As someone who has been a Linux newbie repeatedly over the years, by virtue of periodically installing some distro, trying to get it to work, getting it 90% there and then getting hung up on some random bit of hardware and eventually getting frustrated and going back to my Mac, the lack of coherent established opinions is difficult to work with.
In my longest (and still running) experiment with Linux, on a home server, I went with Debian because I thought it would avoid RPM dependency hell that drove me nuts in previous tries. In retrospect, I think this was a good move -- I'll never use anything that doesn't have apt-get again. However I'll often ask a question on how to set something up or edit a config file, and get the response "Well, that's just because Debian sucks and is broken. Use [insert distribution here] instead."
It's fine to have multiple distributions. It's fine for people to have opinions on which distribution is best. But advocating that others switch distributions constantly in response to what ought to be minor problems doesn't do a lot to inspire confidence by new users in an OS.
Also, I think a lot of users go too far the other way -- so on one hand you have distribution zealots who loudly proclaim that theirs is better and yours clearly sucks, for any reason or none at all, while on the other the people who actually have a soapbox to stand on (in trade magazines, established web sites, etc.) generally refuse to take an opinion on distros one way or another. Once in a while I'll see someone take a wishy-washy stance as to 'this distribution might be a good one' but there's very little clear guidance for new users. If you have an opinion based on real experience, for god's sake say it. But if you just like your distribution because it's yours, shut up.
Sorry -- disregard. I reread the parent posts and realized they were talking about a different issue. They are talking about 47kohm line level inputs, I missed the "k" and thought they were talking about the relative impedances of various headphones, both in the sub-100-ohm range.
I was under the impression that most modern audio devices have buffered outputs, using something similar to a voltage follower, so that their output was virtually unaffected by a small change in load impedance (e.g. 32 vs 47 ohms).
I could be wrong, but my understanding was that the 47-ohm impedance mismatch was, in the case of most modern well designed audio devices, a dead issue.
I don't have any mod points, but I wanted to thank you for your post, it was an interesting read all around.
In particular I wanted to underscore the point you made about what happens if you play a square wave through a speaker system and then monitor the actual acoustic output through a good measurement microphone into an oscilloscope. I have done exactly this, and you would be stunned at how much distortion exists even in some "audiophile" speakers. However, that doesn't make them bad speakers -- on the contrary, they sound great when listening to anything except a square wave.
My point is that square wave testing, and modifications of which the aim is an improvement in the performance of a given device under square wave tests, have limited usefulness to someone who actually enjoys listening to music (as opposed to square waves). I'm sure you realize this, but I just wanted to make the point again.
The best advice I've ever heard with regards to sound equipment optimization was the mantra of a sound engineer I used to work with: "If it sounds good, it is good." Objective tests are great at telling you things like whether your equipment is functioning properly or at peak performance, or giving you hard metrics by which to compare two distant systems, but at the end of the day are irrelevant -- or at best secondary to the very subjective test of the output's effect on your ears, brain, and mind.
I can do better than that. I have a old Apple external 20MB hard drive which interfaces via the floppy disk connector.
It's made to be the same footprint as the Mac Plus, and you can even daisy-chain your external 800k floppy disk drive to it.
I keep it around because the sound it makes while it powers up is just priceless. It sounds vaguely like a large turbine engine spooling up, with some additional clicks and beeps. One of these days I'll get a microphone near it and record it. It's enough to make most people take a step back and look at it suspiciously.
The downside of computer equipment being so quiet these days is you don't get nearly the range of entertaining and impressive sound effects that you used to. Any piece of equipment which sounds as though it should be accompanied by the smell of burning aviation kerosene and a man with ear protectors and an orange reflector vest gesturing with traffic-control wands has a place on my desk.
Huh? You're not making any sense. There are too commercial solutions that support one million accounts (which means a lot more than one million emails), because there are corporations which sell these products and use them, which by themselves have more than one million accounts. Witness IBM, and AT&T.
I would bet that Yahoo and Google built their own systems not because there wasn't a commercial product out there that would scale to the size they wanted, but wasn't a commercial product that would scale to the size they wanted, at the price they were willing to pay. They're high-tech, in particular internet, companies. So the "build it or buy it" equation looks very different for them than it would for, say, the Navy. Or a big consulting firm. Or Wal-Mart. Or frankly, for anyone who's business ISN'T servers and software.
