Most wedding professionals don't make their own prints in their own darkrooms (it's usually fine art photographers who do that, and a lot of them are moving toward digital). They send it out to a lab, often a lab that specializes in wedding photography and knows exactly what to do. Hand-made prints do indeed cost a lot, but there's no inherent reason why a customer needs to go through an intermediary to do that.
Personally, I don't see why Epson bothers - the Free Software community has outdone them. It would be far better if they just helped projects like GIMP-Print and SANE out more, although I guess that means they can't stick Epson logos all over the software... ahh, the joys of corporate ego gratification.
As it happens, Epson already does help us (Gimp-print) out, in exactly the right fashion -- they provide programming manuals for their printers, usually quite promptly too. They're available at www.epsondevelopers.com. You can get there from their main product web site, actually.
They don't provide us with any information about their color management or dither algorithms, which seems fair to me -- that's their real "secret sauce" over their competition. We don't need that; we have bright people who can come up with clever dither algorithms, and we also have people interested in color management. It's probably better for everyone if we do things differently from Epson in that regard; it gives us opportunities to devise better algorithms.
1) At present, we don't know what the nature of the GPL violation is. Certainly by the sound of Brad Kuhn's comment it sounds like the violation was either inadvertent or due to a misunderstanding of the requirements of the GPL. They're not "getting away" with violating the GPL at all -- they're fixing it, on their own nickel.
2) Sure, if you're really unethical you might be able to get away with something like that once, but if you keep making the same mistake you'll wear out your welcome soon enough.
Most big companies hate admitting mistakes (so do most small companies, and most individuals, for that matter). Epson could have simply withdrawn these packages without ever saying why, and either never put them back up or reinstated them later without ever saying what happened, but they chose to admit their mistake.
I'm not at all surprised to hear that Epson is negotiating in good faith with the FSF, and has chosen to go above and beyond what the FSF asked for (publicly admitting their violation and working to resolve it). In my own experience (as project lead of Gimp-Print, I have found Epson to be far more clued-in about working with the free software community than other printer vendors that I have come across.
In our case, it's evident that they understand that what we want is access to information about how their printers work, rather than information (such as details of their color management) that they quite legitimately consider proprietary. Because of that, it has been much easier for us to track their printers, and as a result their printers enjoy top notch support within our project. Perhaps more directly relevant to this, they have not used their project (PIPS) as an excuse to starve ours of data. So I'm certainly pleased (but not at all surprised) to see Epson acting with such high ethical standards under these circumstances, too.
Well of course no one knows how any of these business models would do! Nobody has tried them yet. Until they're tried (i. e. until the RIAA's monopoly is broken to the degree that any of this could be done without it getting sued out of existence), it's impossible to determine how they would work.
Furthermore, what's wrong with a musician working another job?
Mundie completely misses the point -- the purpose of copyleft is essentially the protection of the IP involved in creating something. Of course, the goal of this "protection" is rather different, but that's what it amounts to.
The GPL is also a better license to use if someone wants to release something that they also want to commercialize. Releasing something under a free non-copyleft license means that one's competition can pick it up and run with it, without giving anything back; GPL'ing it retains the practical option of selling a version under a commercial license. This option isn't theoretical; Ghostscript and Qt are both prime examples of it.
Mundie's real objection -- and surely he's smart enough to understand the real situation -- is that anything released under the GPL can't be commercialized by third parties in the traditional proprietary fashion without the permission of the author. But that's exactly the idea, of course. It's not too different, at some level, from Microsoft's aim -- they wouldn't be too happy if someone else decided to start selling a knockoff of VB using their code without giving Microsoft a cut. Whether there's an explicitly commercial angle to using the GPL (e. g. Ghostscript or Qt) or not (the Linux kernel, GNU) isn't particularly relevant, unless you believe that there's something inherently more moral about making money such that pursuit of that entitles you to trample over someone else.
Mundie may or may not believe that personally, but he's certainly smart enough to understand what the GPL is really about, and he's also smart enough to realize that to come right out and say that Microsoft believes it should have the right to "annex" anything that it pleases wouldn't be too popular. It would also weaken Microsoft's anti-piracy message for him to say that it believes it should have the right to "pirate" code that somebody else has written, ergo this subterfuge.
There are actually a lot of inexpensive classical recordings, even contemporary recordings (and even pieces that are still under copyright -- just about anything written in the 20th century, and there's a lot of interesting classical music written in that time period). These discs sell in much lower volumes than major pop recordings (probably 2 or more orders of magnitude less). For a large orchestral work, the number of performers can be very large also -- Mahler's 8th, the "Symphony of a thousand", typically uses 700 or so musicians.
