4 ms·
I pretty much agree with you especially on the point that there are far more effective and nuanced ways to for the maintainer to have made their point, like rel
by dcow 5y ago
I pretty much agree with you especially on the point that there are far more effective and nuanced ways to for the maintainer to have made their point, like relicensing (imagine what would happen if the license suddenly became AGPL and everyone at big corps automatically updated without checking the new license—it would be a raucous for both legal and engineering, author could offer a non-gpl version for money and probably rake in more than a 6 figure salary… but I digress).
What I find most interesting though are the ethics. The way I interpret (1.) on this comment in the context of this thread (other people have said similar things) is that the community considers the author here to be acting ethically, just annoyingly. When someone submits dubious patches to the linux kernel and wastes everyones time, it’s unethical (kernel maintainers are the victim). When researchers submit dubious CCPA requests to people on and waste peoples time, it’s unethical (webmasters are the victim). However, when a package maintainer throws a temper tantrum and fucks up everyones’ stuff and wastes a bunch of time, they’re in their right to do so and it’s pretty much just annoying but totally okay because the maintainer is the victim of something (either corporate greed or mental health issues, or both). Our society very much puts a premium on protecting victims.
- gnubison 5y ago> Our society very much puts a premium on protecting victims. … is this a bad thing?
- dcow 5y agoNot in isolation but it’s important to understand. It can and is abused by people who know how the meta game is played. The biggest thing for me is that it pretty frequently results in an internally inconsistent worldview, and that is a bad thing.
- rezonant 5y ago> it would be a raucous for both legal and engineering, author could offer a non-gpl version for money and probably rake in more than a 6 figure salary… but I digress) For these particular packages? Very unlikely.
- dcow 5y agoThe GLP affects the entire work. If some fancy pants proprietary shop got caught publishing a release that incorporated an AGPL `faker` then a GPL request would render all the source code. This would be a legal nightmare for some places. Have you heard what happens when an acquirer finds a GPL bomb during due diligence?
- rezonant 5y agoI'm very familiar with the GPL, I just meant folks would rather find non-GPL alternatives. For a library that wraps a few ANSI terminal codes to enable color and provides a bunch of fake data for testing, I think finding alternatives (or just writing them yourself) wouldn't be hard.
- Siira 5y agoI guess the OP means that it would be too late to replace the AGPL code if the company has already published a version including it. My unexpert opinion agrees with the OP.
- newaccount74 5y agoI seriously doubt that courts would side with someone who tricks people into downloading code with a different license. Especially since changing the license of a project from MIT to AGPL doesn't suddenly revoke your rights that you had under the MIT license -- only new code would be affected by the license change.
- dcow 5y agoYes the solution would be to fork off at the last MIT/Apache licensed version. I’m no lawyer either and would actually be interested in seeing the court case play out.