jamesravey

Seems like the lawyer thinks that AI models deliberately jumble the Disney logo rather than specific text/artifact/logo generation just being a weakness of these types of models. (He’s wrong, he’s attributing intent to something janky/buggy)

Came for this joke

I used to agree until I saw corporations starting to fork open source projects to run them internally like the “I made this” meme.

If I spend months or years of my life toiling over a project and license it permissively with MIT or such, they can just swoop in one day and take it for free and be like “thanks, we’re going to make mega bucks off your code and give you nothing” (and yes this does happen https://www.elastic.co/blog/why-license-change-aws).

No, screw that! I’m gonna make my stuff AGPL and those guys can damn well pay me for my time of they want to use my stuff or more cynically, do it anyway or go and reimplement it themselves in-house knowing damn well I can’t afford an army of lawyers to actually do anything about it.

Counter counterpoint: Often frontend js code is minified so that it is smaller and more efficient to transfer to the browser. For FOSS projects you should still be able to get access to that code, unminified, from the project git repo. In the same way desktop apps often ship as binary executables but you can still see the code that was compiled to build them if you find the source repo.

It does make things harder to debug for an average user but it makes it faster/more efficient to run for most end users (in the case of the desktop or phone app it makes it possible to run without needing compiler toolchains that mom and pop likely wouldn’t be able to grasp).

The key thing isn’t that what the end user’s computer runs is readable and editable but whether the code used to build that artifact is available easily and what restrictions there are on editing and redistributing that code.