Just because Yahoo and Google went the DIY route doesn't mean that anyone else should; it only means that if you somehow ended up in a position identical to theirs, that it might be worth considering. But if servers and massive distributed systems aren't your company's (in managerese) "core competency," then you ought to go hire someone for whom it is.
I can't tell if this is just sarcastic understatement, or if you actually think that "a single xeon" server (or two of them, with one as a redundant backup) can handle a million corporate email accounts. That's ridiculous.
I'd wager that the whole thing would fail the first morning you had a tenth of the employees in the corporation all attempt to log in and check their email simultaneously. When he means 1M accounts, in a corporate situation, it's not like Hotmail where 90% of the accounts are just idle and never see any activity. The usage is going to be extreme at some points and nearly zero at others, but users are going to expect the same (excellent) performance constantly.
The requirements plainly dictate a distributed, highly redundant system with multiple levels of soft fallbacks. It's a major project, not something you're going to run out of your basement on a dual T1.
Um.... I think it's worth pointing out that eBay doesn't have a "core competency" in maintaining " VERY large number of users, especially in near-realtime transactions." Actually, they kinda sucked at it. Their original architecture didn't scale well. Which was part of why they hired an outside contractor to rebuild their system.
If anyone ought to get credit for that being their 'core competency,' it's IBM and Sun, who actually did the work.
The only time I've ever used Skype out was when I'm traveling internationally. I didn't have a GSM cellphone at the time that I could just pop a new SIM into, and U.S. cellular carriers' rates are atrocious abroad. So I would just stop into an internet cafe, login to my Skype account (downloading it if necessary), connect a cheap headset, and off I go.
A few of the cheaper youth hostels that I visited had actually replaced their phones with cheap PCs and Skype VoIP handsets. It was a great idea, except that with the rates being so much cheaper, the lines were atrocious.
Not just on Slashdot, but to the world in general. Consider how the two companies investors apparently felt about the deals: when Google announced its IM client, its stock went up and is still going up. eBay announces the possibility of it buying Skype, and its stock takes a nosedive.
On one hand, the innovator, on the other, the behemoth. Investors are not completely stupid, they can see the writing on the wall as easily as anyone else.
If what you say is true, then this is truly depressing news for manned space flight in general, American in particular.
It makes me wonder whether we wouldn't just be better taking NASA out behind the shed and getting rid of it completely, in the hopes that a fraction of its budget might get reallocated to private-sector projects, than continue down the current path which (again, if what you said is true) offers nothing but wasted money and false hopes created and broken for the American people.
As to the people who have commented that private sector space flight only puts rich people in space, who cares? You have less chance of becoming an astronaut now than you would under a $1M a head ticket sale strategy, why whine about it? If it produces equivalent or better scientific and engineering gains, and gets people interested in space, I'm all for it. The "average person" is never going to get a ride, either under the government-run or private sector plans. Whether the few who get to go are selected by some byzantine bureaucratic process or straight-out cash on the barrelhead, it's irrelevant and won't make people any less enthuasiastic about space. How many young people have you seen that have a poster of a McLaren F1 in their room, even though realistically they'll never be able to own one? (Or at this rate, even the gas to run it?) That very few people WILL be able to afford spaceflight doesn't detract from the fact that it is available, just as many people thought the Concorde was cool, but never could have flown on it themselves.
As I've thought about it more, the way to use patents in the FOSS world wouldn't be as an 'offensive' weapon, something that you'd use against a commercial company just because they weren't liked, but as a deterrent.
It's the reverse of the above scenario. The commercial company says that some piece of Free Software violates its patents. The FS people can come back and show that the commercial software is also violating a number of their patents.
The result is a sort of 'mexican standoff.'
By not striking first, you save a lot of funds -- since the FOSS community would only have to respond to the occasional threat when a commercial company came after an open source product for infringement, instead of actively litigating every patent in the positive, aggressive sense (as most corporations do).
The presence of the patents themselves, plus enough of a financial 'war chest' (which could be in promised services and not just money -- e.g. if a law firm promised to help with any litigation that might occur) to respond to an attack, would have a deterrent effect. It wouldn't really be much of a weapon against commercial software or infringers, but it would at least guarantee the status quo.
Umm...perhaps I'm misunderstanding, but this is happening on computers owned by the city, and used by city employees. Key word: employees.