The prices for tickets to major orchestra performances are fairly high, but not all that different from the prices to major pop groups. Concert halls are much smaller than the arenas and stadiums that major pop groups perform in, and require substantial upkeep. Nor do classical radio stations really fund very much; their ad income is much lower than pop stations.
There certainly are very expensive classical recordings, although you can find just about any Beethoven, Mozart, Tchaikovsky etc. piece in budget recordings. These budget recordings, BTW, are usually performed by less well known musicians, but the quality is just as good. Certainly this isn't going to be true for less well known composers. I have a 2 CD set of Havergal Brian's 1st symphony (the "Gothic") that I paid $30 for; one of the very few exceptions to my rule not to pay more than $10 for a CD. I had no objection, either; that composer is very obscure (although extremely interesting); the CD surely sold very few copies, and the effort involved to record this was titanic; it requires about as many performers as Mahler's 8th, and it's quite a lot longer.
So no, I don't believe that the costs of recording a Metallica disc are any greater (per copy) than those of a Mozart disc. Far from it, by 2 orders of magnitude. The issue is simply supply and demand; the supply is artificially constrained. Recording companies, and everyone else, would have no trouble making perfectly good profits charging $5-7/disc; at those prices it simply wouldn't be worthwhile for most listeners to copy discs, when the time involved is factored in.
A lot of packages are code-rich, but documentation-poor (such as the project I lead, Gimp-Print). If your writing skills are good, and you understand how something works from a "normal user" perspective, your assistance in that regard could be invaluable.
There's a bit of a danger in sending money to individual developers, particularly on large projects. If somebody sent me money out of the blue for the work I've done on Gimp-Print, I'd find it rather embarrassing, since it's a team effort; I'd turn right around and donate it to the FSF, or EFF, or someone else who's doing some good. Donating money to one of these organizations in the name of the project (as long as you're sure that that's not going to offend the key people on the project; donating to the FSF in the name of someone who's violently opposed to RMS would be most insulting) would be a way of furthering the ultimate goals of the people working pro bono on the project, while not singling out individual contributors to the project.
Re:Maybe, but not as much as you'd think
on
Insanely Audiophile
·
· Score: 3
Carver's an interesting fellow. He actually built a "real" high-end tube amp that everyone raved over (he probably sold a few), and also built a conventional (by his standards -- it used a highly unconventional power supply that didn't need a monster transformer, but I've long since forgotten the details) amplifier with a matched transfer function that he sold rather more of. The high end critics really hated his much cheaper amplifiers, even if they couldn't tell the difference.
Of all the snake oil salesmen in high end audio, though, I think the cable mafia is the worst. Particularly the digital cables (which others have already commented on) and some of the really bizarre speaker cables, which in some cases look like they would be more counterproductive than anything. I will certainly admit that trashy connectors can cause problems, although simply gold plating them and taking care not to run them right next to a power line should take care of just about anything.
I also don't mind requiring code developed under those circumstances to be distributed in the public domain. However, that's a red herring. There's no problem at all (in my view) with the NSA distributing their Linux mods; they can distribute them in the public domain, since that can be incorporated freely into GPL'ed code.
The GPL only requires that you distribute a derived work (containing previously GPL'ed code and your code, which you still hold the copyright to) under the GPL. You can still distribute the code that's purely your own under any additional licenses you please, when that code is distributed by itself. It's only when it's distributed in combination -- as a single work -- that it must be distributed under the GPL. That's why there's no problem with incorporating BSD licensed code into GPL code. The BSD code can still be distributed as it was before; the GPL only applies to the instance that was incorporated into the GPL'ed software. The thing you don't want to do is put GPL'ed code (that's covered only by the GPL) into something else, if you're not prepared to distribute the combination under the GPL.
The potentially thorniest issue that I see is what happens if you use macro, constant, and variable names from GPL'ed header files. Use of the macros per se isn't an issue; they could be rewritten (the header files just describe an interface; that description can be rewritten). However, if the names themselves are considered to be subject to copyright there could be an issue.
IANAL, so take it for what it's worth (i. e. nothing).
The LGPL differs from the GPL in how it treats linking to a library. Roughly speaking, the LGPL permits linking to a library without requiring disclosure of what's linked against it. However, changes to the library itself are still covered by the source distribution requirements. It certainly is not as weak as the BSD/X11 family of licenses.
RMS has no difficulty with the LGPL under the right circumstances, and the FSF has laid out suggestions for when which license is appropriate. GPL'ing libc, for example, would be counterproductive.
