@RandomRanger's banner p

RandomRanger

Just build nuclear plants!

1 follower   follows 1 user  
joined 2022 September 05 00:46:54 UTC

				

User ID: 317

RandomRanger

Just build nuclear plants!

1 follower   follows 1 user   joined 2022 September 05 00:46:54 UTC

					

No bio...


					

User ID: 317

You're not alone: https://youtube.com/watch?v=oXMjtVnLD4o

Harden your heart Putin, Increase your attacks, Banish them all to Palestine and we shall marry Ukrainian women!

By contrast, Gorbachev's decision to undertake reforms were taken because of accurately-identified issues

Yes but the way he dealt with issues was poor. Reducing military spending would've greatly ameliorated the economic situation, it was sucking up a good 10%+ of the Soviet economy. Gorbachev didn't even have the power to control military spending but he thought he could radically alter the whole ideological and economic structure of the Soviet Union - in a controlled way! The man was dreaming.

That the 90s followed doesn't mean that the strategy of change was a wrong decision

If a strategy is launched in an inept and naive way and fails, it's a failure of strategy. A return to hardline Stalinism would be a 'strategy of change' yet that wouldn't have helped either. Change and reform is not sufficient, it needs to be the right change done in the right way. Implementation is important - gradual and controlled marketization beats chaos. Nothing about the Soviet system required handing everything over to robbers in a mad rush to privatize all assets before the communists could be elected, the Yeltsin approach was extremely counterproductive. Gorbachev's ineptitude led to the hardliner coup, he didn't manage the situation sufficiently well. Now nobody had ever done this before, it's a difficult task that he wasn't trained to do. Indeed, the Soviet failure helped inform China's success. Yet it was still a failure.

However, good management is not some made up video-game skill, it requires a sound understanding of the people and institutions that control a country, it requires certain personal characteristics that Deng clearly had. Even Putin did a decent job in cleaning up much of the mess that Yeltsin left behind - Putin is not an exceptional leader but he's not a Gorby/Yeltsin-tier blunder-addict.

The result was a Union where even Crimea wanted to leave it, and when ungraceful suppression was ended, did

That was the result of mismanagement and a certain level of naivete (itself a result of poor management) about how things would be outside the Soviet Union. As late as 2013 Ukraine regretted leaving the USSR.

Anyway, you started this diversion saying the war in Ukraine was the worst disaster for Russia since '41 - did you miss the increasingly frantic rhetoric coming from Macron and the Pentagon about how the Russians are about to roll the Ukrainians?

“There’s nothing that can help Ukraine now because there are no serious technologies able to compensate Ukraine for the large mass of troops Russia is likely to hurl at us. We don’t have those technologies, and the West doesn’t have them as well in sufficient numbers,” one of the top-ranking military sources told POLITICO.

It's not looking good for the rules-based order.

If you mean 'see' as in 'it's not there,' this would be flatly incorrect, and anyone who lived through the early 2000s could probably recount more than a few American examples.

No, American nationalism is not on the same order as Chinese nationalism today or in the 2000s. Not even after 9/11. The US ambassador in Beijing was trapped for days after the Belgrade embassy bombing as hordes of rioters threw rocks. China routinely blows up tiny maritime incidents into completely disproportionate affairs. The most popular movie in US history wasn't a patriotic war story like Saving Private Ryan toned up to 11 with 'the eternal glory of the US Army remains in our hearts forever and ever, amen' on the postscript. What are you thinking of - Islamophobia? China is way more Islamophobic than the US has ever been, as the US govt delights in telling us so often.

in a conventional conflict in Europe, the Russians were likely to be decisively beaten in any conflict with the Americans

Firstly, the Iraqi army is not the Soviet army. Just the arsenal Ukraine inherited from the Soviet Union is a whole other world to the SA-8s and Rolands Iraq was fielding. The Iraqi army was also saddled with Iraqi soldiers, who were not known for excellence under US tutelage either. We've yet to see how Airland Battle deals with S-300s or the arsenal of a proper military. Secondly, conventional inferiority was no problem for NATO in the 1970s or Russia today, they have nuclear deterrence.

Before I posted the link, and before you posted your opening thesis on American strategy, had you ever read or reviewed an American National Security Strategy?

This may be news to you but you do not have to read these documents to discuss national strategy. You can look at what actually happens in the real world. You can interpret govt priorities with your own eyes. This is better than trusting in the documents. Govts lie! The Chinese might say that they're interested in purely peaceful development - yet actually build up a gigantic navy and forces targeting their near abroad. The US might say it's worried about Iraqi WMDs and Saddam's links with terrorists - but have other motivations and goals for invading Iraq.

