@david_chisnall this observation reminds me of @glyph 's article here: https://blog.glyph.im/2026/06/adversarial-communication.html

@david_chisnall On the other hand... If the MGMT give you a bs task to do, that will be judged by weight not merit, a locally run LLM can save you hours... They get what they deserve... 😆

@david_chisnall
Quality:
cognitive value to the audience,
divided by processing effort imposed ?

Sperber and Wilson experimented with this kind of schematic, and the implicit assumptions of polite communication, in Relevance Theory (1995)

@david_chisnall

That's kind of second order rudeness. Because on the surface LLMs are like kind and well educated. But in human perspective it's at least toxic behavior. Or in other words: superficially good and friendly transactions are in fact passive aggressive behavior. Please forgive me my new invented term "second order rudeness".

@david_chisnall thank you. It's amazing how many people don't seem to understand this simple concept. I saw a thread yesterday where someone was complaining they were banned from Wikipedia and their appeal was rejected because they used an LLM to write it. They didn't seem to grasp how rude it is to waste the time of an unpaid volunteer and thought it shouldn't make any difference; it absolutely does

@david_chisnall

Let me tell you a small story: Once, in University, Pre-LLM-time, I handed in an essay. I got a bad grade. I had worked very hard for that essay, but the superior thought I had been sloppy. And that sloppiness indicated I had not worked hard, and so the bad grade was justified. I didn't know what to say, but after the talk, I had a meltdown and cried. What he didn't know, I do have a spelling problem. I cared a LOT and worked very hard and the mistakes he found were just the ones that were left over after all the work and effort I poured in.

Therefore, I would strongly suggest to not make tool-usage (llm or spellchecker or thesaurus etc) an indicator for politeness. Nor should the quality of the result be an indicator for politeness or effort. The intent, which is unlike harder to measure, would be, in my eyes, the best indicator for politeness.

@shapr “impolite behaviour stems from assuming that your time is worth more than other people's” is a special case of my theory of evil: evils is when one person pursues their own goals heedless of the loss or suffering it brings to others.

The behavior Chisnall describes is evil on a very small scale, but it's different in degree, not in kind.

@david_chisnall @tforcworc When I see the acronym LLM, roll the "Naked Video" theme tune, and cue the Baldy Man.

@david_chisnall great framing. In a similar way, I always use the example of LLM use as equivalent to dumping your trash out the window into the street. Sure, the LLM user saved time and hassle, but made the situation worse for everyone else in time and environment. Now someone has to clean up an exploded bag of trash in the public.

@david_chisnall Corollary: rules of politeness originated as embedded shorthand to balance peoples' time and attention?

Therefore old fashioned and outdated rules are predicated on shorthand whose need no longer exists?

For example, social status used to be essential information in daily life, with titles used as part of normal communication to convey what level of attention was socially appropriate.

Now many titles are a relic rather than an important indicator. We still use special honorifics in formal circumstances ("Your honor" in a courtroom), but it's fine for me to address my Christmas card to my friend the academic with just her name even though a student might be expected to use the full title of "Professor".

@david_chisnall

Heh:

"If someone uses an LLM to write some code that looks like it works and expects you to put effort into reviewing a PR, again, they're saving work for themselves at the expense of your time. That's rude."

Heh. If you can't test your own code, then you're not a real programmer. You're not a programmer and I don't want to work with you. Ugh.

@david_chisnall That's such a fascinating definition of rudeness! I never thought of it that way.

@david_chisnall Also works the other way round: if person 2 is much better at doing task B than person 1 is, it makes sense for person 2 to do it so person 1 can concentrate on task A instead of doing task B badly. (We pay a cleaner because they're better at cleaning than we are.)

@david_chisnall For unclear reasons, I have been seeing things as externalities everywhere recently. This is one of the biggest ones.

@david_chisnall Wow, cheers, I'd never made this connection between narcissistic time-efficiency and rudeness.

Been thinking about the self-aggrandising complaint "I don't have time for that" which I've heard recently (from an LLM user, and about various activities), just in terms of my own perspective of, "well, yeah, I like to be time-efficient too, putting in the work up-front to be so". Merely had the complaint down as an attempt to co-opt the principle of time-efficiency, but glad to connect it to rudeness.

@david_chisnall "hey're machines optimised for creating rudeness." yes. good observation !

@david_chisnall This all boils down to:" If you didn't care enough about this to write it, why should I care enough to read it?" To make it worse, all too often the sender doesn't even care enough to read it either, so you just get the unfiltered response to some prompt. And I agree, that is a deeply disrespectful waste of my time.