...I feel that just because the record industry wields enough power to get artists to sign such lousy deals... One [solution] is a legal limit to what a contract can say. Any contract that puts such limits on what individuals can do, (such as go to another record company and get a better deal), should be illegal.
You may be forgetting something -- the reason the record industry has so much power is because distribution with physical media is difficult, and economies of scale pay off well enough to support these behemouths. Online, however, distribution by an individual band on their own is no problem; and they can also make quality recordings much more easily nowadays too, with the appropriate hardware/software. Only thing left is concert tours, but a band that's popular enough could probably foot their own bills there too.
About contracts: by definition a contract is something that limits what the two parties can do. The musician can't just go to another record company, and the record company can't just drop them either. If you make such a fundamental aspect of contracts illegal, you're basically making contracts illegal in general. Which would not be good -- contracts serve a vital purpose; just try buying a house, and you'll see how important the contract is.
Book stores are surviving in the face of amazon.com and it's ilk because books are still best enjoyed as very tangible things. Maybe small music stores can hang on by doing some of the things the book stores have done to make themselves a "scene."
One reason a local bookshop can have a "scene" is the fact that you can open up a book and glance at it right there. If all the books were shrink-wrapped, with no "demos" open for you to look at, could a bookstore really work at all?
That's one of the things that has always annoyed the hell out of me about music stores -- I really have no idea what I'm buying, unless I happen to know the song/group already. Here and there I've tried those headphone setups, but usually the selection is so limited, it's not interesting. That's why an on-line music store, w/ easily-accessible samples for all songs, has a fundamental advantage.
(BTW, I think they should have a lower-quality-but-full-length type of sample too; could even release something like that onto Kazza, as free advertising!).
So I'm not sure if a bricks-and-mortar music store can have a "scene" the way a bookstore can -- it's expensive to have enough playing equipment and opened cd's around to really be able to "browse". I've actually been in a store like that once or twice -- it's cumbersome, and just too expensive to actually buy a cd there.
Perhaps, though, a music store could be re-cast as a "compilation service" store. They'd have a bunch of pc's instead of CD racks, and all the equipment needed to burn song collections, print cover art, etc. And maybe their own databases of songs? Actually, that'd be a lot like a book store -- it'll end up being very quiet, because everyone's trying to listen to their own prospective purchases (w/ headphones or those isolation-dome things)!
... SCO winds up "losing" its entire base of source code to the public domain.
What makes you think that would happen?
Isn't it the case that something that incorporates GPL'd code then falls under the GPL license, and so must be released into the public domain? Or have I got that wrong?
As already stated on LKML here [iu.edu], it's far more likely that they saw something that had been stolen from Linux or other GNU code in SCO, and thought is was the other way around.
I wonder, then, if someone could turn around and sue SCO for infringement of the GPL -- it would at least force them to reveal the code to the plaintiff, who could then make a determination as to whether that's the case.
If it turns out they did put GPL'd code into their proprietary code... now *that* would be justice: in sueing for copyright infringement, SCO winds up "losing" its entire base of source code to the public domain.
all we'd have to do is start quoting our speeds in furlongs per fortnight... our speed numbers would again exceed theirs, and their coping mechanism would be shattered.
Ahh, but then they'd just start using millifurlongs per fortnight.
By contributing to Linux IBM is trying to increase their hardware sales by essentially selling software at a loss (they pay programmers to develop code that they give away). Microsoft will lose business as a result and should be able to sue IBM for antitrust violations since this is illegal.
Naaa, to be violating antitrust, you first need to have the monopoly. IBM hardly has one (these days).
This is not about 10 lines of code, it's about 20 years of extremely valuable intellectual property we're trying to protect...
You know, if you're a homeowner, and you let your neigbor use a piece of your land for a long enough period of time (e.g. for a part of his driveway), he can go to the city and claim ownership of that piece.
Why is this any different? I mean, this is open-source stuff, right? They theoretically should have been able to see the patent "violation" for many years now; if they haven't acted upon it until very recently, shouldn't it be considered public domain? How can they even have a case?
I titled this case "HEMOPHILIA" due to the red of the glass, my many wounds from working on this case, and the fact that I have hemophilia (a rare genetic blood disorder that hinders my blood from clotting.)
