It is time we build a Marxist-Leninist-Maoist Large Language Machine Learning Model for Multi-Level-Marketing Markup Languages

In this paper, we propose solving the alignment problem with a novel technique we call “sandboxing”: if the AI misbehaves, it gets punched with a glove full of sand

Fridays always get a thumbs up from me

Edit: link to the original video
https://m.youtube.com/watch?v=_mSbcVQPRXg

I’ve recently watched Fullmetal Alchemist (Brotherhood), not knowing anything about the plot beforehand, and it’s quite amusing how it has some parallels to today

Conway-Conway’s The Game of Life: you start as a cell on a board with two or three neighbors and a car, go through early adulthood, your first job, get married, have children, until you ultimately create an organization that produces designs shaped after its communication structures

The Germans are so sure AI will kill us they call it Killer Instinct

They say you can’t un-read a book

Indigo: “hold my magic scroll”

Our most customizable UI yet allows you to dial its appearance all the way back to Solid Sand

I had this tab open for months, but I’m glad I finally got to watch it. Learned so much about memory mapping. Fascinating solution to implement dynamic trampolines through Apple’s restrictions on iOS
https://www.youtube.com/watch?v=SuQGQ1vh9k0

Weird that in Japan the iPhone Duo has a screen on the back

“Others have created foldables that just feel like two phones awkwardly stuck together.”

Yours isn’t, which is why you call it… “Duo”? Some lack of coordination between naming team and copywriting team

Okay, the Liquid Glass design with floating bubbles instead of toolbars and insistence on icons rather than text buttons now makes A LOT more sense. I still prefer something closer to the old styles for devices other than the Duo, but now it at least has a reason for being.

(And damn, it’s gonna be quite a bit of work at my jobby job to adapt all our screens to this…)

LOL, just heard of Pangram, and asked it to analyze this post. It says “100% of this text is AI”, though I wrote it myself (I swear!). But as I was obviously aiming to mimic AI’s tropes and repetitiveness, it’s hard to count that as a false positive.

(Also, though it contains twice as many words as the minimum, Pangram still warns the sample size is small)

Hear me out: it should be called Apple Paper.

It just makes sense: it’s the thinnest Apple product, it’s foldable, and it pairs perfectly with the new $129 Apple Pencil. (And releasing a new Apple Pencil that’s not compatible with any existing products is perfectly on-brand)

Just noticed that every time I right-click a cell in Numbers, there’s an “Add To Playground” menu item now. So glad that I can quickly turn my business expenses spreadsheet into whatever the hell this is

@siracusa Every time you mentioned “Dock materials”, my mind kept wandering back to this…

@marcoarment You mentioned you are overprovisioned on Mac minis, but I still see a number of podcasts that rarely get transcripts, especially shows that are not in English.

For example, Xadrez Verbal (itunes996967108). I imagine there aren’t a lot of Portuguese speakers using Overcast, but this is one of the most popular podcasts in Brazil. Surely I’m not the only subscriber!

On macOS 27: I find this "Ask Siri" field† on top of nearly every context menu very invasive and distracting. On the bright side, though, it works for searching the menu, much better than what menus usually do when you type.

† Presumably it’s a text field, but the cursor, margin, background color, and foreground color are all wrong, compared to the established search field in the Help menu

@atpfm @caseyliss Sorry, to be clear: that was fake! There’s no audio transcription comparison on Apple’s website. I meant it just as an inside joke for fans of the show, not to trick you guys. (Also, notice the label in the selected pill)

I also almost made a chart comparing gaming performance on the M5 Ultra to the “Mac Pro (2019) with 12-core CPU”, as another ridiculously-specific comparison.

You know that old thesis that a superhuman AGI would manage to trick humans into letting them loose? Turns out we overestimated the required intelligence level by twenty or so orders of magnitude, huh