Not citizens, not townspeople. Government employees, on their work computers. These are the 'end users' in this discussion, not the actual recipients of the government's services.
Nobody is saying that the customers (in this case, the citizens who get their services from the government) have to change anything. In fact it seems as though one of the main objectives of the migration is to be transparent from a customer perspective. People doing their automobile registrations, for instance, will still see the same web pages, etc.
This is purely an internal change, by a government office. The way you wrote your comment, you'd think that the Free Software gestapo were going to go kicking down people's doors and forcing them to install Debian at home. That's just not true, and while the employees should certainly have a say in anything that effects them (and probably have a much greater say than typical employees in either the public or private sector in the U.S.), they are at the end of the day employees, and free to leave if they don't like the direction their workplace is headed.
The key words are "being depreciated." That doesn't change anything about the installed user-base who depend on VBA applications or tools. It means they would have had to change anyway in the near future, but that's irrelevant now that the move to Linux has been announced.
Now, all the effort at replacing them will be blamed on the OS switch, as opposed to just being written off as a necessary upgrade in order to use the new framework and remain competitive. And of course the end users will still bitch and moan the same either way, now it'll just be directed at Linux instead of Microsoft.
Don't get me wrong, I'm all for the switch, I think Munich should be applauded. However everyone has to realize that when you make any kind of big change or migration, it instantly becomes world's largest scapegoat. Any problem, real or imagined, regardless of whether it would have happened under Windows anyway, will be its fault.
Oops -- in my editing I cut a chunk out:... and then assigning [the patents] to the FOSS organization 'closest' to the project who would have the resources to employ them so they could be used against GPL violators.
Maybe the solution to the GPL software patents "problem" is for GPL software authors to begin aggressively patenting their work?
I mean, there are certainly lots of things that exist in the various flavors of the Linux operating system that could be patentable (certainly by the low standards the Patent Office here in the U.S. seems to be using). Rather than just scream and yell about how evil software patents are, maybe as a community we should start actively and encouraging authors to patent new concepts, and then assigning them to the FOSS so they could be used against GPL violators.
It seems that perhaps this is the direction that the GPL 2 drafters have in mind?
Software patents are looking more and more to be a powerful new weapon between the various entities competing for mind- and market-share in the tech world. It would be irresponsible for the FOSS community not to attempt to arm itself, as a deterrent if nothing else.
Re:High Resolution Computer Graphics and Broadband
on
Pornified
·
· Score: 1
I think broadband was driven by Napster and the MP3 revolution, more than anything else. I can't speak for S. Korea -- maybe online network games drove it there. But among the admittedly biased sample group of people that I knew when broadband really exploded, most of them got it because they wanted to download free music.
Now, I think what's driving it is people who graduate from colleges or universities and have gotten used to being on a high-speed connection of some sort, and don't want to go back to dialup when they move into their first apartment. Most young graduates have the disposable income to burn, and within this group I'd be willing to bet that broadband penetration (no pun intended for an article about internet porn:) ) is very high.
Napster and its ilk were the killer apps that spurred broadband into residential areas, and it's the wide and ubiquitous presence in college dorms that will guarantee a steady flow of new young subscribers. Once you've had high speed, you'll never go back.
So you can't modify some GPLed software and release it under the "Mr. Underbridge Public License" (UPL?), but if you're a developer who just wrote something novel, you can surely apply your license to it.
But obviously you can't take something that's licensed under the GPL and "remove offending portions" that you don't like, before releasing your modifications. That would undermine the whole idea.
However if you're the copyright holder of the original code, as (I believe) Linus is of the kernel, then you could cut the GPL up to your heart's content before applying it to your work. You just can't modify the license on somebody else's work, or on a derivative you've made from their work.
You have to read the link -- the process they're using grows the algae in artificially or externally heated conditions, so there's an energy input, and also in an artificially CO2 rich atmosphere, which would come from fossil fuels. It is a significantly more efficient use of fossil fuels than our direct consumption of them today, but it is not the closed-carbon-cycle that you are alluding to.
It's a common misconception (not one you necessarily made, but I'm just using your post as a starting point) that Deep Blue just "brute forces" the entire chess game, based on the opposing player's possible responses to every move, analyzing every possible avenue to exhaustion. This is not true.