And when govts don't lie, they try to be tactful, they massage their words and adopt a certain frame. The Chinese adopt this supercilious tone where their military may be forced to take action if foreign provocateurs incite a rogue province into illegal independence activities. That's not a lie but it's not straightforward communication. Better to ignore the cheap talk and look at results.

Congratulations. You officially won the argument and convinced key American elites years ago. DEI and migration policy are how the US government under a Democratic administration believe the US will long-term compete with PRC STEM diploma numbers, and it made it into policy.

The migration policy of having a de facto open border? I note this is contrary to what is indicated in your august strategy document. US migration policy isn't primarily about improving the quality of the STEM workforce but about demographic and political change, plus serving certain corporate interests. The vast majority of the millions of people arriving in America (many flown in at state expense) are not trained in STEM. In fact US legal immigration is a rather byzantine and complicated mess, making it difficult for the most skilled to arrive.

This is where the advantage of my 'look at what's actually going on' approach kicks in. I can observe that DEI and migration policy is not motivated by a desire to acquire STEM talent. If they wanted talent, they could adopt a points-based system like Australia and enforce the border. If they wanted talent, they'd favour meritocracy as opposed to diversity quotas and affirmative action. It's not rocket science. This policy isn't secret - its publicly observable and it does get communicated. But people massage the truth, they arrange their intentions in certain ways to make it sound more defensible. Children are taught things like 'diversity makes us stronger' in school and via the media, just like how China is taught nationalism via school and the media.

Furthermore, relying on Chinese STEM talent to counter China has a number of rather obvious flaws. This is what I was pointing out initially. The DEI and Rules-based order strands are in conflict. The US wants to skim off Chinese STEM talent but not end up training them so they take skills back to China, not have them spy for China. They want to whip up popular sentiment against China (another thing you won't find in official strategy documents but which can be observed through funding of various organizations and media slant) but do so without inciting racism or civil unrest. These are the contradictions I've been talking about the whole time.

The purpose of a system is what it does, not what it says in the PR brochure.

Yes, though I honestly thought it was spelled that way, never having read it on paper.

They're either pirating it, streaming it from some sketchy website that pirated it, or watching free preview stuff.

There's loads of porn available for free. Twitter, reddit, boorus, 4chan, discord... People who pay are suckers or have more money than sense.

Should we have a 'should be longer' and 'should be shorter' upvote/downvote button?

A good few reported comments make a point but I believe the real problem with them is that they don't substantiate their claims or elaborate. They can't, nobody can in only a few sentences.

Alternately, there are some top-level and mid-level posts that are so long my eyes just glaze over and I scroll onwards. I'm wary of doing that myself and try to prune things down. That comes at the cost of detail, I sometimes end up letting considerable weight rest on single word qualifiers I add where perhaps sentences are needed. Scylla and Charybdis. I don't know how hard length-voting would be or if anyone else cares. Opinions?

Everyone would have to do it all at once or the Amish are going to get crushed by people who retain modern technology. It's not stable for some people to retain modernity and some to go without.

It's easy to say those things now but when the time comes, will you say 'no, we shall preserve this massive pot of wealth as a national park'?

Imagine a hundred million tonnes of gold and uranium. A trillion barrels of oil. A hundred Australia's worth of coal and iron ore. All of that is a rounding error compared to the Sun. Its value goes up into the wacky numbers, the sextillions and nonillions.

Ukraine has been by far the worse Russian strategic failure.

I'll stop you right here. No, the fall of the Soviet Union and the Yeltsin era was catastrophic for Russia in a way that Ukraine is not. Massive impoverishment of the population, economic collapse, social collapse, demographic collapse, military collapse. Gorbachev had a strategy to reform the Soviet Union with policies of perestroika and glasnost. He wanted socialism with a human face and to preserve the Union. It failed massively and disastrously. I'm not sure if Yeltsin had a strategy other than 'remain in power' but it certainly wasn't good for the country.

https://tradingeconomics.com/russia/gdp-per-capita-ppp

It's possible to reform a socialist country into a market economy without shock therapy, without shelling Parliament, without a decade of chaos. China did exactly that. Ideology is a part of national strategy, if people stop believing that's a problem in and of itself. With better management of the economy and internal politics, the crisis of belief would've been mitigated. They could've transitioned to other sources of legitimacy in a more graceful manner. And hey, there are people in the West today who still believe in socialism. Indeed, the Russian communist party was quite competitive electorally - it took some trickery to keep them from ejecting Yeltsin.

