RE: https://sauropods.win/@futurebird/116304296612633497

A thing that my years working in the software industry have taught me is that sometimes people just want information. Not good information, not accurate information, not useful information. They just want information.

“Here is a transparently fictional timeline for our project!”

“Here is some technical gobbledygook that will make you sound smart if your parrot it with confidence!”

“Here is a slide full of charts nobody will ever critically examine or even superficially understand!”

“Here are some quantitative metrics whose meaning is beyond dubious, but we sure as heck were able to output those numbers!”

“Here is a whitepaper that sounds just like your idea of what a smart person sounds like!”

@inthehands while ai is causing new problems here I think this is also part of how we teach people how expertise and the social value of "smartness" works.

Sure people often need to rely on the expertise of others, but we should emphasize that it's an exercise in community and trust.... and usually we don't, schools present it as an exercise in authority... so yeah people go looking for something that produces authority for them.

Edit: wrote some more about this: https://bayes.club/@wronglang/116307615948005128

Krzysztof Sakrejda (@[email protected])

I contribute to extremely applied research on mundane but politically spicy topics such as health and inequity. I have a degree in chemistry and a PhD in evolutionary biology that gave me enough time wrestling with the gnarlier parts of statistical algorithms to be an expert on most* of applied statistics. I try to respect that work when posting. The way expertise is treated on social media and in daily life is broken, and it's hurting all of us. #Introduction

Bayes Club
@wronglang @inthehands Which is why we invented writing, or the experts are overwhemed
@inthehands Slow down! Are you programmed to invent riddles?
@inthehands The curse of the quantified life. It's so tempting to want more data and more measurement, when most of the time, it doesn't matter that much, and can even make it more difficult to do things like, say, listen to your own body.
@inthehands Except that's not information. It's information-shaped noise.
@dalias
“Information” in the information theory sense, not in the signal vs noise sense
@inthehands It's not information-theoretic information either tho.
@inthehands Like if I have a program that implements a prng and outputs the successive values emitted by it, adding a TB of output from the program alongside the program source adds zero information.
@dalias
I suppose it’s an open question whether business BS has a Kolmogorov complexity that’s high (because it’s random) or low (because it’s predictable)
@inthehands this is how we ended up with a federal chat bot that tells us the healthiest part of the cat is the internal organs, because they have all those micronutrients
@inthehands this is extremely insightful, and absolutely matches up with my experience.