My god, the man's willing to risk death for his art!
Someone should rent a heliocopter, get some big-ass speakers, fly up to the windows of one of these ad agencies, and blast some car-commercial ads or something into their windows at 100db.
The article mentions a few examples of different ways to try to shield against spam, but that doesn't help the ISP's who still have to deal with the mails coming in (and from angry customers who had legitimate email filtered out when the ISP did its own filtering). Plus all the time investment one must do to try to find a good filter, and keep updating it.
Trying to attack the spammers themselves is futile. They'll just change ISP's, accounts, use foreign relays, etc. Also there is a growing trend of innocent victims whose email addr has been put as the "Reply-to" field in the spam emails. Passing laws and other "infrastructure" changes also are a threat to freedoms of speech and general internet use.
No, what has to be done is to go after those who sponser the spams in question. They cannot hide by using false addresses -- if they did, what would be the point of paying the spammers to send email?
The article has it wrong, I think, in saying that we cannot concentrate on the commercial aspects of spam. The only reason it is so pervasive is that the spammers get paid to send their spam. Your average "Jesus-saves" freak is not going to be spending the money to do that. Even if there are some who do it, they aren't businesses; the level of spam that they fund will be below the level at which a spammer can stay in business.
So... how do we go after the spam-sponsors? A coordinated DDOS attack could get one thrown in jail, because it's a knowing attack on the accessibility of a legitimate business. But there's one thing about spam -- what the article called the defining characteristic -- in that it's sent to hundreds of thousands, or millions, of addresses at a time. The spam itself provides the coordination for an attack; no individual attacker has to do more than access the web site a few times. Or enter bogus credit-card orders*. Or anything else that the web-site owner will have to pay for if done in enough volume.
If we all just made it a habit -- receive a spam, go to the web site referenced, and click around a bit, enter in your credit card # with the wrong address a few times, etc, then there would be very real costs mounting, which would quickly overshadow the increased sales that the 0.05% spam respondents would bring. It doesn't even have to be a majority of spam recipients doing this, as the article points out is a drawback of spam-shielding techniques; it only has to be enough to do the required damage to the spam sponsors.
* Just a random credit card # won't work, apparently. There's some kind of checksum technique or something. Perhaps a good way would be to use your own card #, but enter a false name/address. Basically, what needs to happen is for their ordering system to be forced to actually query the bank to make sure it's a valid credit-card account. Each time it comes back rejected, or if too many come back rejected, the bank starts charging them extra.
But if I'm still doing things at a command line in 20 more years, that will just plain suck.
If the command line is the same as what we have today, absolutely. But there is something about a command line interface, I think, that is fundamental to the human brain. After all, what is our primary means of interacting with other humans? A linear sequence of symbols -- also known as speech. Phonemes serve as the symbols, in this case. Our brains are *highly* optimized for this -- we can memorize an amazingly large number of arbitrary symbol strings, deal with all sorts of wierd rules of how to put those strings together, etc.
So there's a real reason that the command line hangs on, I think. The main problem with it is that it's largely context-free. But the more modern shells
have made progress in this area -- look at tcsh. You can hit the tab key to complete a file or command name, or show a list of the completions available. And in vim, you can hit ctrl-P or ctrl-N to complete a word you're typing, by searching up or down the document for another word that starts with the same letters that you've just typed. It's amazing how useful this is when you get used to having this ability. And why are these things useful? They make use of context.
Think about the voice-interface computers on Star Trek. They are extremely sensitive to context. That's how it knows what the user means when he/she gives a command. But other than that, it's really just a command line interface, only spoken instead of typed.
And typing isn't going to go away either. For one thing, it'd be as annoying as hell if everyone was using a voice interface at my company -- the chatter would be incessant! You know, on Star Trek, there was never more than one person in the room talking to the computer at a time. On the bridge, what were all those officers around the edges of the room doing? They were typing, essentially -- if they were all using voice interface, they'd interfere with each other, and even with the captain's ability to think and make decisions.