Even with all of today's computing capacity, until you got to the endgame, the power required would still be prohibitive. Instead, Deep Blue pursues every move it can make to a depth of (I've been told) 8 or 9 moves. It's fairly "intelligent" in the way it decides which avenues are fruitless and which to pursue further. And of course it does many avenues at once, in parallel, and then compares the benefits and gains of each before deciding on a move.
To pidgeon-hole it as just some sort of beefed up desktop computer running Chessmaster 3000 doesn't do the design much justice. It was a wonderful application of massively parallel computing technology, at a time when most people didn't know what 'parallel computing' was.
Yes, and like studying the birds to build an airplane, trying to jump-start a small star in your backyard might not be the way to clean, safe power generation, either.
Sometimes discoveries which seem "right around the corner" linger just out of our reach for hundreds of years, until the right person or group of people has the vision and arrives at the right time to produce something really revolutionary.
Seriously -- this is about the long and short of it.
It sounds less like they really were going to switch to Linux, than they were just playing a rather large and complicated game of chicken with their vendors / potential vendors.
FOSS is in many cases to corporate types less an actual piece of usable technology, than a big stick they can wave vaguely in the direction of their vendors and subcontractors, to keep their costs down. "Your bid is too high! We're switching to Linux, it's free!" "Okay, how about we cut another 30%." "Sold!"
I think you're substantially right, I heard the difference (on the macro-scale, anyway) is mainly in the viscosity and gelling point. On the chemical level it's just slightly different ratios of various hydrocarbons. The reason I heard that mixing Kero and #2 heating oil worked better than straight #2 or straight kero was that it makes the viscosity and flashpoint close to that of diesel. Might not matter depending on what you're going to use the fuel in.
I have actually seen a demonstration of small-scale biodiesel production from SVO, and it was interesting but struck me as being not particularly practical. It requires fairly large quantities of lye and methanol, and the lye is consumed in the process -- it is not catalytic, at least in the small-batch process recipes I've seen.
If someone could come up with a small scale continuous catalytic process, then we could really get a 'cottage industry' of bio-diesel manufacturing going. But right now the batch processes that bio-diesel enthusiasts are championing don't seem ready for prime time.
Slashdot Mirror
I just wanted to second what you're saying.
As someone who has been a Linux newbie repeatedly over the years, by virtue of periodically installing some distro, trying to get it to work, getting it 90% there and then getting hung up on some random bit of hardware and eventually getting frustrated and going back to my Mac, the lack of coherent established opinions is difficult to work with.
In my longest (and still running) experiment with Linux, on a home server, I went with Debian because I thought it would avoid RPM dependency hell that drove me nuts in previous tries. In retrospect, I think this was a good move -- I'll never use anything that doesn't have apt-get again. However I'll often ask a question on how to set something up or edit a config file, and get the response "Well, that's just because Debian sucks and is broken. Use [insert distribution here] instead."
It's fine to have multiple distributions. It's fine for people to have opinions on which distribution is best. But advocating that others switch distributions constantly in response to what ought to be minor problems doesn't do a lot to inspire confidence by new users in an OS.
Also, I think a lot of users go too far the other way -- so on one hand you have distribution zealots who loudly proclaim that theirs is better and yours clearly sucks, for any reason or none at all, while on the other the people who actually have a soapbox to stand on (in trade magazines, established web sites, etc.) generally refuse to take an opinion on distros one way or another. Once in a while I'll see someone take a wishy-washy stance as to 'this distribution might be a good one' but there's very little clear guidance for new users. If you have an opinion based on real experience, for god's sake say it. But if you just like your distribution because it's yours, shut up.
Okay, I'm done.
Sorry -- disregard. I reread the parent posts and realized they were talking about a different issue. They are talking about 47kohm line level inputs, I missed the "k" and thought they were talking about the relative impedances of various headphones, both in the sub-100-ohm range.
I was under the impression that most modern audio devices have buffered outputs, using something similar to a voltage follower, so that their output was virtually unaffected by a small change in load impedance (e.g. 32 vs 47 ohms).
I could be wrong, but my understanding was that the 47-ohm impedance mismatch was, in the case of most modern well designed audio devices, a dead issue.
I don't have any mod points, but I wanted to thank you for your post, it was an interesting read all around.
In particular I wanted to underscore the point you made about what happens if you play a square wave through a speaker system and then monitor the actual acoustic output through a good measurement microphone into an oscilloscope. I have done exactly this, and you would be stunned at how much distortion exists even in some "audiophile" speakers. However, that doesn't make them bad speakers -- on the contrary, they sound great when listening to anything except a square wave.
