5 ms·

> Stallman's conception of free software says that you should be able to charge as much as you want. Yes, he was referring to the service of duplicating and ma

> Stallman's conception of free software says that you should be able to charge as much as you want. Yes, he was referring to the service of duplicating and mailing tapes pre-Internet. You seem to have mistaken that for a key philosophical issue of some kind, when it's secondary. > reacting to the hyperbole I worked for national labs back in the day. Every dept. wrote their own terminal applications, plotting programs, etc. which consumed serious grant budgets. Free and open source software has been a huge boon to science. The most important? GCC and family. Second is probably Linux, which is GPL. See a pattern yet? It really sounds like you need a history lesson on the FSF, because you're way off base.

eesmith 7y ago
Really? Because I think it is part of the key philosophy of free software. Consider #1: I think of an improvement to gcc. I get a copy of the gcc source, modify it, and prove to my satisfaction that it works. I stop at that point, intellectually satiated, tell a few friends, and don't release any software (source or binary). The principles of free software say that I am not forced to release any changes. I am not even obligated to send improvements upstream. Am I wrong about this? Consider #2: The same as above, but I know that Yoyodyne could save $1 million per year with my improvements. I don't tell them. The free software principles say that I have no obligation to tell them. Am I wrong about this? Consider #3: The same as above, but I go to Yoyodyne and say that I'll sell my gcc modification to them - under the GPL - for $250,000. They of course can then redistribute those changes, including back to mainline. Is that contrary to any free software principle? I don't think so, according to my years of reading the history of the FSF and the free software movement. But, suppose you are correct, and Stallman's essay only applies to "the service of duplicating and mailing tapes pre-Internet". If so, there's a simple workaround: Consider #4: I have the idea, implement it, and distribute it to a single organization - a tape duplicating service - and it's that organization which sells the software to Yoyodyne, for $300,000, of which $250,000 comes back to me. (My brother just happens to run the tape service.) Is that against the principles of free software, and if so, how? I don't think Stallman's "selling free software" essay was meant to be so easy to work around. Just because free software is generally distributed at no price or "reasonable cost of physically performing this conveying of source" (quoting GPLv3), doesn't make it the same as saying that there's an free software principle which obligates that. While the open data movement in science does have the principle that the data should be free, or should only cover the "reasonable cost of physically performing this conveying of" data. How am I off base?
- pessimizer 7y ago
  I don't think you are. Trade secrets aren't covered by GPL except in the case of v3 on software accessed through a terminal. Even in this essay, he makes clear than open patents aren't the same as Free Software. That being said, the Free Software movement was obviosly seminal in all of the open data talk, and for that matter open patent talk; it's not the same thing, but a strong and heavily influential idea in one context inspires analogues in other contexts. I think your overall point is wrong, although I think your supporting point is completely correct.
- inimino 7y ago
  You're not wrong, however... if you sell the software, you're obligated to provide source. A better analogy would be if you publish the paper, you're obligated to release your data. Nobody ever suggested that individuals or companies can't do private research at all. You just shouldn't be able to have it both ways. If you want the credit, show your work.
  - eesmith 7y ago
    Yes, that analogy is the argument in open data. My argument is that that analogy doesn't follow from free software principles. If it did, then it would follow that if you publish the paper, you're obligated to release your source. And we don't see that coming out of the free software movement. We do see that coming out of the open data/open science movement.
    - inimino 7y ago
      Yes, if you publish, you're required to release everything necessary for a practitioner reasonably skilled in the art to reproduce your results, without undue effort. For example, releasing a tarball of source code missing the Makefile that runs the arcane build process is not within the spirit of free software and you will be soundly ridiculed and if GPL is involved may be sued. Dumping a database that, together with your paper, still requires n > 1 grad-student-years of effort to reproduce is not really open science, although it's a huge step up from Elsevier. Is this broadly understood? Probably not, but it is at least well known by some. Honestly, free software people have been working out these "reproducibility" issues in practice for decades before the biologists even got on the bus.

eesmith 7y ago

Really? Because I think it is part of the key philosophy of free software. Consider #1: I think of an improvement to gcc. I get a copy of the gcc source, modify it, and prove to my satisfaction that it works. I stop at that point, intellectually satiated, tell a few friends, and don't release any software (source or binary). The principles of free software say that I am not forced to release any changes. I am not even obligated to send improvements upstream. Am I wrong about this? Consider #2: The same as above, but I know that Yoyodyne could save $1 million per year with my improvements. I don't tell them. The free software principles say that I have no obligation to tell them. Am I wrong about this? Consider #3: The same as above, but I go to Yoyodyne and say that I'll sell my gcc modification to them - under the GPL - for $250,000. They of course can then redistribute those changes, including back to mainline. Is that contrary to any free software principle? I don't think so, according to my years of reading the history of the FSF and the free software movement. But, suppose you are correct, and Stallman's essay only applies to "the service of duplicating and mailing tapes pre-Internet". If so, there's a simple workaround: Consider #4: I have the idea, implement it, and distribute it to a single organization - a tape duplicating service - and it's that organization which sells the software to Yoyodyne, for $300,000, of which $250,000 comes back to me. (My brother just happens to run the tape service.) Is that against the principles of free software, and if so, how? I don't think Stallman's "selling free software" essay was meant to be so easy to work around. Just because free software is generally distributed at no price or "reasonable cost of physically performing this conveying of source" (quoting GPLv3), doesn't make it the same as saying that there's an free software principle which obligates that. While the open data movement in science does have the principle that the data should be free, or should only cover the "reasonable cost of physically performing this conveying of" data. How am I off base?

pessimizer 7y ago
I don't think you are. Trade secrets aren't covered by GPL except in the case of v3 on software accessed through a terminal. Even in this essay, he makes clear than open patents aren't the same as Free Software. That being said, the Free Software movement was obviosly seminal in all of the open data talk, and for that matter open patent talk; it's not the same thing, but a strong and heavily influential idea in one context inspires analogues in other contexts. I think your overall point is wrong, although I think your supporting point is completely correct.
inimino 7y ago
You're not wrong, however... if you sell the software, you're obligated to provide source. A better analogy would be if you publish the paper, you're obligated to release your data. Nobody ever suggested that individuals or companies can't do private research at all. You just shouldn't be able to have it both ways. If you want the credit, show your work.
- eesmith 7y ago
  Yes, that analogy is the argument in open data. My argument is that that analogy doesn't follow from free software principles. If it did, then it would follow that if you publish the paper, you're obligated to release your source. And we don't see that coming out of the free software movement. We do see that coming out of the open data/open science movement.
  - inimino 7y ago
    Yes, if you publish, you're required to release everything necessary for a practitioner reasonably skilled in the art to reproduce your results, without undue effort. For example, releasing a tarball of source code missing the Makefile that runs the arcane build process is not within the spirit of free software and you will be soundly ridiculed and if GPL is involved may be sued. Dumping a database that, together with your paper, still requires n > 1 grad-student-years of effort to reproduce is not really open science, although it's a huge step up from Elsevier. Is this broadly understood? Probably not, but it is at least well known by some. Honestly, free software people have been working out these "reproducibility" issues in practice for decades before the biologists even got on the bus.