Re: LLM usage in Debian -- ballot options comparison

[This message contains a few but not most editorial suggestions and
corrections offered by Google GMail Clippy.  For the record, at the
instant I write these words (having already written the body of the
message), there remain ... 14 suggestions offered by Clippy in the
body of the message which I did not choose to make -- 4 "improve word
choice", 7 "make concise", and 3 "correct grammar".  And, I continue
to be a dog on the Internet which nobody knows about.]


On Thu, Jul 30, 2026 at 8:09=E2=80=AFAM Gerardo Ballabio
<[email protected]> wrote:
> Joseph R. Justice wrote:

> > Also FWIW, I didn't make my original message intentionally unreadable (=
as least in my own estimation), nor do I myself find it unreadable.
>
> I believe that what Marc was objecting to was the addition of multiple
> "Clippy does this", "Clippy wants to do that" inserts throughout the
> main body of your message.

Well, yes, of course.

> Those inserts are an annoying distraction
> from the main body and I agree that they make it pretty unreadable,
> and while making it unreadable wasn't your primary goal, your choice
> to add those inserts can legitimately be characterized as making it
> "intentionally" unreadable.

On further consideration, I'll grant that.  I can see that, at least
from a certain POV, what you say makes sense.  Again, TTBOMK, I did
not intentionally or consciously intend to make my message unreadable
(trollish in some ways, perhaps, but not unreadable), but I can see
that an outside observer might fairly and reasonably think otherwise.
(Since, after all, we do not have working telepathy, cannot see into
the minds and hearts of other people, etc etc.)

> Honestly, when I read your message, I believed you were actually
> performing a satire of "AI-assisted writing".

FWIW, it was not my intention or conscious desire to satirize the
notion of "AI-assisted writing".  TTBOMR now of my thoughts then,
rereading that message now in writing this message here and trying to
reconstruct my thoughts then, it *was* my intention to note that that
message did contain some content and/or changes suggested by Google
GMail Clippy, which would in turn meant that the message was in some
sense "AI-assisted" or partially "AI-generated", which as I've
gathered from osmosis reading mail sent to various Debian Project
lists ideally should be noted within the message for the benefit of
those who do not wish to engage with AI-generated content, because
That Is What One Does at this time and in this space.

And, I think it occurred to me that, wouldn't it be interesting to
explicitly note ALL the places and things that Clippy had a suggestion
to make concerning (I'd never actually done that before), so I decided
to do just that, and it kind of snowballed.  (As such a thing often
does.)

I actually had a, at least in my estimation, more interesting thought
and comment to make concerning all the Clippy-isms I'd documented in
that previous message, which occurred to me at some point after I'd
responded to Mr. Haber but before I read your message here.  I intend
to share that thought, but in a different place in this discussion
thread where I think it makes more sense to be.


I have a question for you, if you'd be willing to answer it.  *If* you
yourself are willing to engage with AI-hanced or -generated content
(and I respect that you might not be so willing), did you read my
followup to the Clippy-infested message containing GMail Clippy's
AI-generated synopsis of the Clippy-infested message?  And if you did
read it, did you find it to be any more comprehensible than the
Clippy-infested version?  I'm curious to know.


Thanks for choosing to respond to my message, and for whatever time
(if any) you choose to spend on reading this response to you.  Hope
you find it of some use, interest.  Be well, take care, do good
things.


Joseph