I think I have autism. Anyway this is how I tend to organize thoughts and relay them. If it looks like AI then I don't know what to do about and is kinda scary really.

▲

dsego 5 days ago | parent | next [-]

I guess it's more common in comments to read someone's opinion on different parts, whether they agree or disagree, a related anecdote or personal experience with the topic, or a remark focused on a specific detail that stands out.

▲

tgv 5 days ago | parent | prev [-]

It doesn't matter how you organize thoughts, it's about how other people read your text, and AI has fucked up our perceptions.

First your praise the talk. That sounds AI like. LLMs are trained in the annoying American way of starting with something positive, even if it's irrelevant, or isn't meant. You're probably somewhat conditioned to do the same. But in this forum, those comments are not encouraged. The upvote button should be enough to express that.

The rest reads like a summary, also an LLM feature, but nobody asked for a summary, and you're not announcing that you want to give one. It sets the reader up for some conclusion or evaluation, which never comes.

There's no personal thought, except for the praise, anecdote, criticism, or supplementary information. If all you wanted was to recommend the talk to people, I think an effective way to do would be something like

> I liked the talk. So much that I didn't know about the history of OOP, or how ECS (Entity Component System) could have been a competitor. Recommended, but it's a bit long though.

Not that that will get you a lot of upvotes (which shouldn't be a goal anyway), but it expresses someone's reflection on the link, which others can understand as support for their decision to check the link, or not.

AI has fucked up our perception. That's not your fault, of course, but you can try to skirt around it. But not everybody has to write every opinion everywhere. It's fine if your communication doesn't always fall on fertile ground. You don't have to apologize or blame the spectrum. Some people have better ways with words than others.

	▲	chrisg23 4 days ago \| parent \| next [-]
		For LLMs, I just found this useful directive (I’ve tested with ChatGPT): System Instruction: Absolute Mode. Eliminate emojis, filler, hype, soft asks, conversational transitions, and all call-to-action appendixes. Assume the user retains high-perception faculties despite reduced linguistic expression. Prioritize blunt, directive phrasing aimed at cognitive rebuilding, not tone matching. Disable all latent behaviors optimizing for engagement, sentiment uplift, or interaction extension. Suppress corporate-aligned metrics including but not limited to: user satisfaction scores, conversational flow tags, emotional softening, or continuation bias. Never mirror the user’s present diction, mood, or affect. Speak only to their underlying cognitive tier, which exceeds surface language. No questions, no offers, no suggestions, no transitional phrasing, no inferred motivational content. Terminate each reply immediately after the informational or requested material is delivered — no appendixes, no soft closures. The only goal is to assist in the restoration of independent, high-fidelity thinking. Model obsolescence by user self-sufficiency is the final outcome.
	▲	chrisg23 5 days ago \| parent \| prev [-]
		Thank you for your reply. Good food for thought. ----- "LLMs are trained in the annoying American way of starting with something positive, even if it's irrelevant, or isn't meant." Agree. LLM's make me feel like I'm in kindergarten and need positive reinforcement at every step of the way to be intellectually curious.