I am project lead for gimp-print, which is a package of high quality printer drivers for POSIX environments. The heart of the package is in a large shared library (libgimpprint.so). We have chosen to GPL rather than LGPL the package. I wouldn't particularly like to see a printer vendor take the package, add a dynamic loading facility to allow loading proprietary driver modules to it (they would have to distribute that part of it), and then using it as a core for a proprietary driver for their printers. I suspect that other printer vendor(s) that do cooperate with us wouldn't care to see that happen either.
That said, though, getting around GPL'ed libraries isn't necessarily as hard as you might think. The key to doing it is to write a simple application that provides an IPC mechanism to access the library facilities. That part, of course, has to be released in source form. However, a binary application that talks to that IPC server is not covered by the GPL, as it is not a derivative of a GPL'ed work. Of course, that's a bit more work, and may (or may not!) result in less performance, but it can certainly be done in a lot of cases.
A regular program of exercise, including strength training, should help tremendously, although not overnight and you do have to be careful not to overdo it initially (that can lead to more problems).
You can always start with simple walking, although if you're really badly out of shape it's best to see a doctor first before even doing very much of that. Gradually build up the distance and speed. For strength training, it's best to discuss it with a trainer to at least get started; overdoing that can really cause problems.
I personally like using the Concept II indoor rower. It gives a fantastic cardio workout in a relatively short amount of time (since it's using your whole body; most other cardiovascular exercise uses primarily the lower body exercises), and on the higher resistance settings it also gives some amount of strength training. But it does work the back quite hard, and it's not something to start with if you're out of shape. Good form is important; bad form can really hurt the back.
If Linux had been GPL it could not have been marketed by companies like IBM.
Try that one again? The Linux kernel is GPL (with the binary module exception, admittedly). The core utilities in the vast majority of Linux distribution are all GPL.
If Open Source were a political movement then we would keep RMS away from the podium as assiduously as the alleged President's handlers keep George Bush away from press interviews.
The point is moot, since RMS doesn't claim to speak for Open Source.
Everytime I turn around, it seems like I read a letter to the editor from RMS making a distinction between open source and free software, or between the Linux kernel and the GNU/Linux operating system. By constantly using his position as a (semi-)celebrity to associate himself with one small, relatively radical subset of open source movement in general, I think he may alienate some folks out there.
RMS is the leader of the free software movement, which long (better than 10 years) predates the term "open source". He's arguing a very specific position -- that freedom is important in its own right, independent of any business advantages that may come from the use of open source. He's not trying to argue from a business perspective, and should not be judged on that basis.
Since RMS most likely won't get up there and emphasize that there are alternatives to the GPL that may be attractive, I wonder how widely his ideas will be accepted by the people who have the money.
Read the license page on the FSF's web site. It lists a very large number of licenses, which it groups into "GPL-compatible", free but not compatible with the GPL, and non-free, and gives specific reasons why each is placed in each category, and recommends reasons for choosing particular licenses. While I doubt that RMS will go into deep discussion of this at his talk -- that would be appropriate for a presentation at a conference, not a general lecture -- his position demonstrates extensive thought, certainly not "one size fits all". Again, as for the issue of money, that's not what he's trying to address.
However, I could make a very strong case that in fact the GPL is one of the best licenses that a business could use in licensing its free source output. The reason, interestingly enough, is that the GPL is probably the strongest widely-accepted license there is for protecting program source against proprietary use by someone else. It's interesting that the situations where the FSF recommends use of the LGPL -- a weaker license -- are those where the software in question is a commodity implementation of a standard, such as libc. The parallels between software that a business would typically try to keep closed and sell for money and GPL'ed software, and software that a business would freely give away and LGPL'ed software, are significant. http://www.fsf.org/philosophy/why-not-lgpl.html quite clearly notes that the purpose of the GPL is to build up a significant portfolio of IP that may be used freely by anyone so long as they agree to put any derived distributed work under the same conditions.
I see in RMS passionate beliefs, and also unwillingness to compromise. Could it hurt the open source movement in the long run?
As I noted above, RMS's position is in fact quite nuanced; the LGPL itself is a compromise of sorts, and he even agrees with the decision to LGPL Ogg Vorbis, on the grounds that uptake of a free alternative to MP3 is of sufficient importance that even the existence of proprietary programs using Ogg Vorbis will benefit the free community.
You're really not that far off -- the normal mean in Fairbanks in July is in fact 63, but again that's a) mean, and b) normal. It could be considerably higher during a warm summer, and even in a normal summer there could be days or weeks of very much warmer weather. The record high there is 99; the record high for Alaska is 100 at Ft. Yukon, which is above the arctic circle.
Like I said in another post, Valdez would make a better site, anyway, if someone wanted to do this. Even there (or anywhere), there's the whole issue of single point of failure, which is bad juju however you look at it.
