIAA.
"They claim that Open Source threatens "[The US] continued ability to lead the world in technological innovation/[The US] international competitive position in the global software industry"."
Is it better than open source for the US for a US corporation to produce a proprietary product, implemented by outsourced offshore programmers? It might be better for the investors in that company (who make money on the code as it flows through them), but it appears to make no difference for US programmers (who have no jobs either way), and it's worse for the company's customers (who are paying for something that they could have had for free). Come to think of it, for open source software, at least local programmers would get jobs setting up, configuring and customizing the software...
I agree that "Peer-to-peer gained overnight popularity because of the Britneys and Outkasts", but I'd suggest a few corrections:
"As a matter of fact, it is their growth rate that is down, not overall profitability"
Total recorded music sales (in dollars) dropped by 13.5% from its peak in 1999 until the end of 2002 (the last full year numbers on the RIAA site), and dropped 12% for the first half of 2003 vs. first half of 2002. So the total sales numbers are clearly down quite a bit. Given all of the layoffs at music companies over the last year, I'd have to think that the ~20% drop in sales probably affected profitability a bit.
"that growth began to decline well before Napster"
The music industry started shrinking slightly in 2000, and clearly dropping in 2001 and faster in 2002 and 2003. Napster was launched in 2000. That's not to say that Napster caused the drop in sales (there are plenty of other things that also happened, such as the slowing of the LP to CD replacement cycle), but the sequence makes it possible that Napster accellerated the drop in sales.
"Let's also not forget that truly vast quantities of good older music (from which the music outfits have already made their pile) are traded online regularly because there is no other way to get it. Copyright infringement it may be, but it certainly is no loss to the music industry since they don't sell it anymore anyway."
The music companies keep older music available. Look at Rhino, for example. Sure, not everything is kept in print forever, but back catalog sales are a good business.
Cool! I guess that jasonisnuts doesn't need any software after all... ;-)
On a 500 MHz PPC the CPU will be the bottlenech, not the drive(s). So all that RIPping from multiple drives will do is keep the process pipelined a bit better. That is, instead of being able to ignore your computer for twenty minutes as it RIP's one CD, you can ignore if for an hour as it RIP's three CD's.
I don't have multiple CD drives, but perhaps you could test this -- what does iTunes do if you tell it to automatically RIP and eject CD's and you put CD's in multiple drives? I'd guess (since most Apple software is pretty clever) that it would simply work its way through the inserted CD's, in which case you don't need any software -- just load all three of your drives and let iTunes do its work, and stick in new CD's every so often.
I'm so sick of pretentious twits who whine about how much better looking asian women are. First, it's a matter of taste. Second, asian women are tools. I don't know about you, but I don't buy asian women based on how asthetically pleasing they are.
I'm so sick of pretentious twits who whine about how much better looking Biomorph's desks are. First, it's a matter of taste. Second, desks are tools. I don't know about you, but I don't buy desks based on how asthetically pleasing they are.
I'm so sick of pretentious twits who whine about how much better looking BMW's cars are. First, it's a matter of taste. Second, cars are tools. I don't know about you, but I don't buy cars based on how asthetically pleasing they are.
The real problem with Star Trek is that it's controlled by a pair of egotists who:
1) Think of Star Trek as a franchise to be exploited, instead of as an opportunity to tell great stories. This means that everything is derivitive retreads of existing material, because that's the safest tactic. This is whey don't hire real SF writers (e.g. Niven, Ellison, Gerrold) but instead hire TV writers who slap SF gadgetry and doubletalk over generic TV show plots.
2) resent the fact that everyone likes Gene Roddenberry's work better, and keep trying to create an "original vision" instead of executing GR's vision well. This is why they even took the name "Star Trek" off of Enterprise. They don't want to make Star Trek a success, they want to make something "new" a success, only they don't have the guts to actually create anything new, so they're trying to hijack Star Trek. This same issue is why the movie of Dune sucked (the director didn't want to simply film Dune, but had to get his ego involved), but LOTR was wonderful (Jackson told the original story perfectly, no ego BS), only Enterprise gets to suck weekly.
It's a shame, since Star Trek has so many fans, and the actors and effects in Enterprise are first rate. It's just the writing that sucks.
My advice: hire real SF writers and give them real creative control. Or watch Outer Limits instead. Or Farscape, Lexx, or SG-1....
Eliminating the physical costs, you're right, doesn't save much. Eliminating the retailer could save margin, but then fans would have to go to every artist or label's site to buy music -- the retailers provide a service by aggregating the music and presenting it in an organized manner, covering the cost of actually selling the music (running ads for the store, paying credit card processing fees, staffing customer support, etc.). So if you eliminate the margin that they charge, that means that the labels or artists need to do all of the work that retailers do, which (given that the online retailers aren't particularly profitable yet) would cost about as much as the margin that they're charging.
"The music industry can cut prices far more than they have. A download costs them next to nothing when compared to producing and distributing physical media."
If an album sells for $14, and costs $3 to physically produce (packaging, etc.), that means that a purely digital album could sell for $11 and still pay the same amount to the store, distributor, label and artists. That puts you at very roughly $1 a track.
If you view the digital sales as an incremental business on top of CD sales, you could sell music digitally for albums nothing. But if digital sales _replace_ CD sales, that changes the equation. If an album can sell 100K CD's for $14, that generates $1.4M in sales (split between stores, shipping, label, artist, etc.). If you sell 100K copies of the hit single for 5 cents, that's $5K (i.e. you don't bother spending the $100K to record the album). If you sell 100K albums for 75 cents (15 tracks at 5 cents), that's $75K. Would anyone here spend $100K to produce a product, then risk a $1.4M deal to have a shot at a $75K deal?
