[moksha256]

Yeah I'm as big of a FOSS fan as the next guy on here but you really can't complain about how someone uses your code if you used the MIT License...one of the most permissive licenses in existence.

If someone wants attribution or something then they should use a license that requires that thing.

[brookst]

There’s a gap between what is legally required and what is common courtesy.

I’m under no obligation to thank someone for holding a door for me; if I fail to do so it does not mean that person should switch to a different door-holding license in the future. It just means I’m a bit of a jerk.

When lifting an entire (permissive licensed) implementation it’s good form to say thanks.

[moksha256]

You're not wrong. But the door-holding example isn't really a good one because there's no such thing as a license for door-holding.

For FOSS, on the other hand, licenses are a well-established thing. And developers have free reign to pick a license for their code and they very commonly pick MIT...totally on their own volition. Which strips them of all privileges. It's like writing a book and explicitly setting it into the public domain. If that's what you want to do, that's great, but very commonly I don't think it's what developers actually want to do.

In the world of copyright, the long-standing legal default is for the author to own their work for a certain amount of time, whether or not the copyright is explicitly claimed. Because making public domain the legal default would be utterly insane.

I guess what I'm saying here is my beef isn't with entities that choose to be jerks—that's annoying and always gonna happen to some extent—it's more with the all-too-common decision to use the MIT License. And when I see people complain about it...I understand the sentiment but I also can't help but think that the folks complaining had it coming and it was totally avoidable.

[brookst]

I knew door holding was weak, but I think the principle holds. To me, it is reasonable to release under MIT with expectation of helping lots of people, and also to expect (not require) some credit if a notable company adopts kit and kaboodle.

[azinman2]

Door holding falls under a social contract.

[brookst]

So does saying thanks. That was my point :)

[tarboreus]

I guess I kind of disagree. Some projects might pull in 50-800 deps, and then they will run on servers with utilities written by folks. Who are you supposed to thank and who not? You could literally thank thousands of people after writing a ten-line Python script.

I like that the Kagi folks stepped up and thanked you when you requested it, and I like that you wrote this code and made it available. But going around the internet trying to get explicit thanks seems more like the norm breaking here.

[_Algernon_]

You can if they don't include the copyright header: "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software."

Which, as far as I can tell they haven't done. Their MIT licence claims their own copyright. No reference to the library used in readme.

https://github.com/kagisearch/privacypass-lib/blob/main/LICE...

https://github.com/kagisearch/privacypass-lib/blob/main/READ...

Usually when using apps that use MIT licensed libs they also implement a notice in a user-facing way. Google maps for instance has a (albeit hidden) section in their settings menu referencing at least one MIT licensed library.

[Throwaway123129]

They’re pulling it in as a dependency. it’s not a fork.

As soon as you run cargo build the source code will be fetched including the original license. That’s better than a settings menu with just a license!