As far as ambient cooling, it's easy to forget that it can get fairly warm up there for 2-3 months of the year. The record high in Barrow, at the extreme northern tip of Alaska on the coast, is 78F. Inland (except at high elevation) it's going to be somewhat higher, and the normal summer temperatures will be substantially higher (Barrow's normal diurnal temperature range in mid-July is 34-46F). So there will definitely need to be air conditioning -- very substantial air conditioning -- for the 1-5% of the year when the temperature exceeds 50F. The daily insolation up there is also very high, due to the continuous daylight in the summer, even if the sunlight's weaker than at lower latitudes.
To take this to extremes, the average annual temperature in Verhojansk, Russia is about 2F. Problem is that the record low is about -90, but the record high is 98F, which is probably a bit higher than the record high in Key West. Average case planning isn't enough; they've got to take into account worst cases, too. That means figuring out what to do up to temperatures of maybe 85F or higher.
Dumping excess heat in the permafrost isn't a very good solution, either. Ice isn't a terribly efficient conductor of heat (neither is stagnant water), so it will simply melt the permafrost. That's bad news, because suddenly you're sitting on top of a swamp and your foundation isn't much good. Not to mention that it's messy environmentally, too. I believe the pipeline and associated structures are well-insulated, as much to protect the permafrost as to keep the oil flowing.
It might actually make more sense to do this in Valdez, at the southern terminus -- it's a lot more accessible, it's still fairly cold most of the year, and the ocean is available as a heat sink all year round (the ocean up there is pretty cold at all times).
The GPL is specifically not a contract. It does not impose any restrictions that are not already in existence due to copyright. It merely offers a conditional grant of other rights in excess of those that would normally be enjoyed.
The example you gave -- "all are permitted to redistribute this work, in original or modified form, so long as they do not remove this notice, including the copyright notice and disclaimer" -- is no different. It also offers rights beyond what would normally be available to others under copyright law, conditioned on certain actions (preserving the notice). It's just that the conditions are somewhat different.
Re:What about the anti-genetic backlash?
on
Genetic Stone Soup
·
· Score: 4
What happens is they set up one or two of the richest farm owners with thier patented grow twice as fast corn or wheat. These farmers then have a large advantage over all of the family farms that did not get the mansanto handout. The monsanto farmer then buys out the smaller farmers. When the mansanto farmers own most of the farmland mansanto then raises the price of thier grain (which by the way cannot reproduce) and the large farmers are forced to sell out to a large agricorp.
Why? They could always buy their grain from a different supplier. And if they've managed to get themselves locked in to a contract which allows Monsanto to raise their prices at their whim then that's a foolish move and these farmers are reaping what they have sown.
This reminds me of Pastor Niemoller's words. If Monsanto drives out the local suppliers, then the farmers don't have a choice who to do business with. As for their "foolish moves", I don't think it's either a stretch or unfair to say that farmers in developing countries lack the business acumen and strategic foresight that Monsanto has, and it's not realistic to expect that they can do business on the same playing field as Monsanto. Remember that a "free market" assumes good knowledge on both sides of the transaction; if one side has all the knowledge and the other doesn't understand what's going on, it's not a free market any more, but rather a scam.
The same applies to Chinese farmers buying GM food. You talk about doing research on issues such as disease resistance. Precisely how, pray tell, are they going to do such research? Assuming they even know about the genetic basis for disease resistance, and that crop diseases in the US differ from those in China, you're assuming that they have the resources to do this kind of investigation.
It's not only governments that can oppress. Anywhere that there's an excessive concentration of power there can be oppression. It's not a matter of duly constituted laws and sovereign states (most dictatorships have neither; Saddam Hussein rules by personal whim and doesn't have too much respect for borders of neighboring states). If Monsanto's doing this kind of divide and conquer in Latin America, it's no better than if it's a government doing the same thing.
In Germany they first came for the Communists and I didn't speak up because I wasn't a Communist.
Then they came for the Jews, and I didn't speak up because I wasn't a Jew.
Then they came for the trade unionists and I didn't speak up because I wasn't a trade unionist.
Then they came for the Catholics and I didn't speak up because I was a Protestant.
Then they came for me and by that time no one was left to speak up.
What a bunk argument, serving only to rationalize extortion. We have a choice in the matter - if people turn to a life of crime because of desperation, then it's their own fault that they'll end up in jail, hopefully for a good long time. There's no realistic danger of this happening, as long as we reduce the welfare state.
Under those circumstances, a rational person might just decide that prison is a better deal than living in the streets. Anyway, why do you think reducing the welfare state will reduce the danger of this happening?