As for using a keyboard vs.a GUI, one thing people seem to forget is the fact that a keyboard is a highly parallel interface, whereas the mouse is quite serial -- you only have one mouse pointer on the screen at a time. A lot of the time, you have to position that pointer very carefully (especially on windows), and take care to not move it when you click the button, etc. May not seem like a big deal, but try it with a messed up hand or carpal tunnel. With that, you can really feel how very serialized that interface is. It's so very slow.
With a keyboard, you can reach out and press any button you wish, even multiple buttons at once, in some cases. True, many programs using keyboard interface force you into serialised mode, but then again, you have ctrl keys and F keys and the like -- which can cause the program to jump into doing something completely different. I'd say that games make good use of parallelism on the keyboard too.
So what is my ideal interface? I'm not really sure. A touch screen with a virtual layout of buttons that could change with context (and some kind of tactile feedback for button pushes, and ability to press more than one thing at at time)? Might work, but then again, there's something inherently useful about having a static, physical set of buttons -- our motor neurons work best when they're highly trained for specific tasks. If you had a virtual keyboard that kept changing, it'd be more difficult to be quite as fast with it. You'd basically have to be hunting-and-pecking all the time. You'd be in a tight loop -- looking at the screen for the button to press, then pressing it, then watching to confirm the effects so you can tell where the next button will be to press. That's a very linearizing process, isn't it? Might as well be using a command line.
Ok, I'm done ranting. By the way, here's a reference to that "In the Beginning was the Command Line" article by Neal Stephenson (author of Cryptonomicon). It contains an interesting perspective on the history of the command-line interface (among other things).
Linux itself is a clone of an operating system that is 20-plus years old. That's what it is. That is what you can get today, a clone of a 20-year-old system.
That's what I'm always thinking about when I'm flying on an airplane -- damnit, this is such an old design! 100 years old! I'd really rather be trusting my life to something much more new and innovative!
More staff will be required to process the fake orders/replies, and they'll have a devil of a time weeking out the true from the false responses.
Hmmm, you know a lot of them have on-line ordering, including a step where they validate your credit card number. I wonder, what's the effect of an invalid card number being entered? Obviously they have to check w/ the card company to see if it's valid. Is there a cost associated with that? i.e. does the company have to pay 0.01 cents or something per query?
If so, that could be a way to directly increase the cost of sending spam (or of hiring a "marketing firm" to send spam). You go to the spam-sponsor's web site, fill out the order with bogus information, including dummy credit card info, then repeatedly hit "send". Could a script could be set up to do this? Hmmm.
You're forgetting another key advantage to the seller - rebates generate float. Put simply, they hold your money in interest-bearing investments until the point, 6-8 weeks after the purchase, at which you receive and cash the rebate check.
One other thing I've thought of, too, is that it allows the company discretion as to which fiscal quarter they're going to take the loss in. Think you're going to have better profits next quarter? Push the rebates into April! And such.
Evolution up to know, has proceeded slowly, about one bit of DNA changes every year.
I'm of the opinion that evolution has largely halted for us humans -- at least, those with access to modern medical care. Certainly, most genetic anomalies will result in problems we can't fix, but certainly a greater number of humans who wouldn't have survived otherwise are geting all the way to the reproductive stage. And the set of survivable genetic problems is surely growing all the time.
Not that that's all a bad thing, mind you -- variety is good. But it does mean that the "undesireable" genes are in fact spreading through the population to a greater degree than is "natural". Genetic engineering may in fact turn out to be a necessary tool to prevent the majority of the population from needing advanced medical treatments just to survive.
NASA has come up with a way to wire microchips with nanotubes instead of copper interconnects.
If you read the article closely, you'll see it's not talking about about replacing all copper interconnect on the chip -- only a small portion, in fact: the vias. The carbon nanotube are being used only for the interconnect between metal layers, not between devices on the chip in general.
What NASA has done is to make a switch from copper connectors to carbon nanotube connectors within the chip and (maybe) the boards.
Not quite: it's not the *wire routing* that is being implemented with nanotubes here. Rather, as indicated rather subtly in the article, it is only *vias* -- i.e. the interconnect between layers of metal on the chip. It's a crucial piece of the puzzle, but still limited in scope. The limitations of the copper wires will still dominate.
