Being able to ask questions to an LLM copilot about patterns across several billion rows, 140GB of time series data, like how the market reacts 15 minutes, 1 day, and 1 week after Federal Reserve press releases, using time_bucket() and other aggregation functions with compressed chunks, and having the copilot generate the queries and run them fast, is going to be game changing. I can ask to write a script to scrape the press releases, to run them through another agent to classify them (hawk, dove) and use that information in the time series analysis.
Even though that exploration has not produced any actionable, tradable results for me, which is why I am sharing it, eventually TimescaleDB will be one of the most important tools in fields like medical research. A researcher will be able to ask in natural language for statistically significant patterns in the data, and LLM agents will query it quickly while taking advantage of all the TimescaleDB optimizations.
I installed Claude Code yesterday after the quality of VSCode Copilot Chat continuously is getting worse every release. I can't tell yet if Claude Code is better or not but VSCode Copilot Chat has become completely unusable. It would start making mistakes which would double the requests to Claude Opus 4.5 which in January is the only model that would work at all. I spent $400 in tokens in January.
I'll know better in a week. Hopefully I can get better results with the $200 a month plan.
Not my experience at all. Copilot launched as a useless code complete, is now basically the same as anything. It's all converging. The features are converging, but the features barely matter anyway when Opus is just doing all the heavy lifting anyway. It just 1-shots half the stuff. Copilot's payment model where you pay by the prompt not by the token is highly abusable, no way this lasts.
I would agree. I've been using VSCode Copilot for the past (nearly) year. And it has gotten significantly better. I also use CC and Antigravity privately - and got access to Cursor (on top of VSCode) at work a month ago
CC is, imo, the best. The rest are largely on pair with each other. The benefit of VSCode and Antigravity is that they have the most generous limits. I ran through Cursor $20 limits in 3 days, where same tier VSCode subscription can last me 2+ weeks
Claude Code’s subscription pricing is pretty ridiculously subsidized compared to their API pricing if you manage to use anywhere close to the quota. Like 10x I think. Crazy value if you were using $400 in tokens.
I just upgraded to the $100 a month 5x plan 5 minutes ago.
Starting in October with Vscode Copilot Chat it was $150, $200, $300, $400 per month with the same usage. I thought they were just charging more per request without warning. The last couple weeks it seemed that vscode copilot was just fucking up making useless calls.
Perhaps, it wasn't a dark malicious pattern but rather incompetence that was driving up the price.
Of course they're pitching it like everything's above board, but from the article:
> While many residential proxy providers state that they source their IP addresses ethically, our analysis shows these claims are often incorrect or overstated. Many of the malicious applications we analyzed in our investigation did not disclose that they enrolled devices into the IPIDEA proxy network. Researchers have previously found uncertified and off-brand Android Open Source Project devices, such as television set top boxes, with hidden residential proxy payloads.
I love how its the "evil" Open Source project devices, and "other app stores" that are the problem, not the 100s of spyware ridden crap that is available for download from the Play store. Would be interesting to know how many copies of the SDK was found and removed from their own platform.
This is what I am thinking about this morning. I just woke up, made a cup of coffee, read the financial news, and started exploring the code I wrote yesterday.
My first thought was that I can abstract what I wrote yesterday, which was a variation of what I built over the previous week. My second thought was a physiological response of fear that today is going to be a hard hyper focus day full of frustration, and that the coding agents that built this will not be able to build a modular, clean abstraction. That was followed by weighing whether it is better to have multiple one off solutions, or to manually create the abstraction myself.
I agree with you 100 percent that the poor performance of models like GPT 4 introduced some kind of regularization in the human in loop coding process.
Nonetheless, we live in a world of competition, and the people who develop techniques that give them an edge will succeed. There is a video about the evolution of technique in the high jump, the Western Roll, the Straddle Technique, and finally the Fosbury Flop. Using coding agents will be like this too.
I am working with 150 GB of time series data. There are certain pain points that need to be mitigated. For example, a different LLM model has to be coerced into analyzing or working with the data from a completely different approach in order to validate. That means instead of being 4x faster, each iteration is 4x faster, and it needs to be done twice, so it still is only 2x faster. I burned $400 in tokens in January. This cannot be good for the environment.