My point is that square wave testing, and modifications of which the aim is an improvement in the performance of a given device under square wave tests, have limited usefulness to someone who actually enjoys listening to music (as opposed to square waves). I'm sure you realize this, but I just wanted to make the point again.
The best advice I've ever heard with regards to sound equipment optimization was the mantra of a sound engineer I used to work with: "If it sounds good, it is good." Objective tests are great at telling you things like whether your equipment is functioning properly or at peak performance, or giving you hard metrics by which to compare two distant systems, but at the end of the day are irrelevant -- or at best secondary to the very subjective test of the output's effect on your ears, brain, and mind.
I can do better than that. I have a old Apple external 20MB hard drive which interfaces via the floppy disk connector.
It's made to be the same footprint as the Mac Plus, and you can even daisy-chain your external 800k floppy disk drive to it.
I keep it around because the sound it makes while it powers up is just priceless. It sounds vaguely like a large turbine engine spooling up, with some additional clicks and beeps. One of these days I'll get a microphone near it and record it. It's enough to make most people take a step back and look at it suspiciously.
The downside of computer equipment being so quiet these days is you don't get nearly the range of entertaining and impressive sound effects that you used to. Any piece of equipment which sounds as though it should be accompanied by the smell of burning aviation kerosene and a man with ear protectors and an orange reflector vest gesturing with traffic-control wands has a place on my desk.
Huh? You're not making any sense. There are too commercial solutions that support one million accounts (which means a lot more than one million emails), because there are corporations which sell these products and use them, which by themselves have more than one million accounts. Witness IBM, and AT&T.
I would bet that Yahoo and Google built their own systems not because there wasn't a commercial product out there that would scale to the size they wanted, but wasn't a commercial product that would scale to the size they wanted, at the price they were willing to pay. They're high-tech, in particular internet, companies. So the "build it or buy it" equation looks very different for them than it would for, say, the Navy. Or a big consulting firm. Or Wal-Mart. Or frankly, for anyone who's business ISN'T servers and software.
Just because Yahoo and Google went the DIY route doesn't mean that anyone else should; it only means that if you somehow ended up in a position identical to theirs, that it might be worth considering. But if servers and massive distributed systems aren't your company's (in managerese) "core competency," then you ought to go hire someone for whom it is.
Are you joking?
I can't tell if this is just sarcastic understatement, or if you actually think that "a single xeon" server (or two of them, with one as a redundant backup) can handle a million corporate email accounts. That's ridiculous.
I'd wager that the whole thing would fail the first morning you had a tenth of the employees in the corporation all attempt to log in and check their email simultaneously. When he means 1M accounts, in a corporate situation, it's not like Hotmail where 90% of the accounts are just idle and never see any activity. The usage is going to be extreme at some points and nearly zero at others, but users are going to expect the same (excellent) performance constantly.
The requirements plainly dictate a distributed, highly redundant system with multiple levels of soft fallbacks. It's a major project, not something you're going to run out of your basement on a dual T1.
Um .... I think it's worth pointing out that eBay doesn't have a "core competency" in maintaining " VERY large number of users, especially in near-realtime transactions." Actually, they kinda sucked at it. Their original architecture didn't scale well. Which was part of why they hired an outside contractor to rebuild their system.
If anyone ought to get credit for that being their 'core competency,' it's IBM and Sun, who actually did the work.
The only time I've ever used Skype out was when I'm traveling internationally. I didn't have a GSM cellphone at the time that I could just pop a new SIM into, and U.S. cellular carriers' rates are atrocious abroad. So I would just stop into an internet cafe, login to my Skype account (downloading it if necessary), connect a cheap headset, and off I go.
A few of the cheaper youth hostels that I visited had actually replaced their phones with cheap PCs and Skype VoIP handsets. It was a great idea, except that with the rates being so much cheaper, the lines were atrocious.
You're preaching to the choir, pal.
Not just on Slashdot, but to the world in general. Consider how the two companies investors apparently felt about the deals: when Google announced its IM client, its stock went up and is still going up. eBay announces the possibility of it buying Skype, and its stock takes a nosedive.
On one hand, the innovator, on the other, the behemoth. Investors are not completely stupid, they can see the writing on the wall as easily as anyone else.