Goddamnit man, what's the deal with this stuff gettin' posted over and over. I mean, man, what the fuchk. Friggin fraggin, flaavin, floopin sloothpthhtin, sthhhpptling, slttthhpppptl... oh, goddamnit, look at all this drool over the place. Damnit taco, now look at what you've done to me. Gotta clean up my keyboard before my boss comes and sees me reading your doggam replicated posts all day long, over and over and over and ARgg arrrgg argg arff!
As I read through the postings here, I am increasingly convinced it is a joke. The only people to say they've seen corroborating sources all say that it was in Discover. Which, as has been noted many times, has a history of April Fool's jokes. That page that said the last one was in '98 may even be part of it -- or out of date.:)
Furthermore, that term... "thermal depolymerization"... isn't that another term for "burning"??
Another line of thought: A good fraction of crude oil is benzyne, right? That's what becomes gasoline, anyways. If the process is only one of depolymerization, that would imply that the benzene molecules already exist in the turkey guts, etc, and this process just breaks them out from larger polymer chains. Considering benzene is considered a carcinogen, I don't think it would naturally occur in meat products in any great abundance.
Well, it'd be cool if this was real, but right now I'm leaning more towards the "unreal" side. I mean, you'd think there'd be quite a few articles talking about it, including Scientific American, major newspapers, etc. Especially if that article was really all the way back from Dec. 4.
Slashdot Mirror
One [solution] is a legal limit to what a contract can say. Any contract that puts such limits on what individuals can do, (such as go to another record company and get a better deal), should be illegal.
You may be forgetting something -- the reason the record industry has so much power is because distribution with physical media is difficult, and economies of scale pay off well enough to support these behemouths. Online, however, distribution by an individual band on their own is no problem; and they can also make quality recordings much more easily nowadays too, with the appropriate hardware/software. Only thing left is concert tours, but a band that's popular enough could probably foot their own bills there too.
About contracts: by definition a contract is something that limits what the two parties can do. The musician can't just go to another record company, and the record company can't just drop them either. If you make such a fundamental aspect of contracts illegal, you're basically making contracts illegal in general. Which would not be good -- contracts serve a vital purpose; just try buying a house, and you'll see how important the contract is.
One reason a local bookshop can have a "scene" is the fact that you can open up a book and glance at it right there. If all the books were shrink-wrapped, with no "demos" open for you to look at, could a bookstore really work at all?
That's one of the things that has always annoyed the hell out of me about music stores -- I really have no idea what I'm buying, unless I happen to know the song/group already. Here and there I've tried those headphone setups, but usually the selection is so limited, it's not interesting. That's why an on-line music store, w/ easily-accessible samples for all songs, has a fundamental advantage. (BTW, I think they should have a lower-quality-but-full-length type of sample too; could even release something like that onto Kazza, as free advertising!).
So I'm not sure if a bricks-and-mortar music store can have a "scene" the way a bookstore can -- it's expensive to have enough playing equipment and opened cd's around to really be able to "browse". I've actually been in a store like that once or twice -- it's cumbersome, and just too expensive to actually buy a cd there.
Perhaps, though, a music store could be re-cast as a "compilation service" store. They'd have a bunch of pc's instead of CD racks, and all the equipment needed to burn song collections, print cover art, etc. And maybe their own databases of songs? Actually, that'd be a lot like a book store -- it'll end up being very quiet, because everyone's trying to listen to their own prospective purchases (w/ headphones or those isolation-dome things)!
Isn't it the case that something that incorporates GPL'd code then falls under the GPL license, and so must be released into the public domain? Or have I got that wrong?
I wonder, then, if someone could turn around and sue SCO for infringement of the GPL -- it would at least force them to reveal the code to the plaintiff, who could then make a determination as to whether that's the case.
If it turns out they did put GPL'd code into their proprietary code... now *that* would be justice: in sueing for copyright infringement, SCO winds up "losing" its entire base of source code to the public domain.
Ahh, but then they'd just start using millifurlongs per fortnight.
I haven't seen this one yet. What's is stand for? I have some guesses, but they don't seem right:
Naaa, to be violating antitrust, you first need to have the monopoly. IBM hardly has one (these days).
You know, if you're a homeowner, and you let your neigbor use a piece of your land for a long enough period of time (e.g. for a part of his driveway), he can go to the city and claim ownership of that piece.
