interoperability == API standardization == API homogeneity

standardization != monopolization

Yes, this is why I was mentioning the AGPL being a better fit when the software is a service, for the reasons I explained.

I think many hobby developers also see “hobby” developing as part of their career, so they would happily try and have their hobby align with future employment possibilities. Since companies avoid GPL, those devs will rather choose a license that is more attractive to those potential employers when they see their portfolio.

Thunar is a much better alternative, in my opinion.

To me, the problem is not really so much about “locking people in” (it’s also unclear what you mean by that, if they were already using that ecosystem before using uutils aren’t they already locked in?)

To me, the problem is how the MIT removes legal protections when it comes to ensuring accountability to changes in the source… how can I be sure that the version of uutils shipped with “X Corp OS” has not had some special sauce added-in for increased tracking, AI magic, backdoor or “security” reasons? They are perfectly free to make changes without any public audit or having to tell their users what their own machine is doing anymore.

If you are using a GPL library that is statically linked to code with a different license the result is one binary that has inside both GPL and other license code, which would not be allowed under the GPL terms, because it requires that the binaries that use the source code must have their source code available in full (including other source and modifications that are part of the same binary).

The only case in which you don’t need to provide the source for GPL software is if you don’t actually distribute the binary to customers… private binaries do not have to be published with their source, as long as you never made the binaries public and never gave it to anyone else. Only when you give it to someone you need to provide the code.

This allows for a loophole in which if you are providing a service, then you can run the software privately in your private server without sharing the source code to the clients using the service, since they do not really run the server program although they indirectly benefit from its results. This is why the AGPL was created, since it has a clause to force also those offering services to make the source of the server available to the users of the service.

Yes. Did I deny that?

Can anyone confirm or deny if what I said is wrong?

I get downvoted for stating one advantage an AGPL fork would have, and yet nobody seems to be disagreeing with what I said… *shrugs* 🤷

developers should be able to license code that they write under whatever license they like

What if they choose a license that limits the freedom from all other developers to improve that copy of the software? is allowing a developer to restrict further development actually good for the freedom of the developers? Because I would say no.

The spirit of the GPL is to give freedom to the developers and hackers (in the good sense of hacker). The chorus of the Free Software Song by Stallman is “you’ll be free hackers, you’ll be free”.

the GPL restricts freedom without adding any that MIT does not also provide

“Your freedom ends when the freedom of others begins”

The only “freedom” the GPL restricts is the freedom to limit the freedom of other developers/hackers that want to edit your software. This is in the same spirit as having laws against slavery that restrict the “freedom” of people to take slaves.

Would a society that allows oppression (that has no laws against it) be more “free” than a society that does not allow oppression (with laws to guarantee the freedom of others is respected)?

It doesn’t even say “you should use the GPL”

That sounds a lot more confrontational and less diplomatic than the Github issue, though.

The ticket was actually indirectly asking it, by explaining the potential problems with non-copyleft. They just added “If you plan to carry on…” to introduce a compromise, which actually allowed for at least some minor change to be made, and made it clear that the different license is intentional and not just for lack of awareness (which implies the devs have no intention on switching).

it says “you MUST say GNU doesn’t agree with you”

Somehow you added the “MUST” to this sentence, but not to the first one… even though the github issue did not say they MUST, instead it even used the word “please” and appealed to having some deference to the GNU coreutils.

At least this issue managed to get a change through for clarity… I don’t think you would have gotten anything at all with the other approach.

Note that AGPL can take changes from MIT but MIT can’t take changes that are purely AGPL without following the AGPL.

So, as far as I can understand, any improvements done to the AGPL version cannot be carried over to the MIT version (without very painful and careful re-implementation / re-engineering). That alone would be a big advantage to the hypothetical AGPL fork.

It would be a bit of a legal nightmare, since it’s theoretically possible that, even without really knowing it, the same feature might be implemented the same way in both forks separately, and the MIT devs might have no sure way to prove they did not copy it. So this would be like walking on eggshells.

I largely agree, that’s why I was saying that I’m skeptical that all this will amount to anything substantial.

The will for independence exists in the EU, the problem is that the politicians don’t have the balls for it and they would rather push to maintain the status Quo in all the things that matter. Instead they focus on small things that appear good on paper but don’t really amount to anything. See for example the DMA and all it’s promises of forcing big corporations to bend the knee and stopping monopolies… even when a policy like that is written, it is hardly ever properly enforced. Has any company gotten any serious trouble for not implementing GDPR properly since it was introduced?

There has been a will towards more independence for a long time. Trump was just an extra push (and I’m still not convinced even that will be enough… all these initiatives sound good, but past experience has made me skeptical they will really amount to anything substantial).

But I don’t see it necessarily as anti-american. It’s more like we do need to cultivate local products and services more. Europe has for a while been falling behind in a lot of areas, combined with an aging population and an energy crisis, we really need to try and develop internally if we want to keep ourselves afloat, otherwise I’m not sure we can maintain a stable situation.

Yes, although Ctrl-M would be the “Carriage Return” character (\r). For the “Line Feed” newline character (\n) the Control combination would be Ctrl-J. Both of them would normally produce a new line when you press them on most terminals.

That’s why if you open in nano/vim a file with Windows style EOL (/r/n), you might see a strange ^M symbol at the end of each line.

To be more precise, it’s the “EOT” (end of transmission) control character, the 4th symbol in ASCII, from the non-printable character area.

This is true, but then why not base it off Guix (the GNU distro)? …I’m sure Fedora is full of binary blobs and not-so-free software.

They could still add extra software and blobs sourced by the EU if they needed it… and I think doing that would allow it to carve itself a niche rather than sticking a white label on Fedora and call it something else. I don’t see a lot of value on this over just using Fedora directly.

banned from connecting to wifi networks based on thier mac address

MAC addresses are easy to spoof in a Linux PC though. So you should be able to workaround that limitation if it’s ever a problem.

In fact, in theory you can also spoof the IMEI in a phone via software, it’s just that the software in phones is generally not as open and it would require rooting and some tweaking that not everyone is comfortable doing.

Another thing is that this standard does not allow setting an absolute size… you can only use fractions, for example, 2/3 of normal size, which can still be relatively big if your normal size is big.

I also expect making things bigger for emphasis (or adding headings) would likely be more common than making them smaller (that’s what I hope at least). Outside of things like mathematical notation where superscript/subscript might be useful (see this comment for some examples).

The problem with the web is that some websites use absolute units that might not scale well (like px) to define the sizes.

A compositor is normally a component in a DE, not a DE on its own. For it to be a DE in my book the “standalone” installation needs to, at minimum, provide: a launcher to execute apps, a toolbar/statusbar, and maybe a terminal emulator (or at least call some generic wrapper to automatically hook into one, something like xdg-terminal-exec).

I mean… openbox is used in X11 desktop environments like LXDE… I don’t see why labwc (essentially wayland’s openbox) should be treated like it cannot be a component of one.

And river has almost as a mission statement to become more of a framework than a DE on its own… they even have the goal in the long term to remove things from it and instead expose more to the commands/API to make it more modular… it’s definitely not something intended to work standalone and they expect people to develop third party layout generator programs.

Maybe sway is the one in that list that might be the most “standalone”, since it does have swaybar built-in… but the default configuration still expects you to provide at least something like dmenu to use as launcher, as well as making sure you have your terminal, etc, since it does not list them as specific dependencies of the sway package, so officially they aren’t really part of sway as if it were a DE suite.