On Tue, Aug 11, 2026 at 07:57:59PM +0200, Matthias Geiger wrote:
However, the lintian override must be written manually (for this
proposal). As it should anyway, since an LLM can never "know" what
might be appropriate and what not.
Simple example:
Lintian complaining that a package has no watch file. An LLM might try
to "help" by writing a d/watch file. However, without context, this
can be totally wrong. If the package has no VCS upstream or isn't
developed anymore, a watch file makes no sense. A seasoned contributor
will know that and ignore/override said warning; an LLM can't.
And if I decide as a human that the LLM's suggestion is fine and dare to
insult my fellow DDs by not typing the identical lintian override into
an editor but instead cut&paste the LLM's suggestion?
What if I allow the LLM to commit to my repository and then decide that
the commit done by the LLM makes sense? I _could_ git rebase
--interactive and remove the commit if I don't like the contribution.
An LLM "offering" to override tags which point to underlying package
issues, such as shared libraries not correctly split into their
respective packages. Arguably, a experienced contributor will spot
that,
but a new one might believe the machine.
I am an experienced contributor. Am I allowed to save myself some
minutes of time by taking suggestions from the LLM?
Greetings
Marc
--
-----------------------------------------------------------------------------
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany | lose things." Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature | How to make an American Quilt | Fax: *49 6224 1600421