Timezone handling always has to be validated manually. Every exploration of the data is a train and test split. Here is the thing that hurts the most. The AI coding agents always show the top test results, not the test results of the top train results. Rather than tell me a model has no significant results, it will hide that and only present the winning outliers, which is misleading and, like the OP research suggests, very dangerous.
A lot of people are going to get burned before the techniques to mitigate this are developed.
Overfitting has always been a problem when working with data. Just because the barrier of entry for time series work is much lower does not mean that people developing the skill, whether using old school tools like ARIMA manually or having AI do the work, escape the problem of overfitting. The models will always show the happy, successful looking results.
Just like calculators are used when teaching higher math at the secondary level so basic arithmetic does not slow the process of learning math skills, AI will be used in teaching too. What we are doing is confusing techniques that have not been developed yet with not being able to acquire skills. I wrack and challenge my brain every day solving these problems. As millions of other software engineers do as well, the patterns will emerge and later become the skills taught in schools.
It isn't so much as the rest of the world having easy access. It is what the Chinese want the rest of the world to see. If you are in a South American country using a residential IP in new incognito session, doom scroll, after the initial disturbing content, you will start to notice videos of the United States government physically attacking people born in the country of the residential IP address.
The TikTok algorithm in South America. Content about Tiananmen Square and Tibet gets filtered out. Content about the United States government rolling through protesters in armored vehicles, killing people in Venezuela with bombs, and threatening Greenland, straight to top of feed.
The most brutally honest propaganda is always the most effective propaganda.
> Content about Tiananmen Square and Tibet gets filtered out. Content about the United States government rolling through protesters in armored vehicles, killing people in Venezuela with bombs, and threatening Greenland, straight to top of feed.
There's also the degree of relevance. Tiananmen was over a quarter of a century ago. The USA is killing protestors, bombing Venezuela, threatening Greenland now.
The persecution of Uighurs continues apace. Even if it is not allowed to be called genocide on TikTok. The political elements to this are pretty obvious, but conflating two terrible Minneapolis ICE killings in 3 weeks to the horror that occurred in Xinjiang is beyond the pale. While we may go down the authoritarian path with a Clown King, we're still at least 10-15 years behind China.
But since that's all been defunded by DOGE then by your own argument, it doesn't count any more. Ignoring history is a good way to repeat it, just because it didn't happen this year.
Vietnam invaded Cambodia in December 1978 primarily to stop relentless border incursions, attacks on villages, and massacres of Vietnamese civilians by the Khmer Rouge regime. This led to the end of Khmer Rouge, which was a VERY GOOD THING. China was HELPING the Khmer Rouge, so yes, it actually was rather "imperialist aggressiony". I haven't encountered a Khmer Rouge apologist in a long time.
The thing about Russia is they lie ALL THE TIME. The Khmer Rouge was actively conducting cross-border raids into Vietnam killing civilians. Ukraine was not attacking Russian territory. The Khmer Rouge killed 1.5-2 million Cambodians (roughly 25% of the population) in a well-documented genocide. Vietnam ended a genocide and withdrew most forces within a decade. Russia annexed territory and continues occupation.
China had the less sophisticated tools of groups like the Stasi in that era.. 3 weeks of terror was not much more in retrospect.
Americans who are currently protesting should consider if the apparatus will be subtly manipulating their environment not just in the next months or years but from now on with high quality data it will have perfectly categorized mined and will re-mine.
No, it is not normal, but conflating the detention and forced sterilization of millions for over a decade to the current administration's violently unconstitutional over reach is still a bit much.
Does China go around the world invading countries in the name of freedom?
> Content about the United States government rolling through protesters in armored vehicles, killing people in Venezuela with bombs, and threatening Greenland, straight to top of feed.
China: for Taiwan, they are in the planning phase. (Vietnam, Hong Kong, Tibet, Aksai Chin, Korea, Scarborough Shoal do not count in your view of course). Not saying they are worse than the US.
What China did to the Han Chinese makes them worse than ANY other modern country. The great leap forward and the cultural revolution have not comparison. Add in the chinese invasion of Tibet in 1959 and 1979 invasion of Vietnam and they are butchers and imperialists.
You need far better propaganda materials for your "great leap forward" blames in 2026. There were bad policies, but the intention good, it was all about moving the country forward. It failed horribly with huge consequences, that is just the reminder that a full scale industrialisation for over 1 billion people is not something that can be earned easily.
