PJ said (in part):
"...I thought I'd point that out, in case any of you think the future for Linux would be so great if we could just have the "freedom" to allow proprietary codecs and drivers or let vendors close rights off with hardware. Please remember that the Unix folks tried that semi-proprietary route already, thinking it would increase profits and market share, and it failed miserably. For you "pragmatists" who say there's nothing wrong with closed, proprietary software, here's what's wrong with it: people don't want it. If you give them a choice, they'll choose open every time. They wanted Unix until it stopped being open, and then the market declined precipitously. End users like open. Why wouldn't we? It gives us the opportunity to modify the software to do exactly what we want individually, as opposed to what some vendor guesses the largest group of its customers probably wants..."
If you're reading this, you're probably either family or somebody who already knows about the debate. On the off-chance that you are unfamiliar with the details, here's a synopsis:
- The FSF (Free Software Foundation) is working on version 3 of it's GPL Public license
- The Linux Kernel maintainers generally prefer the current version (GPLv2)
- The FSF is viewing things as idealists, and do not want patent legislation or other legal trickery to hinder the effectiveness of their license in ensuring the freedom of people who use software
- The kernel maintainers are viewing things as pragmatists, and do not want the new license to hinder their ability to write code that may include software designed to limit the way that it may be used
I'm sure I will be corrected if I have that substantially wrong, but that (to me, at least) seems to be the crux of the issue. The FSF stands for freedom to use software, and the kernel maintainers desire a license that permits more freedom for the developers of the software it covers.
I greatly value PJ's insight, determination, and hard work, and she has made me pay *much* more attention to what software licenses say (I opted not to test drive several pieces of software because of that newfound attention, as a matter of fact). I think, though, that in this case, I'm more in the kernel maintainers' camp.
Obviously, GPLv2 has server the Open Source development community well...that's why they continue to use it. It ensures that people can freely use their code, makes sure that it doesn't get stolen outright, and helps drive development resources towards popular projects.
There is a very real problem in software patents (mainly that they exist), and it does need to be addressed. Likewise, there is a very real problem of anti-circumvention and anti-piracy software that makes criminals out of hackers (via the DMCA and its brethren). The freedom to tinker has a long and glorious history of expression in both the hardware and software worlds. I don't think that the kernel maintainers want to see that freedom go away any more than the FSF does.
So, now we come to the disagreements...PJ says:
"...I thought I'd point that out, in case any of you think the future for Linux would be so great if we could just have the "freedom" to allow proprietary codecs and drivers or let vendors close rights off with hardware..."
I don't see proprietary drivers or codecs as being the thing that fractures Linux. The problem that the UNIX vendors that are being referred to were having stemmed from not having compatible UNIXes and demanding vendor support for all of them. True, it benefits many if hardware vendors open their drivers and APIs, but it's the kind of decision that is driven by the market, and if present signs of uptake in Linux on the desktop (Ubuntu, anyone?) are any indication, many hardware vendors are about to lumber in that direction.
Hardware implementations of rights management (trusted computing, etc) are just plain evil, but unless there's effective education for the masses that actually *reaches* people, no one will think it's important until it's too late. We'll all be running whatever version of Windows we're told to. GPLv3 works fine for Linux users, but unless they're in the majority, it doesn't prevent those hardware restriction systems from creeping into being ubiquitous.
It's too early for GPLv3 to solve that problem...Linux has to be *far* more widespread in use for it to be effective at stopping it.
For you "pragmatists" who say there's nothing wrong with closed, proprietary software, here's what's wrong with it: people don't want it. If you give them a choice, they'll choose open every time. They wanted Unix until it stopped being open, and then the market declined precipitously. End users like open. Why wouldn't we? It gives us the opportunity to modify the software to do exactly what we want individually, as opposed to what some vendor guesses the largest group of its customers probably wants.
I think there's a small bit of irritation clouding the most valuable parsing of the pragmatists view. Making a statement that there's nothing wrong with proprietary software does not mean that the person making that statement believes it's the best way to make software, or even a good idea. That statement (again, to me) means that the person saying it doesn't care whether people are allowed to make proprietary software. The reason that it's okay is pretty much exactly what PJ said after the second comma. People *don't* want it. They *do* like open. It *does* give us opportunity without having to depend on a vendor's whim.
Here is one place that I actually see the far edges of free vs. closed meet, in that by taking the GPL that extra step, you prevent certain uses, improvements, changes, and contributions from being made. It's that perverse side of freedom that permits people to create something that denigrates the framework that enables them in the first place.
That's about all I can stand to write, and I feel all the worse for having to have said it, but it's out there. PJ, I hope you can find it within yourself to forgive me, or (better yet) to convince me that I'm mistaken...lord knows you've done it before.