Yep. And, conversely, that software didn't need to get made either, because if it did then somebody would have been willing to pay for it! In other words, there was no loss.
Maybe. Or maybe it would have been a little bit useful to a lot of people, useful enough to pay a dollar or two but not useful enough to spend the time to find eachother and collectively hire someone. Cases like this are not well served by the current system, and it's an interesting question to see if there could be a better system.
First of, let's not make a false dillemma; it's not a matter of all the time devoted to produce those works, or all the time devoted towards something that earns money - at least, not necessarily. One can, for instance, have another job that earns you money, and create 'art' works (or whatever) as an aside. While time is limited, it's seldom limited to the point where one has absolutely NO time left to do something else than 'work for a living'. Oh, sure, I do that myself actually. But I don't know how useful the stuff I make in my spare time is, and since my bank account doesn't know either there's a limit on how much time I can spend on it -- regardless of how useful it actually is.
So, instead of one big sponsor, one can have several minor ones. As long as your product is popular, I think there is a definite chance of that. (As an example; see Freenet; it's paying a full time devl for several years now, just by what people donate to the project.) Cool, I didn't know they did that. Maybe that is the best option, and we just need to reduce the overhead of actually making donations (need to specifically set up a account with a micropayment service, after finding a service that you actually trust and that the project uses, etc...). I mean, I don't have a paypal account (and I'm not sure I really trust them enough to want one anyway), most projects probably can't take credit cards, that one online payment service (e-gold or something?) got shut down for working too well...
Secondly; your assertation at the end is false. There have been examples enough where people did not need to pay for something (well, unless one goes into semantics and conclude that only the sun rises for free). It's not an absolute necessity; though of course, in our capitalistic society (which I agree works much better than a communistic one;-)) as a whole, the market rules, and people pay for products they want. I'm not saying they have to pay to get things, I'm saying that I have to have a source of income from somewhere.
But it must be said that the cost for a product consist of the material, and the time/work one put in it. In this respect, digital 'products' are something outside the normal. (And, in extension, all 'IP' is.) The cost of material there is...well, none. One DOES put time/work in it - in the ORIGINAL, but that is often not in comparison to the number of digital copies that can be made. After the original, the time/work that one puts in it, is virtually nothing. Yeah. So how does one connect the usefulness of the result with funding to support the effort needed to produce it, without blocking some people from benefiting? I suppose an alternative would be to set up a utopia where people don't actually need to work for a living, but that would probably be even harder to get right...
It works fairly well, the question is can it be made to work better. Right now, there are a number of disconnects between making useful things and getting paid. It would be nice if those disconnects could be removed, without adding overhead that lessens the usefulness like our current copyright system seems to do.
"Somebody needs to pay you for something"
This is actually not true. Or rather, you need to do something that someone will pay you for. And if that something ends up being digging ditches instead of writing software, then that software just doesn't get made.
"the transformation to digital methods of production and distribution therefore poses to the twenty-first century a fundamental moral problem. If I can provide to everyone all goods of intellectual value or beauty, for the same price that I can provide the first copy of those works to anyone, why is it ever moral to exclude anyone from anything?"
Moglen
At the same time, how is it possible to produce those works if you need to spend your time producing something salable so that you can eat? Somebody needs to pay you for something, and the most effective way we've figured out to do that seems to be to restrict availability of what you produce to only those who can pay you for it.
I think this model is horribly broken, but what would be a good, general, replacement for it? Not everyone can get sponsorships...
Because the person who created the work didn't give you permission to release it to everyone. If a friend lets you borrow his car, do you loan it out to everyone you see too? I find your view of morality quite skewed towards your own beliefs on copyright. This is fundamentally different, in that there is a single car in your example, but an unlimited number of identical copies of any information. Car analogies do not always work...
Linux is only "free" if your time is worth nothing and you have enough of it to spend learning how to do things you already knew how to do. So if it lets me do things better/faster, does that mean it actually has *negative* cost?
Up next, Frequent Slashdotter finally moves to Ubuntu, feels that this is the Year of Linux on the Desktop.
Uh huh. From the article:
But to this day I've never heard an answer to one question: Since even Linux advocates admit that it's harder to use, what can you do with Linux that you can't do with Windows, to make it worth switching over to? If I was nervous about Vista because some of the interface had changed and some of my old programs no longer worked, it wasn't helpful to tell me to switch to a system where all of the interface would change and none of my old programs would work.
I'd say that it is not harder to use, and that Windows doesn't give you an integrated, easily accessible repository full of useful/fun programs. (And if it did, you'd have to pay extra for them.)
