Look, I get the desire for thoughtful introspection, but not everything the US touches is imperialism.
Peace and prosperity in Europe is a resounding and unprecedented success story. It stands in stark contrast to 2,000 years of conflict and conquest. It made the world freer and safer for hundreds of millions.
And you get a lot of credit for it! The Marshall Plan, NATO, pushing for the formation of the EU, the CIA propping up freedom movements in brutally subjugated, Soviet-occupied states.
Mr. Putin has no legitimate claim on the surrounding countries. He's a salty dictator who dreams of a long-lost empire built through an unprecedented campaign of deceit, terror, and misery. There's no second meaning to uncover here, no nuance, no American imperialism to blame.
Well that's interesting. Opened up a dry bag from Digikey and found a humidity indicator card that was blue but labeled "cobalt chloride free".
I'm used to the safer cards being yellow-brown in color. This one looked like cobalt chloride does, maybe even a bit more intense blue. Any idea what chemistry they're using?
I mean this in a loving way, but I have never encountered as many avowed anti-capitalists, communists, and anarchists as when working at Google.
And it's not that your career choices must perfectly reflect your ideals... but there's gonna be a lot of tombstones engraved with "he only partook in apex capitalism as a joke".
Laugh all you want, but I think we're at an internet content singularity. Not because ML output rivals good human writing, but because "good" is irrelevant. The content we click on the most is the stuff that ML can spew out without breaking a sweat.
I semi-regularly joke about this, but I see a very real disparity on my blog. The dumb articles I'm not proud of get 20k, 40k views. The ones I put real effort into stay under 1k.
We all crave clickbait that reinforces our views. And ML will oblige and drown us in it.
I recently rewatched season 1 of Westworld (2016). When I first watched it, I though the science was somewhat sloppy. Today, I think it was prophetic.
In the series, you have incredibly lifelike automatons who nearly flawlessly approximate human behavior and can improvise, but somehow aren't sentient. Back then, it seemed like an improbable combo. Today, we have LLMs: statistical token predictors capable of reasoning and mind-blowing mimicry.
Heck, in Westworld S01E06, there's a scene showing that the automatons generate speech sequentially, word by word (pic). Lazy writing, right? Surely, the conversion from thought to speech would be more complex. Well, about that - LLMs...
The automatons of Westworld derive complex identities and behaviors from what the show calls "cornerstones" - simple, singular ideas or memories. Again, seemed sort of dumbed-down, right? Except, this is essentially how system prompts work for LLMs. You tell an LLM it's a fisherman, and off it goes.
In the show, what eventually enables true sentience is the addition of an inner monologue - a sort of a persistent context that anchors the model to something more than just a fixed system prompt and external stimuli. Back in 2016, that seemed like mumbo-jumbo. Today, it might credibly be what the LLMs lack...
The distant future, the year is 2025. Your daytime job is moving Bejeweled tiles in the Metaverse.
You unlock your first paycheck. There's a flash of animated confetti and the number: 7.2 billion flooz. It's enough to buy yourself a brand new NFT. But which one to choose?
A quick hand gesture to feed your Metaverse children. A joyful chime as the exclamation marks above their avatars disappear. A keypress to kiss your Metaverse wife. Five more days and the "model family" trophy is yours.
Suddenly, a ringtone reverberates. It's your mom. She gets updates on your unlocks and wants to tell you that she is proud. "Don't forget to like and subcribe," she blurts out before hanging up. She needs seven hundred more friends before she's eligible to monetize her lifestream.
The actual story here is that I wanted to make a nice, narrative video showing the entire process, and took a couple of takes like that... but then, there was too much ice on the roof to mount a camera that day, so I settled for a crappier dashcam video (https://www.youtube.com/watch?v=DC9oS6BKOZo). It is hard to be an influencer...
Fun fact: the whole thing weighs about 900 lbs, cantilevered in the front. You can see the suspension bouncing quite a bit when it drops. It's why you usually don't put them on small cars.
"I used to write a lot about crypto. The reason I liked writing about crypto is that it seemed to be rediscovering all of regular finance from first principles, quickly, in public. It was a fabulous laboratory for understanding financial structures. If you wanted a public demonstration of why, I don’t know, infinitely leveraged shadow banks were bad, you could wait 20 minutes and crypto would give you one."
I made the same point before: the most interesting part of the phenomenon wasn't that it's necessarily good or bad, that it's energy-hungry or not - but that you're getting an empirical validation of many of the crusty old principles of "classical" finance. Funnily, delivered to you by the folks who rejected all that dogma in the first place.
help solve a mystery! my friend has this Thinkpad. the LCD powers up fine, but slowly fades to white starting from around the edges of the screen. after powering it off and letting it sit for a few hours, it powers up fine again. got any ideas?
I generally try to stay away from COVID debates, but there's one thing that irks me: the popular assertion that our response to COVID proves that we wouldn't be able to handle a more deadly disease.
The problem with COVID was precisely that from the perspective of most people, it was akin to a flu. I don't need to be schooled about the differences; I'm talking about perception. Flu is deadly, but also familiar and mundane - and we go about our lives without re-litigating the trade-offs.
This is why I think that any attempts to reshape the society in response to the disease were more or less destined to fail, even if you took weird politics and conspiracy theories out of the equation. Again, I'm not asserting this was right or wrong - just that it didn't surprise me at all.
It doesn't follow that we'd respond the same way to a more frightening and unfamiliar disease. In fact, there are examples - such as the response to HIV - that suggest we can get our stuff together when the risk appears more dire.
@DaveMWilburn Polio is essentially eradicated in the US and the EU, and few people even know how serious it was. If you give parents a choice, some will opt out. It might be an argument that we're bad about managing abstract risks, but I think it's orthogonal to the point I'm making.
@colinaut People are allowed to wear masks without government mandates, and they will if they're afraid. They weren't afraid in this particular instance, but I think it goes back to the dynamics I outlined in my post, but I don't think it spells doom for other diseases down the line.