Second, I have not ever heard anyone complain about any of the other licenses. Even LGPL does not receive so much criticism. Perhaps *the* biggest controversy with the GPL is that it is a viral license. By virtue of using GPLed code in your software, the rest of your code must also be GPLed. How is that for freedom–you can’t even choose the terms under which your own code may be used? It is a relatively common practice for businesses to license closed-source libraries for use in larger projects. Said libraries are incompatible with the GPL and therefore preclude the use of FLOSS.

Third, one of the more interesting arguments that I have heard is that viral FLOSS licenses like the GPL actually inhibit competition. Since software companies generally operate on the principle of closed source, they cannot use GPLed code. As such, FLOSS raises the entry barrier to a given market making it more difficult for competitors to innovate; they have to first reimplement the wheel *before* they can make it incrementally better. BSD/MIT-style licenses on the other hand allow them to leverage an existing implementation and focus on what they want to sell.

Fourth, not everyone who sells closed-source software is evil. I will admit that this is an exceptional case, but in one instance a company open-sourced their product after they were no longer willing to support it. It is a shame that the GPL crowd is busy evangelizing telling people how closed-source software will give them cancer instead of recognizing business realities and trying to compromise. I would be happy as a clam if more companies made a practice of releasing their source code once it reached the end of its life. GPL evangelists will cede no ground, however, and give OSS a bad name among business pragmatists.

The GPL crowd in general seems obsessed with the idea that some business will come along and sell their work without giving them anything in return. That is a legitimate concern, but the GPL doesn’t seem to me to be the right answer. Also, there are many non-GPL OSS projects out there, and with only one exception I have never heard of them being ripped off; said exception was actually a GPLed project that someone in China had changed the name of and was trying to sell.

Personally, it makes me happy just knowing that someone besides myself actually uses my code, and that’s why I put a BSD license on everything.

- To quote from the page linked to in the first footnote, “There are reasons that can make it better to use the Lesser GPL in certain cases.” – like the GNU C library. The GNU project even recommends non-copyleft licences for some software, like with Ogg Vorbis: http://lwn.net/2001/0301/a/rms-ov-license.php3
- I suspect that very few proprietary licences let you sue the authors for very much at all.

My sympathy to pleas against the use of copyleft licences is limited. The authors of such code have no obligation to assist proprietary software authors if they do not want to! Similarly, Faux is free to use non-copyleft licences for his own code as much as he wants. I doubt anyone is going to change anyone else’s mind here.

Now, there seems to be an assumption that all freedoms (however defined) are valuable. I am not sure that “freedom for freedom’s sake” is a very convincing argument (and this would apply to advocates of the FSF’s four freedoms as well). When presented with a choice of possible sets of freedoms, there will have to be different criteria to choose one set over the other. The implicit question is why the freedom to use “free” code in non-free programs was not included in the definition as a fifth freedom – and the answer is that the author of the definition was not interested in protecting the interests of people who deny freedoms 0-3 to others.

Extending the use of your code to proprietary developers who would deny you the same freedom in return is very generous! That does not imply that those who do not do the same are being mean. The Golden Rule, for instance, does not preclude punishing criminals.

Strong copyleft licences in particular are designed to encourage the development of a large body of copyleft software – which in turn is beneficial for the community of free software developers. Note that proprietary software authors still have complete freedom to reimplement the functionality of the copyleft software – nothing has been taken away from them. They have merely not been actively helped – and I would argue that it is not in the interests of developers who value freedoms 0-3 to assist proprietary competitors who would deny them those freedoms.

To push this conversation out of the 1990s cliche – what are your actual goals when you release or use free software? Do you desire compatibility with proprietary or free software licenses? Do you really not desire the four freedoms outlined by the FSF? What about enforcement of “credit”, trademark protections or “artistic integrity”? Do you have reservations about your code being used in a proprietary manner, and why shouldn’t third-parties release their changes in source form? What do you find objectionable about copyleft? Why do you believe warranty and being “fit for purpose” is so entwined with the choice of a free software license?

These are the interesting issues, not some absurd summation from a tally-chart with an unjustified conclusion.