there are two positions one can take on the issue of intellectual property, either you believe in it or you don't. either information is a material resource that can be governed and owned, or information is an immaterial resource that cannot be governed nor owned.
you can't criticize software vendors for restricting access to their source code and invoking intellectual property law to sue people who copy it, and then turn around and invent a copyleft license so you can do the exact same thing. you can't copy the source code from unix and emacs and violate their license to write unauthorized clones and then lock those clones behind your own license to stop others from doing whatever they want with "your" code.
words are not things and language is a means, not an end. the significance of a legal document is not in the text itself but in how the text is interpreted and enforced in a court of law. permissive licenses such as the bsd license can effectively be thought of public domain licenses because they contain no enforceable non-trivial restrictions that could conceivably ever warrant an actual court case, but copyleft licenses like the gpl can and have been involved in real courtroom cases and any civil litigation in any courtroom requires a specific plaintiff, and it cannot be "the people" as in a criminal case filed by the state. the plaintiffs in gpl violation cases are typically non-profit organizations like the free software foundation or the software freedom law center who act as legal representatives for one or more of the developers, and which developers get to be constituents and potential beneficiaries in the lawsuit is decided entirely at the discretion of whomever happens to be in the room when the lawsuit is first filed.
to illustrate the real-world ramifications of this, consider the busybox lawsuits filed against various companies by busybox developers erick anderson and rob landley represented by the software freedom law center. the cases were settled for undisclosed sums and these two developers got paid off, but none of the other hundreds of contributors to the busybox project, including the original authors bruce perens and dave cinege, were represented or given a share of the settlement or even contacted by the plaintiffs. so in practice, enforcement of the gpl does not really represent the open source community and all the contributors to a particular project, it only represents the tiny fraction of contributors who have enough extra time and money on their hands to pursue litigation and are under no compunction to share the rewards with anyone else.
thus, the whole idea of the gpl is just one big contradiction, it is a way for certain privileged people to appoint themselves as representatives of a supposedly democratic community project with or without the consent or even the knowledge of the other contributors and invoke their names and their work to extract financial restitution for themselves. this is not democracy, this is not the commons; this is just private ownership. and there's no way to correct for this, it is the result of a fundamental category error, trying to govern a public resource using a legal system that is only designed to govern private property. you can't have your cake and eat it too, either you release something to the public and accept that the world will do whatever it pleases with it, or you treat it as private property. intellectual property is either legitimate or it isn't, there is no coherent middle position.
>you can't criticize software vendors for restricting access to their source code and invoking intellectual property law to sue people who copy it, and then turn around and invent a copyleft license so you can do the exact same thing
You mean you're against weaponizing these systems against the capitalists benefiting off of them? Having an opportunity to do so and not acting on it is just seems plain stupid to me, because withholding interaction makes little difference, when there is no opposition it can tangibly be invested in.
Still i stand by my prior analysis, that the GNU project has failed its stated purpose. Copyleft appears incidental to the existence of quality software and more often than not the better project is written by a small corporate team or a single person in their spare time. People also tend to underestimate the number of projects GNU have subsumed, like GNU patch being a fork of Larry Wall's code or GNU tar deriving from a "public domain tar" Usenet posting, which was rumored to have come from Sun.
Real power over software is not determined by who has distribution rights, though restricting source access certainly helps. It flows from the governance of the project and ultimately who pays for its development. What does it matter to me if a program's source is available, when that program is a python monstrosity with 5 git submodules, that i couldn't as of yet get to build on my system? While this class of software was often made to address a need, its place in a corporate quagmire is sure to limit its influence and routinely frustrate users, with any sense of community involvement functioning as a ploy to profit from free labor. Quality software in contrast remains one of the few things modern businesses can't reliably produce, which gives open-source projects that have a certain leverage.