• 1 Post
  • 18 Comments
Joined 1 year ago
cake
Cake day: April 26th, 2024

help-circle
  • I’ve been thinking about this topic along similar lines. There’s a race to the bottom, and someone’s fully licit corpus is just going to end up being worse (in user perception, not in whatever gamed metrics/benchmarks are de rigueur) than someone else’s illicit corpus, even if they are broadly similar at a glance.

    There are also likely other, smaller factors that make this difficult. I expect my small body of public work with permissive licensing has been an input to one or more of these things at some point, but I would not be surprised if the fidelity of provenance were not maintained consistently, because that is not easy to do, and it doesn’t buy you much if you don’t intend to provide attribution.