This was meant to be detached from commercial use, just on “training models”.

deleted by creator

Do you think it’s a good idea to adapt licenses to be able to disallow training models on the source code? Do you think this could be enforced? If so, how?

Yes exactly, they’re similar. I guess there’s a universal desire for tree like data stores.

If you’re using a Linux distribution: are you familiar with gsettings or some equivalent?

Haha and Reddit ofc asks you the same. But I have the feeling that it’s more widespread practice nowadays. I haven’t seen this in Europe that strongly yet though. Seems like there are more proective data regulation laws.

What communities did you find so far?

!ShitAmericansSay

Apply Hitchen’s Razor against it:

https://en.wikipedia.org/wiki/Hitchens's_razor