Why is this any different? I mean, this is open-source stuff, right? They theoretically should have been able to see the patent "violation" for many years now; if they haven't acted upon it until very recently, shouldn't it be considered public domain? How can they even have a case?
[shatner]
My...god...it's.....
[/shatner]
Interesting:
I titled this case "HEMOPHILIA" due to the red of the glass, my many wounds from working on this case, and the fact that I have hemophilia (a rare genetic blood disorder that hinders my blood from clotting.)
My god, the man's willing to risk death for his art!
Share with them the user experience.
Trying to attack the spammers themselves is futile. They'll just change ISP's, accounts, use foreign relays, etc. Also there is a growing trend of innocent victims whose email addr has been put as the "Reply-to" field in the spam emails. Passing laws and other "infrastructure" changes also are a threat to freedoms of speech and general internet use.
No, what has to be done is to go after those who sponser the spams in question. They cannot hide by using false addresses -- if they did, what would be the point of paying the spammers to send email?
The article has it wrong, I think, in saying that we cannot concentrate on the commercial aspects of spam. The only reason it is so pervasive is that the spammers get paid to send their spam. Your average "Jesus-saves" freak is not going to be spending the money to do that. Even if there are some who do it, they aren't businesses; the level of spam that they fund will be below the level at which a spammer can stay in business.
So... how do we go after the spam-sponsors? A coordinated DDOS attack could get one thrown in jail, because it's a knowing attack on the accessibility of a legitimate business. But there's one thing about spam -- what the article called the defining characteristic -- in that it's sent to hundreds of thousands, or millions, of addresses at a time. The spam itself provides the coordination for an attack; no individual attacker has to do more than access the web site a few times. Or enter bogus credit-card orders*. Or anything else that the web-site owner will have to pay for if done in enough volume.
If we all just made it a habit -- receive a spam, go to the web site referenced, and click around a bit, enter in your credit card # with the wrong address a few times, etc, then there would be very real costs mounting, which would quickly overshadow the increased sales that the 0.05% spam respondents would bring. It doesn't even have to be a majority of spam recipients doing this, as the article points out is a drawback of spam-shielding techniques; it only has to be enough to do the required damage to the spam sponsors.
If the command line is the same as what we have today, absolutely. But there is something about a command line interface, I think, that is fundamental to the human brain. After all, what is our primary means of interacting with other humans? A linear sequence of symbols -- also known as speech. Phonemes serve as the symbols, in this case. Our brains are *highly* optimized for this -- we can memorize an amazingly large number of arbitrary symbol strings, deal with all sorts of wierd rules of how to put those strings together, etc.
So there's a real reason that the command line hangs on, I think. The main problem with it is that it's largely context-free. But the more modern shells have made progress in this area -- look at tcsh. You can hit the tab key to complete a file or command name, or show a list of the completions available. And in vim, you can hit ctrl-P or ctrl-N to complete a word you're typing, by searching up or down the document for another word that starts with the same letters that you've just typed. It's amazing how useful this is when you get used to having this ability. And why are these things useful? They make use of context.
Think about the voice-interface computers on Star Trek. They are extremely sensitive to context. That's how it knows what the user means when he/she gives a command. But other than that, it's really just a command line interface, only spoken instead of typed.
And typing isn't going to go away either. For one thing, it'd be as annoying as hell if everyone was using a voice interface at my company -- the chatter would be incessant! You know, on Star Trek, there was never more than one person in the room talking to the computer at a time. On the bridge, what were all those officers around the edges of the room doing? They were typing, essentially -- if they were all using voice interface, they'd interfere with each other, and even with the captain's ability to think and make decisions.
As for using a keyboard vs.a GUI, one thing people seem to forget is the fact that a keyboard is a highly parallel interface, whereas the mouse is quite serial -- you only have one mouse pointer on the screen at a time. A lot of the time, you have to position that pointer very carefully (especially on windows), and take care to not move it when you click the button, etc. May not seem like a big deal, but try it with a messed up hand or carpal tunnel. With that, you can really feel how very serialized that interface is. It's so very slow.
With a keyboard, you can reach out and press any button you wish, even multiple buttons at once, in some cases. True, many programs using keyboard interface force you into serialised mode, but then again, you have ctrl keys and F keys and the like -- which can cause the program to jump into doing something completely different. I'd say that games make good use of parallelism on the keyboard too.