Like it or not, the "Exceeding the UK, catching the USA" (超英赶美) goal of the great leap forward has been overfulfilled under the leadership of the CCP with the help of brutal state capitalism. Everything else is just cheap talk.
Having a full scale industrialization larger than the G7 combined is not something handed to China on a silver platter - those very sad deaths caused by the failed attempts during the great leap forward was a part of the costs.
> and the cultural revolution have not comparison.
The cultural revolution is brutal, nothing should be used to defend it. It is just so wrong. That being said, the west is going through the exact same cultural revolution -
* extremely polarised society with everything is politicalised
* populism taking control
* suicidal policies destroying the civilizational foundations
the difference is 99% Han Chinese consider the cultural revolution as extremely bad, while the west is enjoying having its own ongoing cultural revolution.
if you add the recent woke cancer, the western version of the ongoing cultural revolution is far more brutal.
So you agree that 50mm Han Chinese dead makes the CCP brutal, correct? Every brutal regime thinks they are justified. Ask the US if they think they are justified. Woke is dead. China is much smaller than the west. The G7 is maybe 1/4 of the west. The west includes EU, US, first island chain (japan, s. korea, Philippines, etc), all of the western hemisphere.) China has no allies, it probably doesn't need them. However, China doesn't control its vital sea-lanes. It has less water per person than Saudi Arabia. China has more old people than the west and Confucianism prohibit to "great leap forward" them. China is not escaping the middle income trap.
I think you're being sarcastic, but just in case you're not
> Propaganda is the deliberate, systematic manipulation of information—including facts, half-truths, or lies—to influence public opinion, attitudes, and behaviors toward a specific cause, ideology, or agenda.
Sometimes what you choose to show, even if true, can impact how people see a situation or fact. That is what the OP is referring to. Your quote even mentions that propaganda can be made of "facts" and "half-truths" (a half-truth is usually a fact with a portion omitted to change the interpretation of the fact).
A large percentage of Americans are convinced that police will just shoot them if they happen to feel like it.
Even including ICE in this statistic, you will never even meet someone who knows someone who was murdered by a cop. Police encounters that turn deadly, not even blatant murder, are on the order of 1 in 50,000.
However, that stream of police murder videos are definitely real.
Propaganda is often stoking tiny sparks into large raging forest fires.
> police will just shoot them if they happen to feel like it.
Well that's exactly the problem. There's nothing stopping them: no accountability, no justice. Many cops just don't feel like randomly shooting people, and that's good. The problem is if they do, and even if they brag about it, little will be done.
Take for example the latest Sainte-Soline repression scandal revealed a few months back by Mediapart [1] where videos show dozens of riot cops making a contest about maiming the most people, encouraging one another to break engagement rules, and advocating for outright murder. Everybody knew before the bodycam videos, but now that we have official proof, we're still waiting for any kind of accountability.
If i go around and shoot people, there is no way i will avoid prison. If a cop goes around and shoots people, or strangles people to death, prison is a very unlikely outcome.
> you will never even meet someone who knows someone who was murdered by a cop
That's not how statistics work. Police abuse tends to happen in the same low-income social groups (and ethnic minorities). As an example, living in France, i've met several people who had a family member killed by police. Statistically unlikely if i only hung around in "startup nation" or "intellectual bourgeoisie" circles, which is not my case.
Being killed by police is different than being murdered by police.
Police in the US kill somewhere around 1000 people a year. But of those, it's something like 5-10 that are murders. There is maybe 1 every few years where the cop is itching to shoot someone who is clearly compliant and not a threat.
The 990 police killing videos that become available every year now are not particularly compelling, because its bad actors trying to kill police and getting themselves killed.
Sorry, I don't know anything about France and police though. The US has a different dynamic because guns are everywhere, especially where crime is. Every cop knows about the ~50 cops who are killed by guns every year.
The dynamic doesn't look very different here, at least from reading the news. I don't know about the US (though i suspect <1% murder out of all police killings is a gross under-estimation), just for anyone's curiosity, in France police killing of a threatening person is the outlier. [1]
We don't have guns circulating freely around here (though some people have them such as for hunting). Many police murders take place in police custody (such as El Hacen Diarra just this month). According to the most comprehensive stats i could find [2], out of 489 deaths by police shootings (1977-2022), 275 victims were entirely unarmed.