To make the same money as the 100K CD sales with 5 cent tracks, you'd have to sell 20M tracks. That is, that one album would have to sell nearly as many tracks as the iTunes Store's entire catalog has sold since it launched. Seems unlikely this year...
Now, if you argue for 50 cent tracks, that's not as bad a deal. You'd only need to generate 2.8M track sales. That's far more than anything's sold so far online, but doesn't seem completely impossible as the world grows increasingly digital.
If I had to guess, prices will stick at around $1 a track for a while, at least until the volumes ramp up.
"If the record industry wanted to kill illegal file-swapping instantly, they could do it simply by offering up their entire catalog for online distribution at the price of $0.05 per track, with little or no DRM. Boom. Illegal file-sharing gone."
While I agree that this would eliminate illegal file swapping, it's also a huge roll of the dice, betting that in return for charging 5 cents a track, people would buy 20x as much music. And if they're wrong, the music business dies.
This depends on what the fingerprinting technology is. Audible Magic's technology, for example, can supposedly generate fingerprints for audio files that are independent of specific encoding. I don't know whether they'd end up with a single fingerprint for all versions of a track, or a couple, but in either case the number of fingerprints would be much smaller than the number of hash values.
:-)
Also, the system doesn't need to block every single copyrighted song to be effective, because not all songs are equally popular. There are perhaps 1m copyrighted tracks,
There are perhaps 1m copyrighted tracks. That's probably still too many to block. Luckily, tracks are not all equally popular, so if you block the most popular tracks, that would be disproportionately effective. Perhaps the Rolling Stone 500 Greatest Albums of All Time could be a start. That's only, say, 10,000 tracks.
How hard could it be to transmit 10,000 numeric fingerprints, and block sharing of any files that match them? P2P networks are great at transmitting data, so it'd be easy to update the list.
"Did common sense go on holidays?
:-)
Load a fingerprinted file.
Change one bit.
It has a new fingerprint."
Right, and all of the swarmed download technologies go right out the window. And the ability to tell which files have multiple sources. And Bitzi.
So I'd say that yes, common snse went on holidays.
This also misses the point of the discussion. The p2p networks' legal defense has been that they have no knowledge or control over what's going on with the network. Someone just demonstrated that a pretty obvious technical approach can, in fact, control what's moving around the p2p networks by blocking the fingerprints of copyrighted material. So the fact that they don't do this isn't because it's impossible, but because they choose not to do so.
This is important because the p2p networks' defense has been that they're innocent, and that the music company's only recourse is to sue the evil, law-breaking users. If the courts rule that the p2p networks are responsible for not blocking copyrighted materials, the music companies can stop suing individuals...
"So? How did they determine that those IPs should be blocked, and more importantly, that the rest shouldn't be?"
IP blocking wouldn't affect this survey, because the data was collected by (1) running software (with permission) on panel member's computers that watches what they do, and (2) by traditional paper market surveys.
IP blocking would only block external monitoring of the computer. That might effect other surveys, as far as users use IP blocking software...
So many errors, so little time...
"a lot of problems that *would* require multiple machines can run on one machine simply by using C++ instead of Java"
According to benchmarks, even old versions of Java are within a factor of 2 in performance with C++. See Java Performance Comparison with C/C++.
"Sun is big about "scalability" aspects of it. The problem is that this only applies to problems that can't work on one computer, but *can* work on one computer times some constant that's relatively small from a computer science standpoint."
Why? A well written distributed application can scale to thousands of CPU's. So if you have a big problem to solve (weather modeling, Computational Fluid Dynamics, high volume transaction processing, molecular modeling, rendering movies, etc.), you can either solve it by waiting 15 years for CPU's to get 1,024 times faster, or by clustering 1,024 CPU's. So I agree with your point at as CPU's get faster, more and more problems can be solved on a single CPU, but I think that you're underestimating the number of problems that require lots of computational power to solve.
You don't think that eBay, or a bank, or an airline runs on 1 CPU, do you?
"The only benefits I see to IBM's approach is that companies can theoretically replace their local Windows boxes with thin clients and fire their local IT staff"
This isn't about the kinds of applications that you run on desktops! This is about being able to build virtual mainframes out of commodity "desktop" components.
Well, Thinking Machines got bought out (in effect) when parallel computing made the transition from a "niche" to a "mainstream" technology -- once IBM, Sun, etc., all started selling MPP supercomputers, most of the specialized MPP companies went under or were bought by mainstream computer companies. In Thinking Machine's case, my recollection is that Oracle bought the Data Mining team, and Sun bought the compiler and OS team, and Danny Hillis went on to run a really cool group at Disney and to design a really cool clock. So Thinking Machine's technology and architecture won (virtually all of the world's fastest supercomputers are MPP), the company didn't survive the transition. Shame -- the people were really cool, and the food was amazing.
"Or the hardest problem of all: how often does an organization need CPU power beyond that of a typical modern desktop machine?"
This isn't a technique that applies to word processing. Think about enterprise class problems such as data mining. When you're American Express, analyzing terabytes of purchasing data to identify customer behavior models (which they do, and make a fortune on), you don't do that on a desktop computer. When you're an oil company, processing seismic survey data, you don't do that on a desktop computer. When you're computing payrolls for 20,000 employees you don't do that on a desktop computer. You get the idea.
So while desktop computers are wonderful and astoundingly fast, there are important business problems that can't be solved in a reasonable period of time on one. That's why people make servers, mainframes, etc. All this is, is an attempt to use clever software to make distributed computing as manageable as mainframes are, while taking advantage of the astounding better price/performance that you get for using commodity components instead of niche hardware.