4 ms·
Couldn't you circumvent this by sub-licensing the project under the standard MIT license and then use THAT as training data? If you care what people do with yo
by OkayPhysicist 4y ago
Couldn't you circumvent this by sub-licensing the project under the standard MIT license and then use THAT as training data?
If you care what people do with your code, don't use permissive licenses. Use something with strong-copyleft like the GPL, or come up with a license that actually restricts the behavior you want.
- rognjen 4y agoIANAL so I don't know. But imagine the time investment required to do so...
- OkayPhysicist 4y agoZero? The act of licensing something to myself creates no observables, but it's something I'm legally allowed to do. So it just becomes an argument in a court case. Now, personally I would probably respect the clearly stated intent of the author, because I like to think myself basically decent. But I also try to advocate for people not undermining themselves by selecting licenses that actively permit people to do the thing they don't want them to do. If you want people to contribute back to your ecosystem in exchange for licensing your code, don't use a permissive license, use a copy-left. If you don't want someone building a business off your code, use a non-commercial license. If you want to put any meaningful restrictions at all on your codebase don't give people a carte blanche to sublicense it.
- rognjen 4y agoWouldn't you, at the very least, have to copy repo and change the license text?
- OkayPhysicist 4y agoThe license is just an agreement between the license issuer (me) and the person looking for permission to use the text (me). I don't need to publish anything to sublicense it.
- eru 4y agoNot sure if you sub-licensing something to yourself would work out? But I am sure it's just as easy for a company to spin up two shells to play this games.