Equivalence of NEETs and cultural equivalencies isn't an advantage remaining with China, it's the undermining factor of any strategy that depends on social elan by denying it unique advantages proscribed to it.

You don't see the kinds of hysterically aggressive Western nationalists that are found in China. There's a reason why I highlighted Little Pink and not lying flat. You don't see many novels in the West heaping xenophobia and revanchism into their story, demonizing national rivals. Furthermore, US social elan is in a pretty poor condition - January 6th is proof of that.

You seem to be under the impression that the national security strategy is only about the field of security, which is a very important indicator that you haven't read it. And are probably confusing it with the National Defense Strategy.

Above you were saying that the fall of the Soviet Union was a failure of ideology, unrelated to security issues (which is wrong, given how Chechens and Islamists immediately took advantage of Russian weakness). Anyway, you were saying strategy was about security. Now you want to say that US strategy encompasses more than security? I had a look through the document you nominate as the holy text of US national strategy and there's loads of ideological and legitimacy content in there, the need to defend democracy and human rights. Plus there's a fair bit of DEI stuff as well - they want more STEM for girls, there's anti-racism content, they affirm diversity as a national value...

They don't outright say DEI is at the core of their ideology, they say many things they don't really mean. It's a public document, not to be viewed as a window into their innermost beliefs. Only in the implementation, in the specifics of outcomes do we see what they truly desire. That's what's important and why I brought up the DEI/CHIPS article in the first place.

Article prophesising and diagnosing doom on the Indian economy: https://time.com/6969626/india-modi-economy-election/

Some of it is really staggering - 10 million manufacturing jobs were lost even before Covid. Indian manufacturing employment halved in 5 years (up to 2021 but doesn't seem to have recovered that much, though output is rising). Even in output it's still quite low as a % of the economy, falling as a proportion the 2010s. Meanwhile there are 60 million extra farm workers: deurbanization and deindustrialization. India is at the bottom of the Global Hunger Index, below North Korea and above Afghanistan.

I note that the article author wrote a book on Modi despotism so can't be considered unbiased, yet he has a lot of pretty ominous links. Some of them are structured in a deceptive way - 'crude imports' being down 14% goes against the article's overall message of trade increasing, even if goods exports decreased. It's always good to read these kinds of articles with a sceptical eye, economics is so broad that you can paint all kinds of pictures. It looks like India is focusing very heavily on services rather than manufacturing, leaving it rather vulnerable to AI disruption, as self_made_human has predicted. Vietnam exports more manufactured goods, while India's overall exports are much higher.

There's also this fun website that graphs exports by type: https://atlas.cid.harvard.edu/explore?country=104&queryLevel=location&product=undefined&year=2021&productClass=HS&target=Product&partner=undefined&startYear=undefined

India is ICT nation, the tech support stereotype. China and Vietnam do manufacturing. Russia and Australia dig things out of the ground. The US does services and manufacturing.

Ukraine and the West's official war aim is to retake all Ukraine's 2014 territories which include millions of Russians. There are many more inside Ukraine's currently controlled territory, where being Russian is not very popular. There are fundamental differences between the two states that can no longer be resolved diplomatically.

There's also the strategic dimension regarding control in the Black Sea, bases and so on.

Iran's firing off Shaheds that cost $10,000 to $50,000 to produce. Israel's trying to shoot them down with F-15s and F-35s, plus their air defences. Can that possibly be economical? The flying cost of an F-35 is about $30,000 to $40,000 per hour. They can shoot down multiple Shaheds on each trip but I assume they're firing air to air missiles, it's a rather small target for cannon. Sidewinders apparently cost nearly $400,000. The Israelis make their own missiles and price is unclear but one imagines it's somewhat in that hundred thousand dollar zone.

At any rate it's somewhat academic, I imagine the US will be paying the bills.

LCS was horrendously bad, was it even suitable for Iran?

In April 2012, Chief of Naval Operations Greenert said, "You won't send it into an anti-access area"; rather, groups of two or three ships are intended to be sent into areas where access is jeopardized to perform missions like minesweeping while under the cover of a destroyer. The LCS's main purpose is to take up operations such as patrolling, port visits, anti-piracy, and partnership-building exercises to free up high-end surface combatants for increased combat availability.

