> As opposed to just taking it, profiting off it (like ~90% of people who 'eschew' the GPL because they can't make as much money as possible off it) and never doing anything in return.
This right here - though I would extend it to also mention those who take it, extend it, profit off of it, then next-to-never (and many times never) returning any of those changes back to the community.
If we're lucky, they contribute something else back (and those contributions are welcomed, mostly) - but never the changes which allowed them to grow and expand well beyond anyone's expectations. We all know who you are.
I understand the want to see my code used by "the big guys" (or the "up-and-comers" who become "the big guys"); it does look good on one's CV. At the same time, I also understand Stallman's reasonings and message; he has personal experiences which led to the creation of the GPL and his philosophy on software - and he has been around long enough to see how the corporate world really works when it comes to the lifecycle of software. From the birth of an application or system, to its use and adoption, to its death.
Sometimes, its the latter end-of-life phase that we seem to forget about. So often as developers we have to "re-invent the wheel" when had old proprietary software source code been released at EOL, a base to build something better would be in place. Unfortunately, then as now, software has been looked on as an "asset" to be traded and sold for profit, despite the fact that it also depreciates faster than a new car. Yet, like an old Vega, some desperately cling onto that old software hoping to profit "someday, you'll see!".
And in many cases, that source code, and even the binaries - suffers the ignoble fate of "bit rot", sitting in a warehouse on an old hard drive (if we're lucky - if not, it's on some old nearly unreadable 7-bit paper tape literally rotting away, in a format used by a computer system that hasn't been turned on - if one even exists and hasn't been scrapped for gold - in over 50 years).
Original symbolics LISP? Some of the code behind the Apollo program (heck, and that was government funded!)? Cray Research operating systems, compilers, libraries, etc - for the old Cray supercomputers? This list could be extended nearly forever.
The sad thing is, we have lost man-centuries of work in the form of software, due to want-of-profit and bit-rot - and Stallman knows this - we should all know this. Furthermore, even when the code does survive (in whatever form), inevitably, the systems that can run it do not (in general); in some cases, it has become impossible to run code written for certain machines because no examples of those machines, nor their designs (to create an emulator) exist any longer. All we can do is guess at what the program should do (assuming we have at least documentation of the mnemonics and opcodes for the machine language binary to reverse engineer - if you don't have that, nor the real hardware, the binary code might as well be noise).
Crazy enough - in some cases we have the machine and the software extant - but only one operating copy of that machine! For example, there's a company in Texas which uses an old IBM accounting machine to this day; it uses punched-cards for it's data input/output, plus it has a printer. It's "programs" are hard-wired plugboards. A computer history museum delegation has begged them to donate the system to a museum for preservation, as it is the last operating example of such a system. But the company has an attitude of "if it ain't broke, why fix it" - and you can't fault them for that. Even so, one day it will break, and likely whomever is in charge will decide to sell it for scrap rather than donate it (due to likely amount of precious metals in the machine) - and that machine, plus all of its programs - will be lost forever.
Heck - we treat our automobiles and automotive history with greater respect, and that has had arguably a much lesser impact over the last 100 years, than what computer technology has had over the last 20.
/end ranting (and book)