If what you say is true, then this is truly depressing news for manned space flight in general, American in particular.
It makes me wonder whether we wouldn't just be better taking NASA out behind the shed and getting rid of it completely, in the hopes that a fraction of its budget might get reallocated to private-sector projects, than continue down the current path which (again, if what you said is true) offers nothing but wasted money and false hopes created and broken for the American people.
As to the people who have commented that private sector space flight only puts rich people in space, who cares? You have less chance of becoming an astronaut now than you would under a $1M a head ticket sale strategy, why whine about it? If it produces equivalent or better scientific and engineering gains, and gets people interested in space, I'm all for it. The "average person" is never going to get a ride, either under the government-run or private sector plans. Whether the few who get to go are selected by some byzantine bureaucratic process or straight-out cash on the barrelhead, it's irrelevant and won't make people any less enthuasiastic about space. How many young people have you seen that have a poster of a McLaren F1 in their room, even though realistically they'll never be able to own one? (Or at this rate, even the gas to run it?) That very few people WILL be able to afford spaceflight doesn't detract from the fact that it is available, just as many people thought the Concorde was cool, but never could have flown on it themselves.
That's the point.
As I've thought about it more, the way to use patents in the FOSS world wouldn't be as an 'offensive' weapon, something that you'd use against a commercial company just because they weren't liked, but as a deterrent.
It's the reverse of the above scenario. The commercial company says that some piece of Free Software violates its patents. The FS people can come back and show that the commercial software is also violating a number of their patents.
The result is a sort of 'mexican standoff.'
By not striking first, you save a lot of funds -- since the FOSS community would only have to respond to the occasional threat when a commercial company came after an open source product for infringement, instead of actively litigating every patent in the positive, aggressive sense (as most corporations do).
The presence of the patents themselves, plus enough of a financial 'war chest' (which could be in promised services and not just money -- e.g. if a law firm promised to help with any litigation that might occur) to respond to an attack, would have a deterrent effect. It wouldn't really be much of a weapon against commercial software or infringers, but it would at least guarantee the status quo.
Umm...perhaps I'm misunderstanding, but this is happening on computers owned by the city, and used by city employees. Key word: employees.
Not citizens, not townspeople. Government employees, on their work computers. These are the 'end users' in this discussion, not the actual recipients of the government's services.
Nobody is saying that the customers (in this case, the citizens who get their services from the government) have to change anything. In fact it seems as though one of the main objectives of the migration is to be transparent from a customer perspective. People doing their automobile registrations, for instance, will still see the same web pages, etc.
This is purely an internal change, by a government office. The way you wrote your comment, you'd think that the Free Software gestapo were going to go kicking down people's doors and forcing them to install Debian at home. That's just not true, and while the employees should certainly have a say in anything that effects them (and probably have a much greater say than typical employees in either the public or private sector in the U.S.), they are at the end of the day employees, and free to leave if they don't like the direction their workplace is headed.
The key words are "being depreciated." That doesn't change anything about the installed user-base who depend on VBA applications or tools. It means they would have had to change anyway in the near future, but that's irrelevant now that the move to Linux has been announced.
Now, all the effort at replacing them will be blamed on the OS switch, as opposed to just being written off as a necessary upgrade in order to use the new framework and remain competitive. And of course the end users will still bitch and moan the same either way, now it'll just be directed at Linux instead of Microsoft.
Don't get me wrong, I'm all for the switch, I think Munich should be applauded. However everyone has to realize that when you make any kind of big change or migration, it instantly becomes world's largest scapegoat. Any problem, real or imagined, regardless of whether it would have happened under Windows anyway, will be its fault.
Oops -- in my editing I cut a chunk out: ... and then assigning [the patents] to the FOSS organization 'closest' to the project who would have the resources to employ them so they could be used against GPL violators.
I find this very interesting.
Maybe the solution to the GPL software patents "problem" is for GPL software authors to begin aggressively patenting their work?
I mean, there are certainly lots of things that exist in the various flavors of the Linux operating system that could be patentable (certainly by the low standards the Patent Office here in the U.S. seems to be using). Rather than just scream and yell about how evil software patents are, maybe as a community we should start actively and encouraging authors to patent new concepts, and then assigning them to the FOSS so they could be used against GPL violators.
It seems that perhaps this is the direction that the GPL 2 drafters have in mind?