So what is my ideal interface? I'm not really sure. A touch screen with a virtual layout of buttons that could change with context (and some kind of tactile feedback for button pushes, and ability to press more than one thing at at time)? Might work, but then again, there's something inherently useful about having a static, physical set of buttons -- our motor neurons work best when they're highly trained for specific tasks. If you had a virtual keyboard that kept changing, it'd be more difficult to be quite as fast with it. You'd basically have to be hunting-and-pecking all the time. You'd be in a tight loop -- looking at the screen for the button to press, then pressing it, then watching to confirm the effects so you can tell where the next button will be to press. That's a very linearizing process, isn't it? Might as well be using a command line.
Ok, I'm done ranting. By the way, here's a reference to that "In the Beginning was the Command Line" article by Neal Stephenson (author of Cryptonomicon). It contains an interesting perspective on the history of the command-line interface (among other things).
That's what I'm always thinking about when I'm flying on an airplane -- damnit, this is such an old design! 100 years old! I'd really rather be trusting my life to something much more new and innovative!
Hmmm, you know a lot of them have on-line ordering, including a step where they validate your credit card number. I wonder, what's the effect of an invalid card number being entered? Obviously they have to check w/ the card company to see if it's valid. Is there a cost associated with that? i.e. does the company have to pay 0.01 cents or something per query?
If so, that could be a way to directly increase the cost of sending spam (or of hiring a "marketing firm" to send spam). You go to the spam-sponsor's web site, fill out the order with bogus information, including dummy credit card info, then repeatedly hit "send". Could a script could be set up to do this? Hmmm.
Hmmm.
Scenario #1: "Oh, F***! I dropped my TiVo 3 feet onto a tile floor! All my recordings are GONE!"
Scenario #2: To next-door neighbor: "Hey, check out this cool show I recorded last night!" *FLING*
One other thing I've thought of, too, is that it allows the company discretion as to which fiscal quarter they're going to take the loss in. Think you're going to have better profits next quarter? Push the rebates into April! And such.
I'm of the opinion that evolution has largely halted for us humans -- at least, those with access to modern medical care. Certainly, most genetic anomalies will result in problems we can't fix, but certainly a greater number of humans who wouldn't have survived otherwise are geting all the way to the reproductive stage. And the set of survivable genetic problems is surely growing all the time.
Not that that's all a bad thing, mind you -- variety is good. But it does mean that the "undesireable" genes are in fact spreading through the population to a greater degree than is "natural". Genetic engineering may in fact turn out to be a necessary tool to prevent the majority of the population from needing advanced medical treatments just to survive.
If you read the article closely, you'll see it's not talking about about replacing all copper interconnect on the chip -- only a small portion, in fact: the vias. The carbon nanotube are being used only for the interconnect between metal layers, not between devices on the chip in general.
Not quite: it's not the *wire routing* that is being implemented with nanotubes here. Rather, as indicated rather subtly in the article, it is only *vias* -- i.e. the interconnect between layers of metal on the chip. It's a crucial piece of the puzzle, but still limited in scope. The limitations of the copper wires will still dominate.
Though if you happen to really have some data on this -- I'd be interested to hear it. ;)
The Flash business is the (relative) cash cow that's keeping the rest of the company afloat. No way it's going to split.
That's a neat trick -- include an outrageous-but-believeable-to-newbies statment in your story submission, get it posted! I'll have to try that.
Wow, this really is technology of the future!
Furthermore, that term ... "thermal depolymerization" ... isn't that another term for "burning"??
Another line of thought: A good fraction of crude oil is benzyne, right? That's what becomes gasoline, anyways. If the process is only one of depolymerization, that would imply that the benzene molecules already exist in the turkey guts, etc, and this process just breaks them out from larger polymer chains. Considering benzene is considered a carcinogen, I don't think it would naturally occur in meat products in any great abundance.
Well, it'd be cool if this was real, but right now I'm leaning more towards the "unreal" side. I mean, you'd think there'd be quite a few articles talking about it, including Scientific American, major newspapers, etc. Especially if that article was really all the way back from Dec. 4.