They would not be factually wrong. The argument is not that their world must "just exist", but rather that there is no evidence either way, and that the model without a creator is the one with fewer assumptions. Both of those statements are factually correct. The burden of proof is on those postulating the existence of a creator to come up with actual evidence supporting the more complex model.

If this world was somehow created, like your virtual world, then either that creator's world "just exists", or it was created in turn. At some point you either have a world which was not created, or you have a cycle or infinite series of worlds, each creating the next. In the former case, there is no evidence to suggest that we are not in that original, uncreated universe; in the latter, there is no uncreated creator.

If you can postulate a world where it is possible for things to "just exist" without being created, then our universe can be one of those things which "just exists". That is a far simpler answer than claiming that there must be an all-powerful creator who "just exists" and who, in turn, created the universe.

It's not unreasonable to expect people who call themselves "libertarians" to actually hold libertarian principles—which primarily means the Non-Aggression Principle, which is incompatible with threatening coercion against anyone who chooses to enter into a voluntary agreement to provide or purchase labor below your arbitrary price floor.

If your basic physics classes didn't at least mention that the Newtonian formulas are only approximations, and not valid for very small dimensions or very high energies, then you should probably ask for your money back. Similarly for qualifiers about static situations and ideal materials. For the problem domain those basic science classes are concerned with, the approximate formulas are perfectly valid. The higher-level classes require different formulas only because they have an expanded problem domain. Similarly, Econ 101 covers a limited problem domain, with assumptions like purely voluntary interaction. Given the same problem, Econ 102 should give the same answer. However, knowing Econ 102 allows you to address a wider variety of situations, such as specifically how the economy is negatively impacted by coercion.

If Econ 102 contradicts Econ 101, then someone messed up. At worst, the rules taught in 101 may not hold in all cases—but the boundaries should have been part of the 101 curriculum.

Anyway, the concept of supply vs. demand and the optimal price-point are fundamental to all levels of economics. There is nothing wrong with what the GP wrote. The part which was omitted, however, is that the "excuse" of rising costs pushes out the marginal producers, reducing the supply and thus raising the optimal price-point for those who are left. How much that price increase depends on a number of factors, including the elasticity of demand, the degree of competition, and existing production margins (i.e. how much the suppliers can afford to absorb before it becomes more economical to close shop or shift to a different product).

Who says you can't? The ability to enter into such an agreement voluntarily is an inherent natural right. If you did make such an agreement of your own free will, then abiding by it is an ethical and moral imperative, even others consider it void. I think most people just aren't stupid or desperate enough to actually agree to such a thing, outside the music industry, which gets by on lingering memories of glamour and obviously unrealistic hopes of being part of the 0.001% that makes it to the top of the heap.

Then there's the fact that what the musicians are selling is really the copyrights to works produced under their contracts, in exchange for a rather one-sided partnership. They can walk away from that and keep the revenues from any future work to themselves. However, just as you won't continue to receive income from a company you leave, they won't receive any further revenues from the copyrights they sold to the labels.

The straw-man arguments are getting a bit ridiculous here, not to mention repetitive. I never said that every element had to be both novel and non-obvious, just that some element had to be. If the patent consists of one thing (the algorithm) which is non-obvious but not novel—a law of nature—plus something else which is novel ("on a computer") but obvious, then the patent shouldn't be granted.

I'm not confusing the issues, I'm saying that software patents should be rejected for both reasons. Subject matter aside, they add nothing to the state of the art which is both novel and non-obvious. From a subject matter P.O.V., irrelevancies relating to run on a computer vs. run in your head aside, algorithms are math, and math is not a patentable subject matter.

The pro-software patent camp is the group confusing the issue, by trying to claim that the subject matter restriction doesn't apply simply because you use a computer to assist you with the math.

We already know that this is your position. The question is: Why? What's so different about using a computer for assistance, rather than pencil and paper?

LMGTFY: Harm to standards and compatibility, Free software projects harmed by software patents

As I already pointed out, patents don't need to completely block something in order to be stifling innovation or otherwise causing harm, they just need to be making it more costly than it otherwise would be.

Nonsense. We don't benefit from people driving around with hexagonal wheels for the sole reason that someone holds a patent on the round version. If there is a variation which is actually better, it will be invented and used for that reason and not just because someone holds a patent on an inferior version. Otherwise, we're better off without the workarounds. To add insult to injury, at the pace of the technology industry that "relatively short time" is generally much longer than it takes for the subject of the patent to become obsolete.

It's relevant here because publication as part of a free software project accomplishes both the innovation and publication benefits that patents are meant to provide, without the cost of a monopoly. In other words, software patents are undermining a superior alternative, one which places working code in people's hands now rather than obtuse legalese they might be able to turn into something useful in 20 years (with almost the same effort as starting from scratch), assuming there isn't some other interlocking patent in effect by that point, a common situation in patent-heavy areas like pharmaceuticals.

Obviously. Without patents you don't have patent litigation or the need for a patent portfolio for cross-licensing.

Not the same conditions, and I'm not referring to their GDPs. These countries don't exist in a vacuum, and patents have world-wide effects, even in countries which don't have patent laws themselves. Import and export markets, for example. If Switzerland wants to export to the U.S., it has to accommodate U.S. patent laws, even if they don't apply within Switzerland. If you could show two countries with similar initial conditions which were actually isolated from each other, with no direct or indirect trade or emigration, one of which had patents and the other without, then you might have an argument. Of course, we'd also have to define what we're looking for—increased "innovation" or a net benefit to the country as a whole.

Closed source code isn't even involved here. The point is that completely open source code is blocked by the patent.