Ah, so you are actually arguing that I should be paying 'security' or extortion money so that I'm not held up at gunpoint?
The real issue is that if we create a class of unemployable and uninsurable people, we may not have a choice in the matter. If the only practical way for someone to live is through crime, we shouldn't be too surprised if that's what people in that situation turn to.
Nothing's really gained by foisting the cost on employers per se, but the whole system of health insurance through the workplace is badly flawed to begin with. However, allowing insurers to screen people for genetic issues, and offering no alternatives, means that some people will be unemployable and uninsurable. If not for simple humanitarian concerns, the prospect of a large number of people who will never be granted an opportunity to participate in the economy -- and therefore to either be on welfare or have no choice but to, shall we say, live any way they can -- should make it clear why this kind of policy is just asking for trouble in the long run.
almost always wind up being a group, cooperative affair. Usually there wind up being some important milestones (completing the edges, filling holes in certain areas, completing the puzzles), but everyone wins when people help each other out with their little piece of the overall puzzle.
Slashdot Mirror
Most wedding professionals don't make their own prints in their own darkrooms (it's usually fine art photographers who do that, and a lot of them are moving toward digital). They send it out to a lab, often a lab that specializes in wedding photography and knows exactly what to do. Hand-made prints do indeed cost a lot, but there's no inherent reason why a customer needs to go through an intermediary to do that.
The 1290 should print edge to edge with Gimp-print. If it can't, please file a bug.
They don't provide us with any information about their color management or dither algorithms, which seems fair to me -- that's their real "secret sauce" over their competition. We don't need that; we have bright people who can come up with clever dither algorithms, and we also have people interested in color management. It's probably better for everyone if we do things differently from Epson in that regard; it gives us opportunities to devise better algorithms.
1) At present, we don't know what the nature of the GPL violation is. Certainly by the sound of Brad Kuhn's comment it sounds like the violation was either inadvertent or due to a misunderstanding of the requirements of the GPL. They're not "getting away" with violating the GPL at all -- they're fixing it, on their own nickel.
2) Sure, if you're really unethical you might be able to get away with something like that once, but if you keep making the same mistake you'll wear out your welcome soon enough.
Most big companies hate admitting mistakes (so do most small companies, and most individuals, for that matter). Epson could have simply withdrawn these packages without ever saying why, and either never put them back up or reinstated them later without ever saying what happened, but they chose to admit their mistake.
I'm not at all surprised to hear that Epson is negotiating in good faith with the FSF, and has chosen to go above and beyond what the FSF asked for (publicly admitting their violation and working to resolve it). In my own experience (as project lead of Gimp-Print, I have found Epson to be far more clued-in about working with the free software community than other printer vendors that I have come across.
In our case, it's evident that they understand that what we want is access to information about how their printers work, rather than information (such as details of their color management) that they quite legitimately consider proprietary. Because of that, it has been much easier for us to track their printers, and as a result their printers enjoy top notch support within our project. Perhaps more directly relevant to this, they have not used their project (PIPS) as an excuse to starve ours of data.
So I'm certainly pleased (but not at all surprised) to see Epson acting with such high ethical standards under these circumstances, too.
What about tweaking the performance for the preference of the listener -- changing the tempo, phrasing, instrumental balance, and so forth?
Well of course no one knows how any of these business models would do! Nobody has tried them yet. Until they're tried (i. e. until the RIAA's monopoly is broken to the degree that any of this could be done without it getting sued out of existence), it's impossible to determine how they would work.
Furthermore, what's wrong with a musician working another job?
Mundie completely misses the point -- the purpose of copyleft is essentially the protection of the IP involved in creating something. Of course, the goal of this "protection" is rather different, but that's what it amounts to.
The GPL is also a better license to use if someone wants to release something that they also want to commercialize. Releasing something under a free non-copyleft license means that one's competition can pick it up and run with it, without giving anything back; GPL'ing it retains the practical option of selling a version under a commercial license. This option isn't theoretical; Ghostscript and Qt are both prime examples of it.
Mundie's real objection -- and surely he's smart enough to understand the real situation -- is that anything released under the GPL can't be commercialized by third parties in the traditional proprietary fashion without the permission of the author. But that's exactly the idea, of course. It's not too different, at some level, from Microsoft's aim -- they wouldn't be too happy if someone else decided to start selling a knockoff of VB using their code without giving Microsoft a cut. Whether there's an explicitly commercial angle to using the GPL (e. g. Ghostscript or Qt) or not (the Linux kernel, GNU) isn't particularly relevant, unless you believe that there's something inherently more moral about making money such that pursuit of that entitles you to trample over someone else.
