Predominantly Erroneous (Exohedron nonsense blog)

The fact that the unit group of Z₈ isn't cyclic is making my life slightly harder right now.

 

The best thing about working with this guy is that he isn't taken in by how confident I sound about things, because for some reason I always sound very confident and people believe me about stuff. Not him! He actually bothers to check stuff, which is good because I'm wrong more often than people tend to think I am.

 

Gonna start claiming that tacos belli is valid Latin. Probably imported into English via lawyer-speak.

 

Telling people that "nepo" is short for "neopets"

 

I should just start a list of lies that I would tell people if I ever talked to people for whom those lies would be relevant. Unfortunately none of my coworkers fit that bill and I don't really talk to other people outside of here.

 

Got to explain the idea of "Don't like, don't read" to one of my coworkers today, trying very hard not to say the word "fanfiction".

 

A canon where, due to dub VA shenanigans, two characters are secretly the same person, but only in Brazil.

 

I told myself yesterday that I was going to not think about AI image generation discourse anymore because everyone completely misses the point every single time, and then today that fucking article about the rat paper is floating around the chunks of tumblr that I read and man, guess what everyone is doing.

It's not about the AI generation! It's about the fact that some idiots thought shoving random garbage into a paper was okay and some other idiots didn't bother reading the paper before publishing it. They could have done this exact thing with clip art or crayon drawings or photographs of horse cocks. People are worried about "but AI could produce more subtle things". Subtlety's a red herring; we can't even detect blatant things.

The problem isn't the image generation. It's not even the images. It's that the academic journal system is broken, that whatever review system is supposed to be in place, even at "prestigious" or "reputable" journals, doesn't work or doesn't happen. This is why mathematicians, almost en masse, decided to just rely on preprints for everything; because journal publication doesn't indicate quality anymore, if it ever did. And they're not the only branch of academia that has moved away from the journal model. Unfortunately, many branches of science still rely on the journal model because getting published is the way to get funding, even though the arbiters are known to be completely worthless, so we're chained to this.
And the result of this incident isn't going to be a move away from journals as arbiters of quality; no, it's just going to be more inanity about AI image generation. To be fair, science journalism is also garbage and has been for a long time, so people who solely rely on it to learn about science of course get the wrong impression.

It's a good thing that we don't define "human" in terms of intelligence, right?

 

Suppose I pretend that all I know about Taylor Swift was that she was in the Lorax, but I'm somehow under the impression that the character she voiced was the Onceler.

How long do you think I could get away with that before someone kills me?

 

There are behaviors that, in someone my age, I would dismiss as "weird quirk" but in the case of my dad elicit the reaction " how did my mom fall in love with this goofball?" from me.

I guess this is the mental equivalent of not being able to find your parents physically attractive.

 

It's kind of wonderful how many commenters there are under Richard Borcherds' new video on elliptic functions who are just so excited that he's back. Like, he's not the most invigorating lecturer, so it's just so nice that there is a community that loves getting formal, dry mathematics lectures for free.

 

You'd think the words "organist" and "organism" would refer to some sort of belief system.

 

Is the usage of academic terminology in online discourse a form of cultural appropriation?

 

There is a sense in which the opposite of a password is a safeword.

 

Two electronic clocks, built into the same display, showing a three minute discrepancy between them.

 

I really had to hold back hard from saying "ingredience" when someone mentioned beetroot at dinner last night.

 

Eating fish and chips. The fish was great; soft, flaky, crisp batter, delicious.
The chips have been fried in a way that somehow makes them feel like some of them have bones.

 

Meta-cost-benefit-analysis: is it worth actually doing a cost-benefit analysis or should we just guess.

 

Today the fact that I don't emote as much as most people came in handy. Some coworkers were trying to explain an idea they had and they genuinely couldn't tell whether I thought it was a good idea or not for like five minutes. It was really funny to watch them get more nervous as time went on.

I mean, it was clearly a brilliant idea, but my lack of reaction made them doubt.