Software patents are looking more and more to be a powerful new weapon between the various entities competing for mind- and market-share in the tech world. It would be irresponsible for the FOSS community not to attempt to arm itself, as a deterrent if nothing else.
Or, since it's just as likely:
d) Eliminate Third-World countries?
I think broadband was driven by Napster and the MP3 revolution, more than anything else. I can't speak for S. Korea -- maybe online network games drove it there. But among the admittedly biased sample group of people that I knew when broadband really exploded, most of them got it because they wanted to download free music.
:) ) is very high.
Now, I think what's driving it is people who graduate from colleges or universities and have gotten used to being on a high-speed connection of some sort, and don't want to go back to dialup when they move into their first apartment. Most young graduates have the disposable income to burn, and within this group I'd be willing to bet that broadband penetration (no pun intended for an article about internet porn
Napster and its ilk were the killer apps that spurred broadband into residential areas, and it's the wide and ubiquitous presence in college dorms that will guarantee a steady flow of new young subscribers. Once you've had high speed, you'll never go back.
Correct.
So you can't modify some GPLed software and release it under the "Mr. Underbridge Public License" (UPL?), but if you're a developer who just wrote something novel, you can surely apply your license to it.
But obviously you can't take something that's licensed under the GPL and "remove offending portions" that you don't like, before releasing your modifications. That would undermine the whole idea.
However if you're the copyright holder of the original code, as (I believe) Linus is of the kernel, then you could cut the GPL up to your heart's content before applying it to your work. You just can't modify the license on somebody else's work, or on a derivative you've made from their work.
I'll bet you watch network television, too.
You have to read the link -- the process they're using grows the algae in artificially or externally heated conditions, so there's an energy input, and also in an artificially CO2 rich atmosphere, which would come from fossil fuels. It is a significantly more efficient use of fossil fuels than our direct consumption of them today, but it is not the closed-carbon-cycle that you are alluding to.
It's a common misconception (not one you necessarily made, but I'm just using your post as a starting point) that Deep Blue just "brute forces" the entire chess game, based on the opposing player's possible responses to every move, analyzing every possible avenue to exhaustion. This is not true.
Even with all of today's computing capacity, until you got to the endgame, the power required would still be prohibitive. Instead, Deep Blue pursues every move it can make to a depth of (I've been told) 8 or 9 moves. It's fairly "intelligent" in the way it decides which avenues are fruitless and which to pursue further. And of course it does many avenues at once, in parallel, and then compares the benefits and gains of each before deciding on a move.
To pidgeon-hole it as just some sort of beefed up desktop computer running Chessmaster 3000 doesn't do the design much justice. It was a wonderful application of massively parallel computing technology, at a time when most people didn't know what 'parallel computing' was.
Yes, and like studying the birds to build an airplane, trying to jump-start a small star in your backyard might not be the way to clean, safe power generation, either.
Sometimes discoveries which seem "right around the corner" linger just out of our reach for hundreds of years, until the right person or group of people has the vision and arrives at the right time to produce something really revolutionary.
Seriously -- this is about the long and short of it.
It sounds less like they really were going to switch to Linux, than they were just playing a rather large and complicated game of chicken with their vendors / potential vendors.
FOSS is in many cases to corporate types less an actual piece of usable technology, than a big stick they can wave vaguely in the direction of their vendors and subcontractors, to keep their costs down. "Your bid is too high! We're switching to Linux, it's free!" "Okay, how about we cut another 30%." "Sold!"
I think you're substantially right, I heard the difference (on the macro-scale, anyway) is mainly in the viscosity and gelling point. On the chemical level it's just slightly different ratios of various hydrocarbons. The reason I heard that mixing Kero and #2 heating oil worked better than straight #2 or straight kero was that it makes the viscosity and flashpoint close to that of diesel. Might not matter depending on what you're going to use the fuel in.
I have actually seen a demonstration of small-scale biodiesel production from SVO, and it was interesting but struck me as being not particularly practical. It requires fairly large quantities of lye and methanol, and the lye is consumed in the process -- it is not catalytic, at least in the small-batch process recipes I've seen.
If someone could come up with a small scale continuous catalytic process, then we could really get a 'cottage industry' of bio-diesel manufacturing going. But right now the batch processes that bio-diesel enthusiasts are championing don't seem ready for prime time.