@tenobrus Or maybe it's just that we offload half our thoughts to a digital brain, and by releasing frontier open weights it shifts the equilibrium towards a brain that thinks in Chinese?
@tenobrus For example, the canonical intepretation around certain political events. For example, views on trademark and copyright. Or idealogical views on what technological and social progress looks like?
@basilYam@QiaochuYuan Hi! Like any skill it needs to be acquired through deliberate and repeated practice. Most people don't deliberately seek out these challenges, or at scale. But occasionally, when things happen, and we can't understand it, then having this as a reference is useful. Hope it helps.
@heiss_7 The minute they solve the dogshit inefficiency of every new message needing every single previous message to be sent alongside with some sort of memory system memory requirements will fall off a cliff
@mattpocockuk@stefanbinoj Personally I found the name of the AI hero course to remind me of the naming of courses by trading gurus, what made me take a closer look was learning about your previous typescript work. Glad I did though, I find your AI guidance to be the most directly applicable out there.
@AgustinLebron3 Non ivy league CS, would say I'm fairly good at thr engineering side but have no illusions of competing with IMO kids.
I've been hopping around, currently somewhere that's comfy, not very well paid but with a lot of free time, and spending it learning ai agents.
@AgustinLebron3 But it's not all-consuming. In the interests of wanting to walk down one road really deeply, do current market conditions favor choosing something else? (end) /
Thanks for doing this ama!
@AgustinLebron3 Thanks for doing this ama! If it's not too late, I'd like to ask - - I'm a few years past college and I notice many of the established firms overwhelmingly prefer to hire young. Am I out of a shot? I like the problem space of trading /1
Late night tweet. The Indonesian economic question from the 1950s, raised by those who established "economics" as a concept here (Soemitro & Sjafruddin), is growing on me:
"To grow the economy, you grow the amount of capital, or the labor productivity. Ideally both."
Basically, both men arrived at the same conclusion: post-colonial Indonesians are retarded because there's no school and no rice (malnutrition everywhere), so it's pointless to try to grow the economy out of improving labor productivity. From a purely economic lense, the Indonesian inlander population are useless. What's valuable from Indonesia is the natural resources and the plantations, not the humans.
So, it's futile to immediately try to grow labor productivity after Independence, because we don't have money to do it. We should instead focus on capital accumulation.
Out of this assumption, they arrived at different conclusions.
- Sjafruddin thought that we should focus on stabilizing and increasing this capital accumulation out of the existing plantations and oil mines, so that we actually have money circulating around to doing anything.
- Sumitro (a socialist) thought that the State should not sit around, but start actively try to artificially inject capital accumulation in the hands of racially indigenous Indonesians, through e.g. the import licensing regime of the Benteng Program. At the time, more than half of the Indonesian economy was still owned by foreigners.
Both agreed that violently seizing the plantations was a retarded idea, because again, we are retarded and of useless labor productivity. There was no schools. There was no bureaucrats. We cannot actually run the plantations. Their assessment will be tragically proven true after 1957, with cataclysmic cosequences.
This dynamics, I think, is becoming relevant again as Indonesia faces imminent catastrophic defeat and violent destruction and rape by AI - robot accelerationism and Chinese manufacturing expansion accelerationism.
For example, the only promising Indonesian industrial sectors are downstreaming industries i.e natural resource-driven protectionism, such as nickel smelting.
Downstreaming is driven by an ideology that sees the Indonesian population biomass as completely worthless and stupid and retarded and useless and not worth spending capital on in terms of building industrial productivity, and that control over natural resources and capital flows is actually far more important to industrialization.
Is this assertion accurate? Are we Indonesians actually useless and worthless from an economic point of view?
That I am unable to conclusively reject or accept this assertion is my biggest political existential crisis.
Karl Marx once asserted that value comes from the labor of workers.
In Indonesia's case, whether in the 1950s and the 2020s, I think Karl Marx's assertion is simply wrong. Compared to other economic elements like rubber or nickel, the labor productivity of the workers are worthless and are not the primary source of money creation.
Consequently, you can just repress or genocide workers whose labor output are actually totally worthless in an economic sense, and economic production will continue just fine.
This already happened in East Timor. The East Timorese population was useless and annoying to New Order's economic production, so we simply genocide a third of their population.
Another example of this is the 1965-1967 genocide. PKI was particularly popular amongst subsistence farmers in overpopulated Java who don't actually contribute to anything to national wealth and are in fact are eating up the limited remaining quantity of food in the country. This is because of Soekarno's extreme mismanagement, causing the non-existence of schools and roads and fertilizer and harvest.
But it also means that if we kill millions of these Javanese peasants in killing pits, in practice nothing of economic value will be lost since they don't really produce anything anyways, and in fact the nation quickly rebounded economically under early New Order.
Soeharto and the Army was able to industrially exterminate around a million people and jail another 2 million in concentration camps in Buru Island because these people are completely utterly economically worthless and useless to them. The source of money of the Army is the plantations. Occupy the plantations, and it makes no difference whether you then kill 2 million or 20 million subsistence farmers, who are of no value to you.
In a more modern setting, the failure to create formal jobs have resulted in the explosion of thuggery. Thuggery is another form of worthless biomass failed to be integrated into economic production. The one in the wrong here is the government. But the fact remains: if we genocide 5 million thugs in industrial killing camps tomorrow, the Indonesian economy will run just fine, becaue these 5 million people are economically worthless. They don't produce anything. They merely consume.
We can laugh at the genocidal extermination of the 5 million useless thugs, who failed to find a place in modern society. Then, two years from now, all the white collar jobs are taken over by AI. Now what.
It used to be that the solution to this low labor productivity is "education" and "healthcare" or whatever. In the 1950s, this is obviously true. In 1967-1983, this theory was implemented and it worked brilliantly. Today? I have extreme dark doubts about it.
The singualar goal of all 20th century religions is the creation of "New Men", which are people who are able to successfully bloom their full potentials, like a seed sprouting into a massive tree. These religions are collapsing one by one. He who shows the path to salvation, who shows the path towards the explosive realization of the full potentials of billions, is the Messiah.
Industri manufaktur non-hilirisasi Indonesia akan roboh. Musnah.
Alasannya sederhana: kalah sama kompetisi China. China menguasai pasar ekspor & sedang bertransisi ke robot.
Deindustrialisasi, pengangguran massal, dan kehancuran factory belt Cikarang / Tangerang di depan mata
@int32max I run 4 panes, fs / editor / terminal / git (diff on top lens below)
Git diffs open directly in editor pane
I envision the possibility of using a binary window splitter, editor can be nvim or vim, terminal is terminal, maybe tmux for tabs, remaining two pieces to be seen
@int32max Though I must admit my own gut feeling is that as pure terminal tools improve we might be able to get this outcome entirely in the terminal? Won't be clickable and will probably involve a lot of learning shortcuts.
154 Followers 5K FollowingEveryone wants to be a trader but no one wants to focus on managing risk. Not financial advice. @gigachadventure | @0xfishingclub | @veritaslimited
27K Followers 107 Followingidiot. cuda og, harvard val, morgan prize, society of fellows, 1 hilbert problem so far, creating friendly, SAFE, delightful, supergenius ..things @anthropicai
71K Followers 3K FollowingWe're in a race. It's not USA vs China but humans and AGIs vs ape power centralization.
@deepseek_ai stan #1, 2023–Deep Time
«C’est la guerre.» ®1