This is how at LEAST 95% of computer users feel IMO. And going beyond that to the first half of the statement, people keep talking about the console wars in terms of a "killer app" and how that's what the PS3 needs to break through. Whatever, I don't want to get in to that here, but why isn't the same being said about Linux? For the average user, the best Linux can offer is "mostly as good" and often incompatible (and by that I mean, if even something SLIGHTLY doesn't work, people don't want to care how to fix it. Making it "just work" is everything). Yes you can customize, yes you can add all these different things made by people for free, but quite frankly most people don't care! They just want a computer to work "as they expect" and no more. Once ANY expectation is broken, they rebel against it. The only exception is when something is SO good (and often so easy to use as well) that it invalidates the rest. Linux DOES NOT HAVE THIS right now for most users. No, that's a load of crap. We're just too lazy to replace something that's "good enough", especially when it's something we don't understand well. A replacement doesn't have to be perfect, or bug-compatible, or have a "killer app". It just has to be enough better to overcome people's laziness and fear of the unknown. Or I suppose coming pre-installed would also work, even without being any better.
Although it means hobbyists could no longer tinker, we are at the point where that hobbyist tinkering could have significant implications for the international system of computing infrastructure. Why should unlicensed software authors be any different from unlicensed doctors? Both can cause harm; in the former case, potentially more harm.
Doctors always work on people, and will harm/kill people if they screw up. Software people can only mess up whatever their software is given access to mess up. Hobbyist tinkering wouldn't be good in the control systems for large hydroelectric dams, but there's absolutely no reason to forbid people from writing, say, games for fun. The rules belong on the other side, about what software can be used for a particular task in a particular situation rather than about who can write software.
If you have $100 Billion to spend, and you build tanks, bombs and combat jets, you are helping the economy, but only a small amount. Once you use a bomb, it will not add value to the economy. When you build a combat jet, it will not add (much) to the future economy. A bullet shot, is worthless.
This is actually very interesting, because it is "good for the economy" to have people do work that's just going to be thrown away. This seems to imply that they otherwise wouldn't have had anything useful to do. It feels like something is horribly broken (most likely related somehow to how uneven wealth distribution is between people who do work and people who own infrastructure), but I'm not sure exactly what.
I much doubt whether the average user cares (never mind understands) if the kernel is monolithic, microkernel, or heated corn The "heated corn" type tastes like paper.
This gets particularly interesting if you don't allow globals. Then you get a capability system, and can do all sorts of cool things.
And I'd guess that doing this with a type-safe bytecode instead of hardware protection would mean that there's minimal or no performance penalty compared to what we have now, because of the lack of copying and TLB flushes and the optimization tricks that having a runtime code profile could let JIT compilers do.
The GPL doesn't make a distinction between static and dynamic linking. You're not allowed to dynamically link to a GPL-ed library unless your code is GPL-ed as well. That was the whole point of the LGPL, to allow dynamic linking without requiring the entire end result to be released under the GPL. The only thing the GPL can do is grant permission to do what would ordinarily be forbidden by copyright law. If copyright law doesn't restrict me from dynamic linking to something, then the GPL also cannot restrict me from doing that, because I don't have to accept it to be allowed to.
Incidentally, a related question that someone raised with me in another discussion was whether you can, legally robustly, donate a work to the public domain. At that point, copyright would no longer apply, so a licence of any kind would be unnecessary to make a legal copy. However, try tracking down a law in any major jurisdiction that actually has an explicit provision allowing someone to voluntarily and permanently give up a copyright they hold and leave a work in the public domain, in the same absolute sense that it would be after any applicable copyright expired or for, say, government works that are legally public domain from the start. I've never found one, which raises the interesting (and somewhat scary) prospect that no-one can actually give away their work for free in a manner that any recipients can rely on, even if all parties wish this to be the case...
I believe this is pretty similar to the line of thinking that led to the WTFPL. "Public domain isn't always valid, so here's a license that does the same thing."
What he does is link to libmpeg2, a library under the full (not L) GPL. This qualifies him as a derivative work of libmpeg2 just as much as if he had borrowed the whole lot.
A "derivative work" is when you take something and make changes to it. I could see static linking doing this, but if dynamic linking did wouldn't you not need a separate copy of the library being linked to?
of course, this is a good argument for contributors to GPL projects to either retain the copyright in their own name (or assign it to the FSF who they can trust to keep it free) so that projects they contribute to find it very difficult to go closed-source. Not everyone does trust the FSF to keep their software free.
I am guessing that he has found a new business model/investor and now wants to change.
But he will not be able to revoke the GPL for the old work. The reality is that he used lower level GPL software to build with. As such, he entered into a contract that said, I am re-paying you by adding to the work. Once he released it, it was payment.
