Re: GR -- Allow AI-Assisted Contributions

On Sat, Feb 21, 2026 at 04:56:44PM -0500, Theodore Tso wrote:
>Exactly.  There are some humans who might code something that takes a
>little from example A, some from example B, and some from example C,
>where examples A, B, and C were things that said human may have seen
>years ago, and where they might not even have concsious memories of
>having seen these examples.

Or just pullling up the design patterns book from 20 years ago. I don't 
think we ever had a discussion about taking boilerplate code from a book 
(which is clearly unacceptable as well when you say we don't want LLM 
contributions).

By the way, my boilerplate code is not only at least one order of 
magnitude worse than the same code coming out of an LLM, it also takes 
me ten times as long to create.

I would probably seek a project to contribute to that does allow me to 
make more efficient use of my time and talent than requiring me to 
regurgitate boring boilerplate code myself manually.

>Now, there are *some* LLM's that might do that last thing, which is
>clearly wrong, regardelss of whether it is done by a human or an LLM.
>But they could also do the first example where they are merging
>examples from multiple sources to create a new implementation.

The sad part of that thing is that I am pretty sure that the (few) 
models that are useful to aid in software development today, at least 
the ones you dont have to pay for, don't care about this aspect of their 
input data, and that a model that is documented to just use dfsg-free 
input will be vastly less useful than Claude, for example.

Greetings
Marc