Sounds like it was supposed to be handling Yemen...

By May 2022, the Navy shifted its plans to decommission nine LCS warships in Fiscal Year 2023, citing their ineffective anti-submarine warfare system, their inability to perform any of the Navy's missions, constant breakdowns, and structural failures in high-stress areas of the ships.

Or not.

Yeah I saw it, the intro theme is something else.

https://youtube.com/watch?v=eqONgYHYo88

Surprised by what everyone else is saying. Power sounds very difficult! An oil-rich country with what, some S-300s and Mig-21s and no allies? Could LARP as Equatorial Guinea's fantastically corrupt ruler but you couldn't even play the race card when someone decides to launch some regime change. And what about threats from within - there are going to be ministers and officials. Governing a country is hard, how would you know to deal with the locals? Do they speak English? This is the kind of thing an air force colonel or a baron could pull off, I doubt many of us have the right management skills.

For me it's Pleasure, though I was tempted by Comfort. Pleasure is literally maximum fun. Normally I'm a power and adventure man but they seem high-risk, high risk in the sense that you'll have tedious and/or crushing problems to deal with.

Reading economics is like reading politics or history, there are many opposing schools of thought. Comparative advantage vs industrial policy and so on.

I'd just be wary of reading one book and thinking 'that's it, whew, I've got my economics'. Like politics or history, there's a great deal of it floating around, you're most likely to only find the orthodox versions. Yet orthodoxy changes! Tariffs and protection used to be despised and derided, yet they're coming back into fashion. Game theory is related and something to consider as well.

Sure, the US does do this but not quite at the scale and commitment China's working at. How much money has the US promised on the CHIPS act? Tens of billions. Is the US willing to build nuclear plants? Not really. Other considerations come before aggressive implementation of technology. Musk has to file his environmental impact statements before launching his rockets.

China is spending in the hundreds of billions on chips. They're building the most nuclear plants in the world. They've got a huge industrial policy machine, they really put in effort when it comes to technology. When it comes to high-speed rail, they don't just talk about it, they build it. They don't care about environmental impact in China like they do in the US. Technology comes second only to communist party control.

The US still spends more on research but they're not exactly growing their research spending like China is:

https://data.worldbank.org/indicator/GB.XPD.RSDV.GD.ZS?locations=US-CN

Good points. India's fertility is anything but stable though. There's getting old before you get rich and then there's getting old before you're out of poverty. https://data.worldbank.org/indicator/SP.DYN.TFRT.IN?locations=IN

I have no problem with Modi personally, running such a big country democratically is a big ask. US democracy is not exactly high performance governance, how is anyone supposed to manage 4x the population with a small fraction of the wealth?

IMO India needs to follow the standard industrial route out of poverty. Special economic zones, foreign investment, light industry -> heavy industry -> high-tech industry and then services. Doing this in a democracy is very difficult, as you point out. Getting the tax and judiciary right is easy for autocracies, hard for democracies. Plus there's an excessive focus on IT and software. Software is all well and good but what about hardware? What about making the solar panels? Without manufacturing there's no firm base for development, skipping from agricultural to service-economy has never been done AFAIK. Vietnam has been on the right path and is pulling ahead.

https://data.worldbank.org/indicator/NY.GDP.PCAP.PP.CD?locations=IN-VN

The only one I'm aware of is The Fame of a Dead Man's Deeds. It's quite interesting, the author got to interview Pierce for many hours at a time, I think he was around for the funeral. Pierce explains a fair bit about his time with Rockwell, about the trouble of forming organizations, selecting recruits, getting the right people... It includes the whole story of his life, his experience with conservatives, the books he wrote.

You can get the book physically or read it on Unz: https://www.unz.com/book/robert_s_griffin__the-fame-of-a-dead-mans-deeds/

True, there are those people who'll spend hundreds or thousands on gacha girls to get their waifu. Still seems like being a sucker.

I gave Genshin a go, there's some fun to be had. Lots of effort went into the game, it's very big and very pretty. But it's not worth paying for more spins on the roullette wheel, as many have remarked: https://youtube.com/watch?v=M5Hfd4wX2GE

Paying for these emotional relationships is still kind of hollow and artificial. Whether it's pokimane or Beidou or some onlyfans girl it's all still fake. The other party doesn't care about you, they care about your wallet. They're exploiting an emotional weakness in a way that a disciplined and discerning man should observe and reject.

