31–40 of 110 posts
This seems like a storm in a tea cup to me. Distros are made of thousands of packages. The maintainers are supposed to get the license details correct but sometimes humans error. When these errors get buried deep into things and no one is auditing them, they go unnoticed. No need for anyone to lose their mind over that. It's life.
At least it's just unlicensed, not something worse like AGPL /s Yes, it's a joke and it's funny when several companies, including Google [1] are very scared of that license. [1] https://opensource.google/documentation/reference/using/agpl...
> So, to recap, we’ve got distributions of a go package that includes entirely unlicensed code, which I will remind you does NOT mean you get to just use it because you feel like it.
It also does not automatically mean you do not get to use it. It may be illegal and it may not be, depending on multiple factors. It's prudent to be conservative and require proper licenses like the policies of the distributions state but that does not make it illegal. What's considered "fair use" and what that entails differs a lot from place to place.
I'm fairly certain that even in jurisdictions where stuff like this would be problematic, a judge or jury would not rule guilty. Intent matters. Now, if the copyright holder contacted the distributions, made them aware of the situation, and asked them to stop distributing it or amend the license situation, and the distributions don't comply in a timely manner, that would be a completely different story.
IANAL anyway and I fully agree slip-ups like these are not a good look and should all be addressed but I have a strong hunch that the main claim in OP is false.
Author of ripgrep here. Maybe, umm, don't depend on auto-detection tools as the ultimate source of whether something is appropriately licensed or not? encoding_rs is clearly licensed, otherwise I wouldn't have used it in ripgrep: https://github.com/hsivonen/encoding_rs Take it up with the auto-detection tools. Or at least do your due diligence before claiming ripgrep is using "unlicensed" code. The Debian copyright n…
The encoding part is one concern, but what about ryu? Is “Apache-2.0 OR BSL-1.0” not accurate? Neither of those showed up in the grep for license lines? Sure, BSL is an obviously free license that is basically just a wording variant of MIT, but it is not showing up in that grep. Note: I’ve not actually read the copyright file, So perhaps It is clearly accounted for there, but is just not formatted in a way that the g…
Indeed they are not a lawyer. A lawyer would know that this is not illegal. It violates the rights of the copyright holder, but that’s not a crime.
You make it sound like copyright infringement is never a crime, when that isn't true at all. 18 U.S.C. § 2319. Now, the commercial use or intent of the software in question isn't analyzed in the OP (or in your comment) and I haven't dug into it. In this particular case it may or may not meet the elements of the five-year federal felony. But your comment is far too quick to dismiss the possibility. (Additionally, as a…
I know the US is trying (and often getting away with) unilaterally setting rules on the global stage but let's try to not play along.
Author of ripgrep here. Maybe, umm, don't depend on auto-detection tools as the ultimate source of whether something is appropriately licensed or not? encoding_rs is clearly licensed, otherwise I wouldn't have used it in ripgrep: https://github.com/hsivonen/encoding_rs Take it up with the auto-detection tools. Or at least do your due diligence before claiming ripgrep is using "unlicensed" code. The Debian copyright n…
The encoding part is one concern, but what about ryu? Is “Apache-2.0 OR BSL-1.0” not accurate? Neither of those showed up in the grep for license lines? Sure, BSL is an obviously free license that is basically just a wording variant of MIT, but it is not showing up in that grep. Note: I’ve not actually read the copyright file, So perhaps It is clearly accounted for there, but is just not formatted in a way that the g…
It is not included in the Debian copyright file. I suppose this is because the library is not included in the source package, and the BSL does not mandate notice preservation for a binary output. And Debian policy for the copyright file is that only notices or info that must be preserved are required to be present in the file.
Earlier quoted context omitted.
edit: I stand corrected
Speeding in the US ̶g̶e̶n̶e̶r̶a̶l̶l̶y̶ ̶i̶s̶ (edit: can be) a criminal summary offense.
Indeed they are not a lawyer. A lawyer would know that this is not illegal. It violates the rights of the copyright holder, but that’s not a crime.
You make it sound like copyright infringement is never a crime, when that isn't true at all. 18 U.S.C. § 2319. Now, the commercial use or intent of the software in question isn't analyzed in the OP (or in your comment) and I haven't dug into it. In this particular case it may or may not meet the elements of the five-year federal felony. But your comment is far too quick to dismiss the possibility. (Additionally, as a…
For anything to be criminal there either has to be intent or the crime should be strict liability tort where intent does not matter.
It is arguable whether copyright infringement is strict liability tort in practice or not - there are arguments on both sides, so context of the particular case or similar cases that can set precedence matters .
So I wouldn’t be quick to just to say even say illegal let alone criminal.