Re: GR: Ban LLM contributions from Debian

Gard Spreemann <[email protected]> Tue, 28 Jul 2026 15:41:34 +0200

Newsgroups	gmane.linux.debian.devel.vote
Message-ID	<[email protected]>

Matthias Urlichs <[email protected]> writes:

> On 27.07.26 10:24, Gard Spreemann wrote:
>> a person is
>> still writing a precise set of instructions when they write C or
>> JavaScript
>
> I would contend that there are heaps of edge cases in both C and JavaScript
> where the intended meaning and/or the result is *not* precise[ly specified]. At
> all.

Sure. But there are large, well-specified portions of those
languages. We agree that staying away from edge cases in e.g. C is no
easy task, but it is certainly possible (at least in principle).

With the "source language" being natural prose, however, all bets are
entirely off.

> Yes, translating a more-or-less-correct/complete specification to code is (for
> the most part) not going to be one-shot and/or deterministic when you give the
> job to an LLM, but the same thing happens when you hand the spec to a junior
> programmer.

But now you're comparing LLMs to junior programmers! My issue is with
comparing the role of LLMs to that of compilers or calculators. We
definitely interact with junior programmers through natural language
already, so nothing changes there. My point is that we interact with
compilers and calculators through *formal languages*, and that therefore
it's not helpful to compare LLMs to those.

 Best,
 Gard

signature.asc (application/pgp-signature, 857 B)