The reason I wouldn't be a good financial planner is that I don't really come up with new ideas that often. If I was in an office people would just see me doing nothing 90% of the time rather than busily making new reports. But laziness can work really well. Imagine the stock picker who just said 'buy Apple' for ten years in a row, he'd beat SPY and the sweaty actively trading index fund managers.

Right now I am basically all in on AI and crypto, my theory is that it's still undervalued. I believe that OpenAI is cooking something big, GPT-4 is still a top-tier AI and it's a year old with a few updates. What are they doing with all their huge infrastructure spending if not producing next gen models? Just the other day I saw a paper about how you could push up accuracy by having AI models vote on the right answer, getting the wisdom of the crowd. The bitter lesson of AI scaling is that pumping in more compute beats clever fine-tuning, this is the kind of simple trick that works well.

There have got to be a tonne of killer apps yet to be produced with this technology. AI Dungeon for instance, what happened there? It was running off GPT-3 before censoring down to oblivion, there's clearly a market out there for it. Klarna is replacing its customer service people with bots. We've got Suno in music... Yes, NVIDIA stock went down 10% the other day - lots of people seem to think it's a bubble but I disagree. My AGIX went up 15% (what a brilliant name, AGIX, people are sure to buy in on press releases about AGI!). I'm happy to live with volatility, same with crypto.

I also think fossil fuels are undervalued. I have only a small position there since I think tech is worth more but since all the attention and prestige is going towards renewables, I think coal and gas deserve more love. Yes, everyone in the developed world is racing to decarbonize. But industrializing countries are raving about coal, Modi was boasting about reaching a billion tonnes of coal production: https://twitter.com/narendramodi/status/1774844651394228422

China is building up more coal too: it's a reliable, cheap baseload energy source and you can place it anywhere you like, right next to the factory. It needs to be replaced in the long term of course but replacing coal is hard. Germany's been scrambling to get more coal and they've been a huge investor into renewables. DEI funds loathe coal and universities try to divest... There are also wartime price surges as we've seen with the Ukraine war and energy shortages.

There are some born-in-Russia Russians actively fighting against Russia, that doesn't mean they're not Russian. The commander in chief of the Ukrainian army is Russian! There are also many Ukrainians (in the geographic sense) fighting against Ukraine. This conflict has dynamics of both a civil and interstate war, identity is complicated.

Those who fled to Finland would logically be anti-Russian. The Ukrainians who fled to Russia would presumably be the opposite.

I was going to say... don't you control the troops? Just Lincoln the media if they go against the party line, corporations can eat an Alex-Jones sized fine, unions can be nationalized.

Because much of Ukraine is Russian. They speak Russian. They are Russian ethnically and live in a region historically called Novorossiya. The Eastern half of Ukraine is particularly Russian and there are considerable nationalist feelings within Russia about their co-Russians - which prompted the initial civil war in 2014. Strelkov and his band showed up and joined with locals to fight the Ukrainian army in Donetsk and Luhansk, now annexed. Strelkov is not the biggest Putin supporter in the world, he was imprisoned by the authorities. There's grassroots nationalist feeling in Russia that Putin has to respond to - formerly by suppression and now by encouragement.

The western part of Ukraine actually speak Ukrainian and can't be considered Russian. They hate Russians for a bunch of reasons, including the Holodomor. They sought to celebrate Stephen Bandera as a founding father. The Russians (and Poles) consider him a genocidal war criminal. The new 2014 regime sought to restrict the Russian language and Ukrainize the population, prompting the unrest in the east of Ukraine. Russia does not want a Russia-hating state ruling over large number of Russians right next door, aligned with the West.

Furthermore, the Eastern half of Ukraine is fairly industrialized. In the Soviet era it was supposed to be interoperable with the rest of the military industrial complex, engines for Russian helicopter gunships were made there amongst other things. There's lots of mines, coal and factories, the west is more agricultural. Eastern Ukraine also is the gateway to Crimea which is the most Russian part of Ukraine. Eastern Ukraine controls water and power supplies to the quasi-island. The land bridge and Mariupol region Russia took back in 2022 is key to holding Crimea, also a major naval base.

Those are civilian general-purpose computers, military tech has to use hardened tech against EMPs, they need special secret software and configurations, electronic countermeasures and counter-countermeasures. Can't exactly go to Taiwan, the chips are supposed to be made in the USA for supply chain security... It's super complicated.

t. Lockheed Martin shareholder