> You are fine with the policy because it reinforces your fundamental belief (I
> assume) that LLMs are bad, horrible, corrupting, and society damaging
> tech-bro inventions.
No, my beliefs are more nuanced than that. And the policy does not reinforce my
beliefs. It is mostly void of all my concerns I have about LLMs, like the
erosion of taste, values and motivation.
> I am not, because it tells people that they are less welcome if they use LLMs
> in any form ("we strongly prefer"), I object especially to *P3*'s wording in
> the document.
Yes, I absolutely think we should prefer human written code. Otherwise I do not
see how we would be able to protect people who:
- cannot use LLMs due to their political beliefs (e.g. the negative effects on
the environment)
- cannot use LLMs due to their economic situation (frontier models are not free)
- cannot use LLMs effectively due to their personal characteristics (although I
believe LLMs have devastating effects on most people there's certainly groups
that are more vulnerable and have to naturally avoid them or who simply do not
derive joy from using them)
All these people now need to worry they will be left behind by the community,
because they can't keep up with those "10x" engineers who are burning tokens.
A "neutral" stance would in fact not be neutral... and it would cement that the
primary thing we care about is patches and productivity (just as Linus declared
on the linux ML). If that is the view of the project, then it leaves a bitter
taste in my mouth that I probably won't be able to get rid of.
_______________________________________________
ghc-devs mailing list -- [email protected]
To unsubscribe send an email to [email protected]
