Awwwwww thank you!
I had trouble trying to find my own MakerTube toots today as well, was trying to check a reply to a comment. I think something is wonky on the server. Says I haven't posted anything since June 14.
Here's the same video you shared on MakerTube: https://makertube.net/w/u3WMTJ8wNhKcaGuWVMFaoe
It's no secret that I'm a big fan of @ewen and I've been meaning to share this video for a while but every time I try, for whatever reason, I can't find it on his peertube instance via Mastodon. Anyway here's the YouTube video (it's also on his peertube if you use a peertube browser like newpipe).
OK preamble aside, Ewen has a very particular look for his images, and he's often alluded to his digital darkroom process, but in this video he has very clear instructions. It's really good, concrete advice, and it is very clear how those steps form the aesthetic. I got a lot out of this one.
It turns out the oil crisis was just continuing and then Trump did the thing to "solve" it and then we got news coverage and AFAIK it's unsolved again and the news is quiet again. The weird twist is that everything is fucked when the news *isn't* covering the issue.
Still unsure how the petrol prices are so low right now, and worried about if grocery prices are going to just shoot up in the next few months.
It's a monoidculture in the category of endofunctors, what's the problem?
So a couple of folk (including Gina Reinhart) have talked about how TSMC should just set up a new fab in Australia, and there was some conventional wisdom that the world could only really support one Fab, because the actual pipelines are so expensive to set up, they cannot possibly recoup their costs. AMD and Intel, for example, have started using Glofo and Intel's local fab to do things like L3 cache / the chiplet substrate or whatever. The point is that as you go down the reticle sizes, you basically have to pay more to allow the Fab to pay off the lithography machines. TSMC has $250 Billion in assets.
Except, AI already blows those numbers out of the water? Like people are talking about nationalising AI or treating chip manufacture as a strategic asset. It's all funny money now. I used to think "holy fuck, a single machine is over a billion dollars odd", and now there are trillionaires and AI needs trillions.
Not to mention Aukus costing $400 Billion odd for second hand subs?? In the realm of "normal money" the world can support only one Fab, but as defense spending it's a drop in the bucket. I feel like I'm going crazy.
Guys did you know that curing cancer is really hard? We should just not worry about it, like sorry to all the people with cancer but it's just too hard. Like so many scientists have been studying it for so long and there are so many medications, like has anyone looked at the cost-benefit here?
Guys did you know going to space or the moon or mars is really hard? We really shouldn't be spending our effort on that stuff it's super duper hard you guys. It's taking entire industries to prop up the idea of going into space and there's not even really any reason to do it. Has anyone even considered the cost-benefit here?
Guys did you know physics is really hard? Like so many people are spending so much time studying this stuff, there are telescopes and radio telescopes worth millions. Why are we doing this? There is no cost-benefit here, only cost.
This argument is ridiculous for everything else. Why is it OK for climate change?
You can apply this to all hyperscalers btw. I can barely afford an ebike, and one that is "safe" is just barely out of reach, but hyperscalers can apparently afford so many ebikes that they can piss you off enough so you toss them in the river and they'll barely notice. This isn't just their investment being stupid, they can literally afford multiple ebikes to rent to you, and you can't afford to buy one.
Originally I was thinking about hyperscalers as the fault of the RAMPocalypse, the argument being that if we had good anti-trust law, no one would attempt hyperscaling and therefore we wouldn't be "outbid" for RAM.
However, I think the reality is more grim. These hyperscalers get investment from people who can afford it. What we're seeing today is literally our wealth against the wealth of the bourgeois. If the rich need a resource, they can literally pay double or triple what all of society combined can pay for it. If we throw tomatoes at them, they can literally buy all the tomatoes.
Normally, we don't see the money of the rich conflict with the needs of regular folk, but here we can see it. This is what it looks like when you're going against someone "with the means". "The means" is entirely different scale to what we expect. It's like a "slice up the pie based on how much you think people own vs what they actually own", but now you have to survive on the crumbs.
I want to contrast "bro science" vs "scientific metaphors" for why and how bro science gets it wrong.
Bro science tends to simplify by hyperfocusing on one variable, then optimising that variable. A classic one is "protein". It's too hard to figure out the rest, so we need to focus on the one superfood I can eat over and over again. Boom! Chicken. Problem: solved. I could eat vegetables, but then I'd need to eat *two* vegetables and that's too complicated, so out it goes in favour of the one "solved" food.
Similarly for energy. I just want lots of energy. Boom! Nuclear or Fossil fuels! Don't think about it, just spam the "I win" button.
The problem is that this isn't how a lot of things work. There are a lot of variables and often even the small variables, or the items which have a small number, matter. You're trying to simplify optimising for 12 different things, and that's why actual scientists prefer using the "scientific metaphor". An example here is the food pyramid or the plate method. For climate change it's things like the Greenhouse effect, though that does ignore variables, or the water cycle, energy cycle etc.
This probably partly explains why Engineers tend to be anti-science / climate deniers. They use the "Bro Science" method. They don't like the variables, they remove them, and now the model is simpler but wrong.
There's possibly also a masculine / feminine dimension to it. The Metaphors are more "airy fairy" (but maintain the complexity), but bro science sticks with maths (through ignoring complexity).
There's also an aspect that Economics is actually a soft science, but because Bro Scientists started working on it, they "solved problems" by ignoring complexity and adding maths. I can't recall now but there was a story of an economist who basically refused to add math to her work, and she was pretty roundly rejected as a result, even though it had a lot of predictive power.
