The products are largely illegal to sell in the EU. They regularly fail health standards and safety standards because those are expensive.

That’s why the customer is the importer instead for these apps. You’re allowed to import stuff that is illegal to sell to consumers after all.

Oh, I didn’t consider the “any other” aspect.

Welp, I can still register several distinct legal entities in different EU countries, can’t I? Maybe one could be a “Taking every EUPL work on the internet and relicensing it under LGPL as a service” company. That’s bound to make some money from SaaS companies if it would be this easy to purge the EUPL terms.

Though the “ideology” quote is a bit awful, I’ll give you that. The matrix itself does look fairly neutral though, especially with this part under “Discussion of Linking”:

We made the assumption that, by selecting a Gnu license, licensors follow the FSF position and want to consider that most cases of static linking create a derivative.

I’d also argue the 27 legal systems might not be too relevant since copyright law is generally equal in the different member states. The remaining legal issues (e.g. warranty) are irrelevant for interoperabilith between licenses. Also, most importantly, there are only 24 languages in the EU.

If the official guidelines are recognized by courts as legally binding then I think the EUPL is superior to even the AGPL. Sadly that remains to be seen due to the lack of EUPL projects out there (and the lack of corresponding lawsuits).

A lot of the discussion seems to be US centric though, such as this quote:

Courts have a strong tendency to read licences on their faces, that is to say, they look only to the text of the licence to determine rights and obligations. In some cases, courts have explicitly refused [link to https://creativecommons.org/2017/07/06/cc-amicus-brief/] amicus briefs from the authors of the licences who wished to clarify the intended interpretation of their text.

Since this links to a US court decision, I believe the first part also refers to the US. As little as I know about EU law, what I do know is that it is often intentionally vague to ensure the spirit of the law cannot be violated. That seems to be the same for the EUPL where vagueness is preferred over concrete definitions which may hold up even worse in (EU) court due to the limitations on copyright law.

The weaker licenses don’t even mention SaaS so they aren’t in conflict with the EUPL there. As such, the EUPL’s copyleft protections should still remain for code published under it, even when used in an MPL project.

I’m imagining it as a pseudo-dual license permission and the EUPL as some sort of Affero-LGPL.

Though SaaS vendors would probably comply maliciously and only send EUPL code snippets back when requesting source code, if they are used in a differently licensed project.

I am not an expert in copyright law, which is what these licenses are based upon and cannot analyze the text.

Still, couldn’t you make it even more straightforward by forking twice yourself?

1. Take the original EUPL code and fork it under the LGPL
2. Take the LGPL code and fork it under the LGPL
3. This second fork has all EUPL conditions removed

I’d by surprised if the license authors did not consider this. Lawyers wrote this with consideration of EU law after all, not some laypeople.

If I had to guess: Any inclusion of EUPL code in another project would have to be marked as being under the EUPL. This is solely to inform anyone who wants to fork this section and distribute the code in form of SaaS to abide by source code requests.

It’s like an EU variant of the AGPL whose many conditions about linking apparently don’t hold up in EU court. The GPL’s are all primarily considering US copyright law after all.

I can only answer the first question:

The great thing about the EUPL is: Its terms prevail if the other license does not have conflicting provisions. Any code licensed under EUPL will keep the SaaS restrictions:

However, according to the EUPL, the compatible licence that is applied to a derivative work will prevail “in case of conflict” with the EUPL. For example, when the EUPL licensor has its seat in Germany, the applicable law is German and the court is Berlin, but if the code is reused in a French project distributed under CeCILL, the French law will be applicable and the competent court will be Paris. But on the strongest open source EUPL provisions, like the coverage of SaaS and the obligation to publish and share the derivative source code, none of the listed compatible licences enters in conflict with the EUPL: for example, they may not “impose” code distribution in case of SaaS distribution, but they do not prohibit it. Therefore the EUPL obligations are persistent.

https://interoperable-europe.ec.europa.eu/collection/eupl/matrix-eupl-compatible-open-source-licences

That’s wrong. The EUPL’s strictest provisions remain in effect even when used in other licenses:

However, according to the EUPL, the compatible licence that is applied to a derivative work will prevail “in case of conflict” with the EUPL. For example, when the EUPL licensor has its seat in Germany, the applicable law is German and the court is Berlin, but if the code is reused in a French project distributed under CeCILL, the French law will be applicable and the competent court will be Paris. But on the strongest open source EUPL provisions, like the coverage of SaaS and the obligation to publish and share the derivative source code, none of the listed compatible licences enters in conflict with the EUPL: for example, they may not “impose” code distribution in case of SaaS distribution, but they do not prohibit it. Therefore the EUPL obligations are persistent.