Mundie may or may not believe that personally, but he's certainly smart enough to understand what the GPL is really about, and he's also smart enough to realize that to come right out and say that Microsoft believes it should have the right to "annex" anything that it pleases wouldn't be too popular. It would also weaken Microsoft's anti-piracy message for him to say that it believes it should have the right to "pirate" code that somebody else has written, ergo this subterfuge.
There are actually a lot of inexpensive classical recordings, even contemporary recordings (and even pieces that are still under copyright -- just about anything written in the 20th century, and there's a lot of interesting classical music written in that time period). These discs sell in much lower volumes than major pop recordings (probably 2 or more orders of magnitude less). For a large orchestral work, the number of performers can be very large also -- Mahler's 8th, the "Symphony of a thousand", typically uses 700 or so musicians.
The prices for tickets to major orchestra performances are fairly high, but not all that different from the prices to major pop groups. Concert halls are much smaller than the arenas and stadiums that major pop groups perform in, and require substantial upkeep. Nor do classical radio stations really fund very much; their ad income is much lower than pop stations.
There certainly are very expensive classical recordings, although you can find just about any Beethoven, Mozart, Tchaikovsky etc. piece in budget recordings. These budget recordings, BTW, are usually performed by less well known musicians, but the quality is just as good. Certainly this isn't going to be true for less well known composers. I have a 2 CD set of Havergal Brian's 1st symphony (the "Gothic") that I paid $30 for; one of the very few exceptions to my rule not to pay more than $10 for a CD. I had no objection, either; that composer is very obscure (although extremely interesting); the CD surely sold very few copies, and the effort involved to record this was titanic; it requires about as many performers as Mahler's 8th, and it's quite a lot longer.
So no, I don't believe that the costs of recording a Metallica disc are any greater (per copy) than those of a Mozart disc. Far from it, by 2 orders of magnitude. The issue is simply supply and demand; the supply is artificially constrained. Recording companies, and everyone else, would have no trouble making perfectly good profits charging $5-7/disc; at those prices it simply wouldn't be worthwhile for most listeners to copy discs, when the time involved is factored in.
A lot of packages are code-rich, but documentation-poor (such as the project I lead, Gimp-Print). If your writing skills are good, and you understand how something works from a "normal user" perspective, your assistance in that regard could be invaluable.
There's a bit of a danger in sending money to individual developers, particularly on large projects. If somebody sent me money out of the blue for the work I've done on Gimp-Print, I'd find it rather embarrassing, since it's a team effort; I'd turn right around and donate it to the FSF, or EFF, or someone else who's doing some good. Donating money to one of these organizations in the name of the project (as long as you're sure that that's not going to offend the key people on the project; donating to the FSF in the name of someone who's violently opposed to RMS would be most insulting) would be a way of furthering the ultimate goals of the people working pro bono on the project, while not singling out individual contributors to the project.
Carver's an interesting fellow. He actually built a "real" high-end tube amp that everyone raved over (he probably sold a few), and also built a conventional (by his standards -- it used a highly unconventional power supply that didn't need a monster transformer, but I've long since forgotten the details) amplifier with a matched transfer function that he sold rather more of. The high end critics really hated his much cheaper amplifiers, even if they couldn't tell the difference.
Of all the snake oil salesmen in high end audio, though, I think the cable mafia is the worst. Particularly the digital cables (which others have already commented on) and some of the really bizarre speaker cables, which in some cases look like they would be more counterproductive than anything. I will certainly admit that trashy connectors can cause problems, although simply gold plating them and taking care not to run them right next to a power line should take care of just about anything.
Try "sextuple-u" -- it's easy to pronounce.
I also don't mind requiring code developed under those circumstances to be distributed in the public domain. However, that's a red herring. There's no problem at all (in my view) with the NSA distributing their Linux mods; they can distribute them in the public domain, since that can be incorporated freely into GPL'ed code.
The GPL only requires that you distribute a derived work (containing previously GPL'ed code and your code, which you still hold the copyright to) under the GPL. You can still distribute the code that's purely your own under any additional licenses you please, when that code is distributed by itself. It's only when it's distributed in combination -- as a single work -- that it must be distributed under the GPL. That's why there's no problem with incorporating BSD licensed code into GPL code. The BSD code can still be distributed as it was before; the GPL only applies to the instance that was incorporated into the GPL'ed software. The thing you don't want to do is put GPL'ed code (that's covered only by the GPL) into something else, if you're not prepared to distribute the combination under the GPL.
The potentially thorniest issue that I see is what happens if you use macro, constant, and variable names from GPL'ed header files. Use of the macros per se isn't an issue; they could be rewritten (the header files just describe an interface; that description can be rewritten). However, if the names themselves are considered to be subject to copyright there could be an issue.