First, it's not a matter of "refusing" to pay. The requirement to pay is impossible to comply with while remaining compliant with the open source license. If others can't use or modify the software with your changes without paying a license fee, you can't incorporate those changes into an open source project.

Second, if those changes would have enhanced the state of the art—and being open source, they are necessarily available to the public—and were independently discovered, not based on the patent or related publications, then innovation has clearly been stifled. Publication as part of an open-source project would have accomplished everything the patent is supposed to do, without the cost of a monopoly. If the changes were based on the patent, then there is a question of how long it would have taken for someone else to come up with an equivalent algorithm in the absence of the patent. Obviously I think it would have been developed anyway, due simply to the demand for the algorithm itself (without which the patent is also worthless) and regardless of the limited incentive of a patent monopoly, but I'm willing to grant that there is room for debate.

Finally, I think you're promoting the Broken Window fallacy by saying that it's good to be forced to reinvent someone to circumvent a patent. Yes, the final result was better that .gif—but what might .gif turned into if others could have improved on it directly, rather than starting over from scratch? Quite a lot of effect was spent on ways to compress images without infringing on the .gif patent, not to mention reimplementing all the other features already in .gif like transparency and animation, and then pushing support for the new formats into new versions of every image-handling program. That effort could have been directed toward improving on the original format, or working on other software, if not for the patent. For that matter, just look at how long it took to get to the point where you could assume decent .png support. (Are we even there yet? Which commonly-supported raster format handles both transparency and animation?)

For me the question is actually how anyone can possibly justify the aggression necessary to enforce a patent monopoly, regardless of the costs or benefits of granting one. However, we're talking hypotheticals here.

The flip side of your question is, how much is NOT invented today because of patent protection? You have to consider both, and both are ultimately undecidable without an alternate reality to compare against. I have yet to see a study, however, which could safely conclude that patents do anything to promote innovation. Even working within the low standards of modern politics, to justify infringing on the property rights of everyone by telling them they can't implement a patented process or device (even if they were completely unaware of it and invented the same thing independently), I would expect some sort of solid evidence that the program is actually working.

Not for a bank loan, no, but an IPO works somewhat like this. Whatever share of the company is sold in the IPO, that fraction of future profit belongs to the new shareholders from then on, not the original owners of the company. In exchange they receive whatever those shares initially sold for at the time of the IPO.

When an artist signs with a label they're essentially selling stock in their future music career.

What is the significant difference between evaluating an algorithm in your head, and doing the same thing with computer assistance, which would justify a patent? What is the novel, non-obvious change in the state of the art which such a patent would document? We agree that it's not the algorithm, or the computer; what's left?

More specifically, what is it about "evaluate this algorithm with a computer" which would make it patentable, which would not also apply to "evaluate this algorithm with pencil and paper"?

This is an obvious straw-man argument. There is no new rule prohibiting patents which build on prior patents. The only rule being applied here, which is hardly new, is that the patent must add something to the state of the art which is both novel and non-obvious. The automobile in your example is not novel, but your transmission is presumably both novel and non-obvious. The patent would cover just the transmission, not the automobile.

On the other hand, if you tried to get a patent on combining an existing transmission with an existing automobile design, I would expect such an application to be summarily rejected unless there was something both novel and non-obvious about the concept or method of combining the two. This is where software patents fail. You have two pre-existing components, the algorithm and the general-purpose computer. The patent application only covers combining these two existing elements together in the most obvious way possible, as running algorithms is the sole purpose of a computer.

Actually, yes. For one obvious example, just look at the areas where interoperability has suffered due to Microsoft's FAT filesystem patents, or patents on media codecs, or lossless compression algorithms, or rendering techniques. Software patents are also inherently incompatible with free and open-source software; any license fee whatsoever will block legal free software development in that area. Driving people to reinvent the wheel, often in an inferior way, isn't progress. Start-ups and other small for-profit organizations have a similar problem; they don't have the portfolio necessary to enter into the standard cross-licensing agreements, or the financial resources for drawn-out patent fights in the courts. Their only real option is to settle for being bought out (under duress) by an larger incumbent.

It's obvious that patents, and especially software patents, impose costs and place roadblocks in the way of innovation. It is hoped, but not proven (and not for lack of trying), that the incentive of a state-backed monopoly is enough to compensate for this. What we lack is an alternate world to compare against, one with the same initial conditions but without patents.

Exactly, because no retraction would be justified. That a particular product eventually makes it to market does not imply that patents do not stifle innovation—which is not the same as saying that no innovation can occur, only that the patents impede innovation rather than contributing to it. To decide that we would need a parallel world without patents to compare against, and we unfortunately do not have access to such a world. All we have are guesses regarding the relative costs and benefits, the costs being much concrete.

We know how much is spend on patent litigation, and about some things which are avoided due to patent concerns. There is more we don't know about specifically which was never created because the cost of dealing with the patents was deemed too high. On the other side, it is rare to be able to claim even that something was invented which would probably not exist were it not for the patent, much less that it certainly would not have been invented.

So... you're saying that patents on software (with or without a computer involved; it makes no difference) do not stifle innovation in hardware? I might have to give you that one, though one could argue that lack of software interoperability due to software patents stifles demand for innovative hardware. The effect is likely to be small, though. What you should be arguing, though you'll have a hard time of it, is that software patents drive hardware innovation. If the effect isn't positive then there's no justification for the patents. It's not enough to argue that they aren't particularly harmful. The algorithms covered by software patents generally run just fine on existing computers, so it's not like software patents drive demand for faster or more efficient computers, either.