[1] Not very scientific method: any case of police being assaulted and using "self-defense" is widely spread in the media, and those few cases per year don't account for the dozens of deaths every year.
>though i suspect <1% murder out of all police killings is a gross under-estimation
It's easy to track because anytime it happens it's instant major news on the internet. Trust me, in the economy of social media clout, few things rank as valuable as police murder.
Pretti was frontpage of reddit within 30(!) minutes of being shot. Even without bystanders there is a whole group of creators whose whole channel is combing bodycam footage for wrong doing. These videos are worth (tens) of thousands in ad views if nothing else.
Some, maybe all, ICE agents were body cams, but I haven't seen any footage. I'm not sure what the process would looks like, this whole ICE violence thing is only a few months old, whereas most regular police have had bodycams for 5+ years now and getting the footage is well established.
Police also definitely don't turn it off when it suites them, although some have, but again, it's a Streisand effect when they do. I really cannot stress enough that police doing bad things has extremely high monetary value for the people who find it, and you also get paid for the crazy bodycam videos you find along the way. If you're a cop and you turn off your cam before breaking the law, you are almost certainly going to be the face of a 1M+ view youtube video. People, like yourself and me, gobble that up.
It doesn't matter much anyway, because there is 100x more footage of cops doing bad things with their cam on.
> Even including ICE in this statistic, you will never even meet someone who knows someone who was murdered by a cop.
Uh.. I know someone who was murdered by cops while having a bad LSD trip (not violent, just incoherent). He was restrained too tightly despite protests of his family, loaded into a vehicle, and suffocated to death.
> Police encounters that turn deadly, not even blatant murder, are on the order of 1 in 50,000.
>Even including ICE in this statistic, you will never even meet someone who knows someone who was murdered by a cop. Police encounters that turn deadly, not even blatant murder, are on the order of 1 in 50,000.
That just shows that people's social circles aren't that wide. 1 in 50,000 is rare in your personal bubble. For a town of 1 million people, thats 20 people.
Sounds tiny, but if we were to line up 20 people and have them murdered by law enforcement, it'd pretty much end the careers of anyone in that chain of command. Because that's not a behavior you want to let spread and expand.
I am not being sarcastic at all. It is a common misconception that propaganda means lies. Propaganda is information designed to get you to believe a certain thing or feel a certain way. The best propaganda uses entirely truthful statements to manipulate your beliefs and emotions.
One of the best examples of this were the endless photos and information about stocked store shelves, filled with fresh goods at dirt cheap prices, during the Cold War. In general truth is the best propaganda, because when you lie there's always a rubber-band effect when somebody realizes, sooner or later, that they've been had.
It's using information to influence public opinion in a calculated manner. Said information can include facts. It can even be entirely factual.
Manipulating the feed of a social media website for the purpose of swaying the viewer's opinion is a cut and dry example of propaganda. Doesn't matter who does it or whether the information displayed is factual or not. Those things make zero difference.
This really doesn't pass the sniff test. It reminds me of a recent post I saw: "what are movies people like only becsuse it is good?", calling it "quality slop". It's contradictory.
If people are given a wide perspective of a situation and adjusts bias for the Overton window (aka, we don't let Nazis have an equal platform to a more progressive group), then we just call that good reporting. The act of convincing people isn't inherently a bad thing. How you do it matters a lot.
You're subtly misattributing me though. "Convincing someone" is a superset which contains intentional manipulation of the information someone is exposed to but also lots of other things.
As you said, how you do it matters a lot.
You've also gone and (IIUC) equated the general biases of an outlet with propaganda which I certainly wouldn't agree with. They're similar, and the former can certainly morph into the latter, but they aren't the same thing.
That can be part of it, but usually it's not necessary - certain facts, or certain aspects of facts (e.g. exclude some context) can just get exaggerated to have the desired effect on a larger population.
That might be, but if it's amplified through social media it becomes propaganda.
Example, 99% of people are normal, but if all you see is the 1% that isn't you'll start to believe more than 1% aren't normal. Especially if that 1% is of a recognisable ethnicity / religion / background. This is why there's a shift to the right.
