2006-09-29 20:18Licence choicesFreedom and copyrightI have long intended to blog about my personal views on licensing software, not least because the software I write and publicise here should be accompanied by the reasoning behind the choice of licence. Previous posts have probably made clear that I have benefited from [Free] Software, that is, software which is licensed in a way that gives licensees the Four Freedoms (no, not those Four Freedoms). It is thus easy for me to see the advantage to society and the overall economy that Free Software brings, and correspondingly hard for me to justify licensing my own software in a way that disadvantages society and the economy. If my software were of unusually high value (or I were unusually remote from society and the economy) there might be some net value to me in restrictively licensing my software, but to base my decision on that alone assumes that I am not prepared to make the slightest sacrifice (or equivalently, the slightest effort) to improve society or the economy. Nevertheless, I would like to make the case that, on any practical level, I do not appear to have any rights under copyright law, be they the temporary right of an author to exclusively publish my work, or the Fair Use rights that you have if you own a legitimate copy of someone else’s work. I’ll address the latter in another post some time, but as for the former, I do feel that any software or other artistic work I create might be copied about the Internet without me having the resources to fight worldwide legal battles against those I believe have infringed my rights. My work may even be plagiarised by large companies, against whom I would have little chance of success if I tried to take legal action. That is not to say I would just give up; on the contrary I want to stand up for my rights under copyright law and exercise those rights as I see fit, as everyone is permitted to do. I only try to point out that to an individual, copyright law can do more harm than good, and the limited amount of cultural innovation we do have is often due to authors giving away their legal rights to publishing companies and cartels (with some encouraging exceptions). Copyleft and the GPLSo, given that I am not against copyright per se, but I wish to give users of my software Freedom, the obvious thing would be to use a licence on the Free Software Foundation’s long list of Free licences. The FSF understandably categorise the licences by their compatibility with the GNU GPL (henceforth “GPL”), but hint at another deciding factor of whether the licence is “copyleft” or not. I think the reasons given in that link for licensing the GNU project under a copyleft licence are very convincing, and I can personally think of examples where the choice of the GPL has lead to positive results that wouldn’t have been possible under a non-copyleft licence, and the choice of a similar but non-copyleft licence instead of the GPL has lead to results which have hampered Free software. So that leaves the question of whether to be GPL compatible or not. I don’t see any reason to choose to have my software not usable by the many GPL projects out there, and GPL incompatibility just makes things more complicated. If that isn’t reason enough, others have given much better reasons than I, also pointing out the popularity of the GPL. Having no other strong feelings about which licence to use, and with recent legal successes about the enforceability of the GPL, I think that it is the obvious choice. Also, it could be argued that the sheer popularity of the licence protects it in some cases, perhaps the best example of which is the irony that if SCO had proven the GPL unconstitutional, the judge that issued the ruling might have had to use illegal software to do so. This safety in numbers could be called “the network effect”, but I prefer to call it “Hurd immunity” (I can’t believe I just typed that). Specifying and applying the licenceBut the unexplained decisions from last time are why I chose the GPL version 2 in particular, and how I applied that licence to my word count program. With the new (but as yet not finalised) version 3 of the GPL, there are multiple choices for specifying which version or versions apply to your software. The FSF recommend an “auto-upgrading” statement whereby you give the licensee the choice of the current version or any later version. Currently that means the most common wording is “either version 2 of the License, or (at your option) any later version.” , and so choosing to limit (for now at least) licensees of my software to the conditions of version 2 of the licence was a very deliberate choice. The most basic reason for this is that “auto-upgrading” involves a certain amount of trust for the FSF, which they try to alleviate by saying in the licence “The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.” . So, while I have a great deal of respect for what they have achieved, that doesn’t mean I unconditionally trust them or think they are incapable of mistakes. My particular reasons for doubting that their interests will always intersect with mine are the shamefully non-Free GFDL they have created (requiring a questionable compromise from Debian) and the chance that they could use version 3 or later of the GPL to make a seamingly well-meaning stand for Fair Use but end up merely putting pointless restrictions on user’s Freedoms. As for how I applied the licence, which is non-trivial to someone starting out at publishing their own software and trying to keep to the letter of the GPL, my main resources were the advice of the FSF themselves, and the examples of other software written in the same language. Basically, as it was a single file to be used alongside another, bigger package, I included the copyright notice and disclaimer of warranty actually in the file itself, at the top. The alternative would have been putting that message in a separate file, adding “Unless otherwise stated, the following applies to all files that are part of this software” or similar. I mentioned last time that I felt I had included “a copy of the GNU General Public License along with [the] program” by putting a link to the licence on the blog page containing the source. That meant the only area of uncertainty was finding the correct wording for specifying only version 2 of the licence. Naturally I did this by Googling until I found a sufficiently canonical source, namely a GPL-based copyright notice that had been sanctioned by Stallman himself. After all, you do not really receive the full benefit of a widely-used licence if you then apply it to your work in a way that is non-standard and adds legal uncertainty. PostscriptIt is that mention of “(non-)standard” (application of) licences (and the fact that this section of the document is headed with the name of an open standard) that somewhat justifies my classification of this post under the category “Standards”. I think it is also fairly reasonable to put it under “Programming” too, as I am discussing the licences to use for software. I have two final thoughts. Firsly that I look forward to following up this post with one about the technological and legal assault on copyright and Free software and Free media, and to whet your appetite, secondly this: Isn’t the War on Copying Data just like the War on Drugs? (Answer: No, because drugs can’t be duplicated a million times a second and travel at the speed of light). Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
[...] Given my previous comments about licences being a standards issue, I think that the Creative Commons group of licences can be considered popular standards these days. [...]
|
QuicksearchCategoriesSyndicate This BlogBlog Administration |