"Source delivered" can mean many different things though. Stallman's position is essentially that the source must be delivered in a way that respects the four freedoms and any compromise on that is "unethical" and not worth pursuing. My position is that we should start with just having access to the source under any terms, no matter how restrictive, and then move from there.
> I didn't downvote you
Just to clarify: I wouldn't mind if you did. It was just an observation about the knee-jerk response from someone who clearly hasn't even bothered to read the comment.
> we should start with just having access to the source under any terms
Lol. Windows source code has been available under severely-restricted terms, officially for academic purposes, for what, 20 years? It didn't do any good.
I've seen these debates for 25 years by now, they are always the same.
There is no "right to repair" without the 4 freedoms. You can't "repair" if you cannot inspect and run (0 and 1), and your repair is fundamentally meaningless if you cannot communicate it (2 and 3).
> your repair is fundamentally meaningless if you cannot communicate it (2 and 3).
I have spent significant amount of time working around bugs and limitations in proprietary software that could have been fixed easily otherwise, and simply being able to fix it for myself or our company would absolutely be beneficial.
This is so staggeringly blindly obvious that it pains me that this is even a point of contention. Of course things would be better if you can also share, but merely being able to repair is also useful in and of itself. This is exactly the sort of Stallmanistic attitude that's impeding any sort of progress, and why we'll still be having the same discussion 30 years from now with no significant movement in the grand scheme of things.
> simply being able to fix it for myself or our company would absolutely be beneficial.
That's basically the old Unix model, in practice it was unsatisfactory to the point that... people just wrote their own Unix and made a point that fixes should be public - and we got GNU.
Companies could give you the old Unix model tomorrow, if they so wished; but it's effectively lose-lose, because they don't gain anything meaningful (if anything, their support costs are increased by having to deal with thousands of potential user-introduced variations) and you don't gain anything meaningful either - your fix is forgotten the minute you move on and you won't be able to use it with your next employer.
The default today is not the Unix model, it is the Windows model and increasingly the SaaS model. You have none of the 4 freedoms by default.
GP argued that by law by default you should have at least freedom 0 and 1. GP argued that it should not be legal to withhold freedom 0 and 1. GP argued that freedom 0 and 1 should by default actually be facilitated by providing source code. They are arguing for changing the status quo and that doing so gradually is more feasible than all at once.
(Source code is not actually necessary for freedom 1 but it makes it orders of magnitude easier.)
You are arguing that the law is fine because we can just make our own sandbox (GPL software) where we have all 4 freedoms and that is enough. You are arguing that the status quo is fine because you can ignore it while you are in the sandbox.
Turns out the world is a lot bigger than that sandbox.
Your approach towards a world of free software might be to just grow the sandbox until everything else doesn't matter but the GNU GPL family continues to decline in popularity. For this approach you need popularity and for popularity you need effective advocacy and the FSF is not effective.
> They are arguing for changing the status quo and that doing so gradually is more feasible than all at once.
This has been argued over and over and over, and it has made no difference whatsoever. All the limited-rights licenses that have been tried from time to time (MPL et al) have simply fallen by the wayside: because, in reality, nobody is actually interested in this "right to repair" - it was around in the Unix days and it was discarded, because it did not effectively serve users nor businesses.
What have had an impact are effectively two licenses: the MIT/BSD ("do what you want") and GPL ("do what you want but your changes should be public"). By adopting and leveraging those two licenses, the world was slowly steered where we are now - which is a lot better than where we were in the '90s but also where we were in the '70s. We got here by being radical in a punk sense: creating a world of software developed in the open, and fuck what was there before. It's not by bowing to established interests that we ended up with Microsoft developing in the open, it's by bombing their commercial spaces with open software. The minute you stop doing that, they will happily retreat behind the firewall.
> Turns out the world is a lot bigger than that sandbox.
The sandbox is the world at this point, closed software is retreating every day to smaller and smaller niches. You use browsers developed in the open, on websites built with opensource software. Even on mobile, the foundational frameworks are open - the most popular mobile OS is opensource. Obviously it's not a perfect state of things, but it's a world apart from the bad old '90s, and we did get here by keeping the Overton window firmly rooted at one end, thanks in large part to the FSF. It doesn't really matter how much software is GPL, what matters is that other licenses are defined in relation to the GPL.
> The sandbox is the world at this point, closed software is retreating every day to smaller and smaller niches. You use browsers developed in the open, on websites built with opensource software. Even on mobile, the foundational frameworks are open - the most popular mobile OS is opensource. Obviously it's not a perfect state of things, but it's a world apart from the bad old '90s, and we did get here by keeping the Overton window firmly rooted at one end, thanks in large part to the FSF.
The only open source browser with significant usage is Firefox and it is in the single digits. Chromium might be open source but it is not what people use, it's in the other category in statistics. People use Chrome and Edge and Safari and Opera and Vivaldi and none of them are open source. The number of people using AOSP is also vanishingly small. Most Android users use one of the proprietary variants of Android. The year of the Linux desktop never came. The most popular Linux variant after the proprietary Android is the proprietary ChromeOS. And, worst of all, many services that before didn't even involve a computer (ordering food, calling a ride, banking, etc.), now are done through closed source software. It is getting harder and harder to live without proprietary software.
So absolutely not, the sandbox is not the world, the sandbox is getting smaller and smaller in comparison to the world.
The argument was not to no longer hold the Overton window pinned. The argument was to advocate for consumer rights and protections in LAW. Free software advocates are vary much interested in right to repair because right to repair also counters tivo-isation. And the world is much more tivo-ized than it ever was. That the world became so tivo-ized is also a sign of how ineffective the FSF has been at countering a threat they were among the first to draw attention to. You can have all the free software in the world and it is of no use if you can not run it on any hardware.
> I didn't downvote you
Just to clarify: I wouldn't mind if you did. It was just an observation about the knee-jerk response from someone who clearly hasn't even bothered to read the comment.