>First, let me proclaim by bias: I'm a Free software fanatic. I do not ever want to run software that I can't read and, if I want to, modify. I just won't do it.
I can relate to that. I guess you're are also having problems developing AI, because it is next to impossible to set up a full FLOSS AI stack [then I have to use proprietary stuff - conflicts of conscience ensue]
>Free software started when RMS wanted to fix his printer and Xerox said, "No."
cf. James Mickens' USENIX Security '18-Q keynote speech: "Why Do Keynote Speakers Keep Suggesting That Improving Security Is Possible?" https://www.youtube.com/watch?v=ajGX7odA87k
Anyway, they're just a pile of linear algebra and a massive pile of Other People's Data, eh? (And piles and piles of hype, my god, so much hype. Even in the "academic" papers!)
The difficulty would lie in amassing that data, not in developing software. You would have to solve that "conflict of conscience" first, no?
But more to the point: none of my goals can be advanced by talking computers.
- - - -
> Is there an AI stack that RMS would approve/use?
I have no idea. You could ask him if you're really interested.
I can relate to that. I guess you're are also having problems developing AI, because it is next to impossible to set up a full FLOSS AI stack [then I have to use proprietary stuff - conflicts of conscience ensue]
>Free software started when RMS wanted to fix his printer and Xerox said, "No."
Is there an AI stack that RMS would approve/use?