The EUPL 1.2 also allows directly relicensing to GPLv3, I don’t know where they got the idea you cannot do so.

Large parts of the GPLs are also void (e.g. providing “zero warranty/liability” and provisions regarding dynamic linking) according to EU case law.

The EUPL is effectively a interoperable strong copyleft license. It tries to prevent license incompatibilities due to “virality”.

I strongly recommend reading these two articles, they are much more accurate than the FSF’s:

https://interoperable-europe.ec.europa.eu/collection/eupl/licence-compatibility-permissivity-reciprocity-and-interoperability

https://interoperable-europe.ec.europa.eu/collection/eupl/matrix-eupl-compatible-open-source-licences

“Not feeling well? I’ll give you something to not feel well!”

You might want to avoid !lemmyshitpost@lemmy.world

Mostly because Apple’s update policy is superior to A LOT of Android companies. OEMs are really slow when patching known vulnerabilities.

Quick study I found when trying to find evidence:

https://www.ftc.gov/system/files/ftc_gov/pdf/16-Acar-A-Device-Centric-Analysis-of-Android-Security-Updates.pdf

Example from that study:

Compared to the top three OEMs we examined so far, Google is the one with the most stable support behavior. All of the Pixel devices receive monthly security updates without any delay or missed SPLs [Security Patch Levels]

It’s utterly insane this is noteworthy. Not delaying security updates for KNOWN vulnerabilities should not be exemplary.

It’s literally free fame.

Would you have ever heard of this local African politician if he didn’t have this name?

Yeah, I don’t mind it too much. Especially because tax evasion is a far bigger harm to social welfare systems.

As long as he actually needed that money and wasn’t some already well-off fucker that wanted to hoard even more money that is.

Yeah no, that’s not how a pension system can be sustainable.

Some people must get less than they paid so that others don’t have to starve to death because they live to the age of 100.

And that’s honestly OK. The same way someone who never gets sick in their life has to pay for all the chronically ill.

You misunderstand, the project aims to have them eradicated by 2050, not starting to eradicate them in 2050. The project isn’t called Predator Free 2050 for no reason.

limits the harm

Harm reduction isn’t really sufficient when every single feral cat kills presumably hundreds of criticially endangered birds per year, is it?

I really like bash when dealing with even somewhat advanced scripting. Like the 300 LOC scraper I have written over the past two days which horribly parses HTML files using grep | sed.

It’s genuinely so much more fun to do this with Bash than, say, Python. I have once written a scraper using Beautifulsoup and I have no desire to do so ever again.

Honestly, only Haskell manages to beat Bash in how satisfying it feels when you manage to get something working well.

Gambling addictions have very high suicide ideation and suicide attempt rates. The exact numbers seem ti vary from study to study but I haven’t found any with percentages below double digits.

Gambling addictions are not uncommon either, with 1-3% of the general population being affected [Wikipedia]. Risk factors include starting young and online gambling by the way. Guess who encourages both?

If I had to guess, thousands of people have committed suicide at least partially due to Valve. “Partially” because case gambling could’ve been the entry into other forms of gambling that cemented gambling addictions.

Disagree slightly, human created content can have intent but doesn’t automatically have it.

A corporate ad does not have any artistic merit besides grabbing as much attention as possible. Actually creative ads where some thought was put in are the very rare exception.

The same goes with a lot of pop music today. I cannot speak too much about English language pop but German pop is nothing more than fast food. See the Wikipedia article of Menschen Leben Tanzen Welt.

Or take a look at video games. How much artistic effort is put into AAA games? Maybe someone spent 40 hours making the lootboxes as satisfying as possible to open but that’s probably where the most thought was put in.

And movies? Aren’t Disney’s recent “live remakes” of their old, successful animated movies anything but CGI slop? Sure, I admit it takes a lot of effort to make and animate all these models. Just like it takes effort to shit when you’re constipated.

Honestly, the only thing distinguishing AI from megacorp content is that the latter has more consistency and fewer “mistakes” than the former. The sole intent of both is making money.

Ah, thanks. Searching for “national DPA” just yielded some “National Data Privacy Agreement” or - when adding the term “journalism” - the German press agency DPA.

That’s your article. I was talking about the articles OP linked which do not contain this phrase anywhere.

I’m not about to discuss an article that is completely irrelevant for OP’s point.