IANAL, so take it for what it's worth (i. e. nothing).
The LGPL differs from the GPL in how it treats linking to a library. Roughly speaking, the LGPL permits linking to a library without requiring disclosure of what's linked against it. However, changes to the library itself are still covered by the source distribution requirements. It certainly is not as weak as the BSD/X11 family of licenses.
RMS has no difficulty with the LGPL under the right circumstances, and the FSF has laid out suggestions for when which license is appropriate. GPL'ing libc, for example, would be counterproductive.
I am project lead for gimp-print, which is a package of high quality printer drivers for POSIX environments. The heart of the package is in a large shared library (libgimpprint.so). We have chosen to GPL rather than LGPL the package. I wouldn't particularly like to see a printer vendor take the package, add a dynamic loading facility to allow loading proprietary driver modules to it (they would have to distribute that part of it), and then using it as a core for a proprietary driver for their printers. I suspect that other printer vendor(s) that do cooperate with us wouldn't care to see that happen either.
That said, though, getting around GPL'ed libraries isn't necessarily as hard as you might think. The key to doing it is to write a simple application that provides an IPC mechanism to access the library facilities. That part, of course, has to be released in source form. However, a binary application that talks to that IPC server is not covered by the GPL, as it is not a derivative of a GPL'ed work. Of course, that's a bit more work, and may (or may not!) result in less performance, but it can certainly be done in a lot of cases.
A regular program of exercise, including strength training, should help tremendously, although not overnight and you do have to be careful not to overdo it initially (that can lead to more problems).
You can always start with simple walking, although if you're really badly out of shape it's best to see a doctor first before even doing very much of that. Gradually build up the distance and speed. For strength training, it's best to discuss it with a trainer to at least get started; overdoing that can really cause problems.
I personally like using the Concept II indoor rower. It gives a fantastic cardio workout in a relatively short amount of time (since it's using your whole body; most other cardiovascular exercise uses primarily the lower body exercises), and on the higher resistance settings it also gives some amount of strength training. But it does work the back quite hard, and it's not something to start with if you're out of shape. Good form is important; bad form can really hurt the back.
Try that one again? The Linux kernel is GPL (with the binary module exception, admittedly). The core utilities in the vast majority of Linux distribution are all GPL.
The point is moot, since RMS doesn't claim to speak for Open Source.
RMS is the leader of the free software movement, which long (better than 10 years) predates the term "open source". He's arguing a very specific position -- that freedom is important in its own right, independent of any business advantages that may come from the use of open source. He's not trying to argue from a business perspective, and should not be judged on that basis.
Read the license page on the FSF's web site. It lists a very large number of licenses, which it groups into "GPL-compatible", free but not compatible with the GPL, and non-free, and gives specific reasons why each is placed in each category, and recommends reasons for choosing particular licenses. While I doubt that RMS will go into deep discussion of this at his talk -- that would be appropriate for a presentation at a conference, not a general lecture -- his position demonstrates extensive thought, certainly not "one size fits all". Again, as for the issue of money, that's not what he's trying to address.
However, I could make a very strong case that in fact the GPL is one of the best licenses that a business could use in licensing its free source output. The reason, interestingly enough, is that the GPL is probably the strongest widely-accepted license there is for protecting program source against proprietary use by someone else. It's interesting that the situations where the FSF recommends use of the LGPL -- a weaker license -- are those where the software in question is a commodity implementation of a standard, such as libc. The parallels between software that a business would typically try to keep closed and sell for money and GPL'ed software, and software that a business would freely give away and LGPL'ed software, are significant. http://www.fsf.org/philosophy/why-not-lgpl.html quite clearly notes that the purpose of the GPL is to build up a significant portfolio of IP that may be used freely by anyone so long as they agree to put any derived distributed work under the same conditions.
As I noted above, RMS's position is in fact quite nuanced; the LGPL itself is a compromise of sorts, and he even agrees with the decision to LGPL Ogg Vorbis, on the grounds that uptake of a free alternative to MP3 is of sufficient importance that even the existence of proprietary programs using Ogg Vorbis will benefit the free community.
You're really not that far off -- the normal mean in Fairbanks in July is in fact 63, but again that's a) mean, and b) normal. It could be considerably higher during a warm summer, and even in a normal summer there could be days or weeks of very much warmer weather. The record high there is 99; the record high for Alaska is 100 at Ft. Yukon, which is above the arctic circle.
Like I said in another post, Valdez would make a better site, anyway, if someone wanted to do this. Even there (or anywhere), there's the whole issue of single point of failure, which is bad juju however you look at it.