I'm not sure I buy that argument. (What contract? Payment for what? GPL doesn't require anyone to write&release further software.)
Imagine if MS sold you an application, and then later decided to jack up the price you paid for it i.e. they charge you again. That is illegal (though you may have to pay for certain extras).
Here, there is a very direct connection between the payment and what is being paid for.
In fact, if he could retract the license, then why do commercial companies with their big fancy lawyers not retract your right to use their software when they want you to upgrade? In particular, MS sells you a app say MS word. License says that if you paid us for you have the right to use this on one system. Later, MS wants you to upgrade. How do they encourage it? They stop support it for it. But if they could retract the license and say that you are now illegal and must get rid of this, don't you think they would? In fact, IBM and others would be doing it ALL the time. Point is, the GPL was legally applied to this app. It has been there for a long time.
He has ZERO rights to pull it back. The only right that he has is change the license of future code.
In those cases, you paid for the software. In this case, it was a gift. That's a pretty big difference, and probably invalidates your logic here.
Imagine a database with its main column being VARCHAR(255) and using about full length of it, then search using a lot of LIKE and AND, picking various short pieces out of that column, and the database being terabytes big. Try to invent a way to index it. Convert it to an HTML table and put it where googlebot can see it.
Slashdot Mirror
Yep. And, conversely, that software didn't need to get made either, because if it did then somebody would have been willing to pay for it! In other words, there was no loss.
Maybe. Or maybe it would have been a little bit useful to a lot of people, useful enough to pay a dollar or two but not useful enough to spend the time to find eachother and collectively hire someone. Cases like this are not well served by the current system, and it's an interesting question to see if there could be a better system.It works fairly well, the question is can it be made to work better. Right now, there are a number of disconnects between making useful things and getting paid. It would be nice if those disconnects could be removed, without adding overhead that lessens the usefulness like our current copyright system seems to do.
Moglen
At the same time, how is it possible to produce those works if you need to spend your time producing something salable so that you can eat? Somebody needs to pay you for something, and the most effective way we've figured out to do that seems to be to restrict availability of what you produce to only those who can pay you for it.
I think this model is horribly broken, but what would be a good, general, replacement for it? Not everyone can get sponsorships...
Sold more copies of what? XP?
It's the hardware devices that have Compatibility issues with Vista.
Similarly, it's the applications which have trouble integrating with Vista.
How low can you go?
Compared to the hardware you'll need, the OS is downright cheap!
Microsoft is a business, and they certainly like it!
This one's correct, because it is reliable -- reliably slow.
Due to the size of these issues, this "rectification" has produced many goatse look-alikes.
You can't get in trouble online if the computer doesn't work.
I'd say it'll take until the second or third.
But the snarky comments are all mine.
Either sqlite or custom binary, depending on what the data is and why a database is "not an option".
Doctors always work on people, and will harm/kill people if they screw up. Software people can only mess up whatever their software is given access to mess up. Hobbyist tinkering wouldn't be good in the control systems for large hydroelectric dams, but there's absolutely no reason to forbid people from writing, say, games for fun. The rules belong on the other side, about what software can be used for a particular task in a particular situation rather than about who can write software.
A real programmer would have taught the computer to write the code itself.
This gets particularly interesting if you don't allow globals. Then you get a capability system, and can do all sorts of cool things.
And I'd guess that doing this with a type-safe bytecode instead of hardware protection would mean that there's minimal or no performance penalty compared to what we have now, because of the lack of copying and TLB flushes and the optimization tricks that having a runtime code profile could let JIT compilers do.
It looks like it's an iframe of http://section.slashdot.org/images/iframe/$section.html .
Incidentally, a related question that someone raised with me in another discussion was whether you can, legally robustly, donate a work to the public domain. At that point, copyright would no longer apply, so a licence of any kind would be unnecessary to make a legal copy. However, try tracking down a law in any major jurisdiction that actually has an explicit provision allowing someone to voluntarily and permanently give up a copyright they hold and leave a work in the public domain, in the same absolute sense that it would be after any applicable copyright expired or for, say, government works that are legally public domain from the start. I've never found one, which raises the interesting (and somewhat scary) prospect that no-one can actually give away their work for free in a manner that any recipients can rely on, even if all parties wish this to be the case...
I believe this is pretty similar to the line of thinking that led to the WTFPL. "Public domain isn't always valid, so here's a license that does the same thing."
Why? It's not like (dynamic) linking to something means that you make copies of it or creative modifications to it...
Which kind doesn't go away when you close your eyes?