I’m confused with some of the comments here - isn’t OP asking why the BSD 3-Clause doesn’t include a clause preventing patent treachery, which would be a good clause to have?

In any case, this is why the FSF recommends the Apache 2.0 license over other permissive licenses: https://www.gnu.org/licenses/license-list.html#apache2

If you actively try to avoid software that is licensed under BSD or ISC, only Gnu’s Marxist license, then that means you won’t use OpenSSH or the security tools ported over from OpenBSD into both Linux and Android and will not use an operating system or programs that have been compiled with LLVM/Clang.

OP did say “wherever possible”, not “without any compromises”; I agree that it would be very challenging to try and live on purely GPL software, but it just seems like they’re looking for potential alternatives. I think it’s admirable given their stance on libre software.

Gnu’s Marxist license

This seems like such a blanket statement designed to put down copyleft advocates “because communism”. Do you think the right to repair movement - which advocates for control over one’s hardware (in contrast to software) - is also Marxist? I consider these two movements as practically adjacent to one another.

I prioritize system security, and for that I only care about open source and reject free/libre software due to all of its built-in political implications. But also defend the right for people to make proprietary software. I won’t use it, but defend people’s right to make it.

I ask the following more out of curiousity than argumentatively, because I still don’t quite understand this position: do you still encourage people to use libre software in this case? If libre software is better than proprietary software, why shouldn’t we strive for a world where all software is libre? Is there any reason for software to not respect its users’ freedoms? I’m not saying that we should all be hardcore GPL-only-or-die enthusiasts, either; rather that it seems reasonable to aim for a life free of proprietary software, one step at a time. In my opinion, copyleft can accomplish that better than permissive licenses would.

I think you may be misunderstanding the “free” part of FOSS. FOSS - also known as free software - is free as in freedom, not beer (this confusion is also why I refer to it as libre software). It has nothing to do with money - it is all about having control over the software that you use.

Some here have already pointed out the massive proliferation of libre software that forms practically the entire foundation of the Internet, but I would also like to mention that there are some projects that might even say that being libre software has made it more sustainable; for example, here’s a talk about how the GPLv3 (a copyleft libre software license) keeps the Samba project alive.

There are certain monetization approaches that are infeasible with libre software, yes, but I would argue that this is only ever the case with practices that are anti-consumer. Games as a service is a good example of this; I think it’s absurd you can buy a game that you should rightfully own indefinitely, only for it to become literally borked because it was specifically designed to always be connected to the game company’s servers which could be taken down at any moment. With libre software? You have access to the source code, so it’s not impossible anymore to get your own server running if someone else hasn’t already made the necessary modifications to make it happen.

The philosophy of (and reasons to care about) libre software isn’t quite the topic of this post so I won’t elaborate too much on it (unless you want, of course), but feel free to take a look at this page which discusses just that if you’re interested.

As @[email protected] commented, the official definitions of free software and Open Source actually overlap quite heavily; the concerns made by many - including Stallman/the FSF and even Bruce Perens (author of the Open Source Definition) - involve the belief that Open Source has detached from the values associated with the free software movement.

If you are in fact specifically addressing the fairly small subset of open-source-but-not-free software, I would guess that the overlap is great enough for it to not detract from discussions, and “open source” is simply more commonly used.

Just a note, I’m also pretty sure some people in the comments have mixed up free-as-in-libre software for free-as-in-beer software, which is why I prefer to say “libre” instead.

Is this not the same kind of argument companies make against piracy? I think it’s not so much a pricing problem than it is a service problem. In the same way that people rely on Steam rather than pirating every single game on the market, it’s the services that are offered rather than the price that has to be paid.

Say you go ahead and do this - what guarantees will you make with that price? Guarantees like priority support and timely package updates cost money, which doesn’t sound viable unless the big company is setting absurdly high prices, in which case, that just sounds like competition.

I’m aware I’d be restricted from future binary updates. But my point goes to what the legal definition of what a “restriction” is in this context. The language in this sentence to me is plain, “You may not impose any further restrictions on the recipients’ exercise of the rights granted herein.” RH is attempting to limit my behavior by imposing contractual penalties for exercising rights granted under the GPL. Those penalties being imposed are not just restricting access to future binary updates, but forfeiting all current granted privileges and monies already paid under contract, and the termination of that contract. I would consider that a “restriction”, but IANAL either.

I see how it could be interpreted as that now, but I still disagree. To say that restrictions also include the consequences of doing some thing (e.g. you receive no further binaries) implies to me that the definition would be completely up to the user. One might intentionally break the contract to obtain a one-time copy that they weren’t technically “restricted” from getting. This seems rife with ambiguouity, which from my understanding would not work well in court. In contrast, “ability to copy/distribute/modify this snapshot of code regardless of what happens afterwards” remains a guarantee.

It has been a day since I last replied, and I am still not a lawyer. I absolutely agree that input from the FSF or a GPL-specialized lawyer would be insightful.

I’m not sure you said here what you mean to? I certainly still have access to the binaries I already downloaded. After all, they’re still running on my box! And also under the RHEL contract (old or new), I’m still able to run those binaries indefinitely. But that means I’m also still entitled to the source from RH for those installed binaries since they were originally distributed and installed by RH under the terms of the GPL.

I thought you’d still have access to the source for the binaries on your box, just that you couldn’t ask for the source of newer binaries since they would have already stopped supplying those. I could also be misunderstanding how Red Hat’s systems work.

But out of all this, my biggest concern of all this by far is the can of worms RH is opening. What’s going to happen when other even less scrupulous companies attempt to run with this possible GPL loophole RH is trying to bust open?

Alex makes points I would make regarding this - I don’t see any loopholes that would violate the four freedoms. At no point would you be unable to copy, distribute, or modify code for the binaries that you have.

In the event that I’m wrong, surely the FSF would be able to handle this and find a solution if they consider it a problem; perhaps in the form of a GPLv4?

Thanks, I appreciate your insights. Was surprised that it’s not new from Red Hat.

Section 6 of the GPLv2 states the following:

Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients’ exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

The way I interpret this, nothing about your ability to “copy, distribute or modify” the code would be violated. “The Program” wouldn’t refer to the live, changing copy; only the copy that you received at that point in time which you are still able to redistribute. I point specifically to the wording of “Each time you redistribute the Program” - since newer versions of the Program are no longer being distributed to you, the terms no longer apply for these newer copies. Neither this section nor any other section forbids you from denying access to binaries.

No, I am no “free” to do whatever I want. I want to distribute that source is strictly allowed under the GPL, but then RH penalizes me for exercising that right by terminating that account. That’s a restriction. How is being penalized for doing what I’m allowed to do not a restriction?

They only said “free to do whatever you want with the source code provided for the binaries they distributed to you”, which is true? You wouldn’t be receiving any further binaries that would require releasing code to you, and you’d still be able to copy/distribute/modify the code that you got up until that point.

How about yet another angle for you. For example, I download and distribute the source RPM for gcc for the version running on my box. RH terminates my account. Now I want to download and distribute the source RPM for the kernel running on my box. How do I do that with a terminated account?

As mentioned above, you’d be perfectly able to redistribute the code associated with the binaries already provided to you. Once your account is terminated, you’d no longer have access to the binaries that you could request the code for.

Please correct me if I’m wrong; IANAL.