As far as ambient cooling, it's easy to forget that it can get fairly warm up there for 2-3 months of the year. The record high in Barrow, at the extreme northern tip of Alaska on the coast, is 78F. Inland (except at high elevation) it's going to be somewhat higher, and the normal summer temperatures will be substantially higher (Barrow's normal diurnal temperature range in mid-July is 34-46F). So there will definitely need to be air conditioning -- very substantial air conditioning -- for the 1-5% of the year when the temperature exceeds 50F. The daily insolation up there is also very high, due to the continuous daylight in the summer, even if the sunlight's weaker than at lower latitudes.
To take this to extremes, the average annual temperature in Verhojansk, Russia is about 2F. Problem is that the record low is about -90, but the record high is 98F, which is probably a bit higher than the record high in Key West. Average case planning isn't enough; they've got to take into account worst cases, too. That means figuring out what to do up to temperatures of maybe 85F or higher.
Dumping excess heat in the permafrost isn't a very good solution, either. Ice isn't a terribly efficient conductor of heat (neither is stagnant water), so it will simply melt the permafrost. That's bad news, because suddenly you're sitting on top of a swamp and your foundation isn't much good. Not to mention that it's messy environmentally, too. I believe the pipeline and associated structures are well-insulated, as much to protect the permafrost as to keep the oil flowing.
It might actually make more sense to do this in Valdez, at the southern terminus -- it's a lot more accessible, it's still fairly cold most of the year, and the ocean is available as a heat sink all year round (the ocean up there is pretty cold at all times).
The GPL is specifically not a contract. It does not impose any restrictions that are not already in existence due to copyright. It merely offers a conditional grant of other rights in excess of those that would normally be enjoyed.
The example you gave -- "all are permitted to redistribute this work, in original or modified form, so long as they do not remove this notice, including the copyright notice and disclaimer" -- is no different. It also offers rights beyond what would normally be available to others under copyright law, conditioned on certain actions (preserving the notice). It's just that the conditions are somewhat different.
This reminds me of Pastor Niemoller's words. If Monsanto drives out the local suppliers, then the farmers don't have a choice who to do business with. As for their "foolish moves", I don't think it's either a stretch or unfair to say that farmers in developing countries lack the business acumen and strategic foresight that Monsanto has, and it's not realistic to expect that they can do business on the same playing field as Monsanto. Remember that a "free market" assumes good knowledge on both sides of the transaction; if one side has all the knowledge and the other doesn't understand what's going on, it's not a free market any more, but rather a scam.
The same applies to Chinese farmers buying GM food. You talk about doing research on issues such as disease resistance. Precisely how, pray tell, are they going to do such research? Assuming they even know about the genetic basis for disease resistance, and that crop diseases in the US differ from those in China, you're assuming that they have the resources to do this kind of investigation.
It's not only governments that can oppress. Anywhere that there's an excessive concentration of power there can be oppression. It's not a matter of duly constituted laws and sovereign states (most dictatorships have neither; Saddam Hussein rules by personal whim and doesn't have too much respect for borders of neighboring states). If Monsanto's doing this kind of divide and conquer in Latin America, it's no better than if it's a government doing the same thing.
In Germany they first came for the Communists and I didn't speak up because I wasn't a Communist.
Then they came for the Jews, and I didn't speak up because I wasn't a Jew.
Then they came for the trade unionists and I didn't speak up because I wasn't a trade unionist.
Then they came for the Catholics and I didn't speak up because I was a Protestant.
Then they came for me and by that time no one was left to speak up.
Under those circumstances, a rational person might just decide that prison is a better deal than living in the streets. Anyway, why do you think reducing the welfare state will reduce the danger of this happening?
The real issue is that if we create a class of unemployable and uninsurable people, we may not have a choice in the matter. If the only practical way for someone to live is through crime, we shouldn't be too surprised if that's what people in that situation turn to.
Nothing's really gained by foisting the cost on employers per se, but the whole system of health insurance through the workplace is badly flawed to begin with. However, allowing insurers to screen people for genetic issues, and offering no alternatives, means that some people will be unemployable and uninsurable. If not for simple humanitarian concerns, the prospect of a large number of people who will never be granted an opportunity to participate in the economy -- and therefore to either be on welfare or have no choice but to, shall we say, live any way they can -- should make it clear why this kind of policy is just asking for trouble in the long run.
Life isn't an economic textbook folks.
almost always wind up being a group, cooperative affair. Usually there wind up being some important milestones (completing the edges, filling holes in certain areas, completing the puzzles), but everyone wins when people help each other out with their little piece of the overall puzzle.