So if they had left that line in, everything would be cool?

To me, licenses like MIT or BSD pretty much imply "do whatever you want with this" I know it's not exactly that but if you really care to keep some control over what others do with the code, you need a more restrictive license (and even then people are still going to copy it, especially in the LLM era).

You can "do whatever you want with this code", but there's a catch: you have to give credit to the original author. You might not care about the credit, but lots of people care.

You can't just cherrypick the things you like about a license. All of the conditions of the license apply.

You're thinking about what people can do with the code, like copying, editing, and distributing. This is not it. We're talking about giving credit to the original author, as per the license.

> So if they had left that line in, everything would be cool?

It certainly would be better.

Forks tend not to have -perfect- relationships and tend to cause a bit of mutual annoyance. But attribution is important-- it's the most basic step.

When this maintainer is asked how the projects are related, it'd sure be nice if both projects are telling the same story, instead of one illegally lying about it.

Well, it’s the difference between plagiarism and attribution. If your goal isn’t money but a bare minimum recognition for what was your work vs someone else taking credit for it, yes it’s enough.

A lot of open source software operates on the same principles of academic research. Most academic research is considered freely available, and other researchers can generally use your work as they please, so long as they cite the original author.

In this context, not "citing the original author" in the copyright statement, labeling the repository as a "fork" on GitHub, clearly crediting the original author in a way that clearly describes the fact that a significant portion of their code is used in the new project isn't just a violation of the license, it's plagiarism.

So in that sense it could be better potentially.

Yes, it would be cool, and it's the usual way to do these things. You can license code under a more restrictive license, and clarify licensing by adding an extra section to the main license, adding the license to a subdirectory, or adding license headers to the individual files.

Whether the MIT license is the right one to choose is probably a different debate.

Microsoft credited the original author and project in the README, which is far more visible than a hidden copyright line somewhere in the terms and conditions. If attribution was what he wanted he should be really happy about he outcome, but clearly that's not what this is about. He is simply pissed that Microsoft used his project.

If they had been factual I the credit I'd agree. When it's actually a fork, why not just say so. "This project is a fork (or based off) Spegel. Thanks to the authors etc" Maybe with a rationale why they forked it. You know, just common decency...

Because it's not a fork. They copied the API and like 100 lines of unit test code.

Maybe not a fork, but the author writes "It looks as if large parts of the project were copied directly from Spegel without any mention of the original source".

So they are exaggerating?

Still, it's illegal for Microsoft to remove the copyright as per the licence.

If I owe you $100 by contract, I can't just pay you with 1 ton of steel slab delivered to your garage and argue that this is worth more and therefore you should write the debt off.