Come July, and probably a bit before and after, people are going to try and convince you that the current government changing energy prices is somehow a bad thing. I would like you to know that if you have the money and actually believe in climate change, there is no better way to spend your money than through electrification and embracing new energy infrastructure. If you can afford batteries, get batteries, put as much solar on your roof as possible. Yeah, you're no longer getting decent Feed in Tarriff, but it's still good for the environment. Switch out Gas heating and Gas hobs for Electric, switch to an Electric car (or even better, an ebike if you can).
Also, some judicious use of smart appliances like Home Assistant will also help you use energy effectively.
Invest in electrification. It'll probably save you money in the long run, and it's good for the environment.
You can tell Iran won because K-Mart has just launched the Khome-store.
Whilst I was a proponent of "guys" as gender neutral greeting, it appears society at large has decided on both "babes" and "bro" as gender non specific. Sure why not.
OK hypothesis time. People treat money as a yardstick, but the big problem with it is that it changes based on what it's measuring, where, and when, which makes it kind of useless as a yardstick.
Case in point "Musk is a trillionaire". He's sold some stock at a price where, if he sold all of his stock at that price, he'd have a trillion dollars, but here's the thing, people wouldn't buy all of it at that price. The vast majority of the people in the world (and, sadly, only about half the money) would pay nothing for that stock because they need the money for food.
There's a thing that Musk could "give a million dollars to the homeless", like, no, he could only really give a million dollars "worth" of stock to the homeless, which is basically useless because the homeless would need to sell it immediately for food and that would tank the value of the stock.
So, left and right, people don't understand money. So this is the hypothesis: It's better to understand money as a trinity of Money, Taxation, and Regulation.
Money is money, but that money has no meaning without taxes. There's a myth that you need some sort of scarcity to create a currency, but we can see that bitcoin eg doesn't work that way, and a currency very much is inflationary, so it's not scarce. Ultimately the point of it is that the government "handles" the monetary system, that's what you "pay" for with tax.
And the third is regulation. This is monetary regulation, yes, but it's actually the overall regulatory landscape. The milk isn't poisoned, the water isn't poisoned, etc. That's the part that makes people want to live in the country and spend the currency. And monetary regulation means your money is "safe". These three things need each other to work properly.
Now that I say this out loud, I don't think it's novel at all, but I do think that when we talk about money, we talk about it like it exists on its own as a yardstick. It's not, it's all 3 things working together. It's not "a Trillionaire", it's "a trillion shithole dollars for a shithole country".
I don't understand people who dislike Rick and Morty or Family Guy. Sure, the gags are visual but they're still top notch. Today, I think you'll find more recent memes from either show than from the Simpsons. The Simpsons had the benefit that often the gags were spoken, so you could reference them in speech, but on the internet, the visual gags work better, and honestly they're good gags from good shows.
When ONP supporters are like "I'm not racist, I'm just a bootlicker" like THANKS BRO!
I have a horror dentistry hypothesis. In Horror a lot of things are mouths, in weird ways. I guess the idea is that they'll eat you. But there's a key split between mouths with "carnivore teeth" and "herbivore teeth". The Carnivore teeth are generally just sharp teeth but not really built for purpose, just kind of hanging out there randomly because it's gross, and herbivore teeth are lined up perfectly in a menacing grin, but also, not really built for eating flesh?
My theory is that Herbivore teeth deal with our guilt, having done something wrong and the herbivore wants revenge, or otherwise is a mundane inhumane system which we have created, whereas carnivore teeth are fundamentally about a thing which wants to consume us, use us as prey.
There are parallels between "fast zombies" (the zombies are nature) and "slow zombies" (the zombies are society).
Ever wonder what would happen if the people who say "Gyatt" grew up and got an engineering degree? Well, wonder no more...
I find myself constantly frustrated with javascript because devs value a lack of abstraction. I've termed this software nihilism, the idea that the only thing that matters is <some metric> and all abstraction is useless.
This is ironic because JS is one of the most wasteful languages to get an outcome which does not aid in development speed. Yes, there's a sandbox, but it's clear that you can significantly increase speed if you use the same sandbox with web assembly (for example).
This makes working on a js project twice distracting for me. First, because I want to add the abstraction (and know I'll be met with resistance), and second, because actually reading concrete code like that all day long is laborious.
Sorry for AI posting. I try not to dwell on a single topic forever, but I guess I've been subjected to it and now I'm subjecting you to my thoughts on it.
If computers are like bikes for the mind, AI is like cars. It's isolating, it consumes a lot of resources, it's almost built to separate the classes, it changes the landscape to create affordances for the AI. Before you know it, your entire social fabric is "AI dependent", similar to car dependence. In some number of years, we'll have the same discussions like "OK maybe AI can go further from the city but that induces demand" or "the reason you are far from things is because of the amount of real estate given over to cars" or "15 minute cities" but with AI.
AI is cars.
Honestly the public charging network has been one of extreme frustration. While the EV Council of Australia says that PAYG charging should be available everywhere, in practise I haven't found something which doesn't require an app.
Then, you look at the costs, and right now Petrol is actually cost competitive and sometimes cheaper than EV charging if you use the fast DC chargers.
It alarms me that you can have an EV, a charger, and cash, but you cannot pay to get your EV charged.