I am responding to the fact US TikTok does not show videos of an armored vehicle driving through a crowd of protesters standing in front of it like the lone man in Tiananmen Square. They are being removed.
This ability to control what information TikTok users are presented with is the reason TikTok was originally banned in the United States.
I am being objective discussion how TikTok is being used as a propaganda tool whether or not I personally agree with China influencing people in South America or whether or not what the United States government is doing to protestors is good or bad. I'm not putting a value on it. I'm pointing out that when I'm in South America and someone links a video in a text message and I start to doom scroll after a while I will start to be introduced to videos of the Unites States government committing violence against Spanish speaking people.
> might be immensely popular in South America
Objectively the current United States regime was hugely popular in Spanish speaking countries like it was in Spanish speaking Florida. Up until a couple months ago, people would tell me how much they support and admire the current regime in the United States. That has changed recently which likely has to do with the content they receive via TikTok which is controlled by the Chinese government which is why it was banned in the United States. After being sold, it is not surprising that the United States is using it the way they accused the Chinese of using it.
On mastodon, with the non-algorithmic feed, following mostly accounts that aren’t particularly political, those things are still at the top of the feed. If you’re not seeing those topics at the top of your feed you’re probably being misled by your algorithm.
Another reason why feed ranking algorithms should be published. If we can see the algorithm we can stop playing these yes/no games. The real enemies are social media companies, not the other side of politics.
> Content about the United States government rolling through protesters in armored vehicles, killing people in Venezuela with bombs, and threatening Greenland, straight to top of feed.
Aren't these recent events? A better example would be showing US atrocities from the last 50 years, but not Chinese.
Or hiding the suffering of Ukranian and Iranian peoples.
If I doom scroll TikTok without cookies from a residence in South America, after a while, I will be presented with anti American propaganda showing videos of recent events or people speaking in Spanish about the atrocities that the United States is committing against Spanish speaking people that is recent.
I'm am describing objectively what I see.
The United States didn't want TikTok controlling what is visible to people in the United States so they banned TikTok. Later the United States offered allowing it to be sold to an American company.
Currently, there are two extremely influential forces for people under 25 years old in Spanish speaking Latin America, TikTok, a Chinese company, and an American music artist, Bad Bunny, who likely is the single most influential person in the Spanish speaking world. Let's stay tuned for the Superbowl.
I think most media is talking about the mess the US is in with ICE right now. For what’s worth I am in Europe and on X more than half of what I see is about American cops and ICE , most against ICE but some in support of it.
>It isn't so much as the rest of the world having easy access. It is what the Chinese want the rest of the world to see.
If your prosperity depends on using technocracy to deny 1.3 billion people the ability to communicate and share ideas with your citizens, a few things are true:
1) You have created a digital iron curtain
2) You are doomed because information wants to be free
3) If you succeed the result will be war, the only thing left when communication breaks down
I think some people live in movies where the bad guy always loses. Reality doesn't work this way. Bad situations where information is denied from people can last lifetimes.
With modern technology we may be creating systems that end up imprisoning our minds for generations with no escape because you'll be killed the moment your technological monitor realizes you're going to fight back.
"Information wanting to be free" is a concept less rooted in idealism and more in a cynical view of human nature. Even the most closely guarded secrets eventually leak, and as the utility of knowing the secrets increases, the pressure to leak also does. The physical universe itself appears to favor disclosure and abhor secrecy.
>With modern technology we may be creating systems that end up imprisoning our minds for generations with no escape
That has been the goal of authoritarians for a long time. Orwell's vision of it involved obliterating even the capacity to think or speak about anti-state themes.
I see people saying this a lot, but I've also seen videos demonstrating that you can easily post and search for Tiananmen Square content. I don't use Tiktok myself but it seems like this is basically untrue.
This will likely depend on the country, I presume it wouldn't work in China.
But this isn't new either, western services operating abroad will often comply with local laws, which includes country or region specific laws on acceptable content. Google pulled out of China for a good while because they didn't want to, but they eventually cracked and complied with their content laws. Of course, by then the competition was dominant already.
key word is "search," tianamen square will never be recommended in a feed. This is the illusion of "choice." Most people think they can "train" their feed, this is not true.
I'm confused. I thought there was Douyin in China and TikTok for the rest of the world. TikTok used to be under Chinese control but now is essentially under US control. Isn't western TikTok a single entity?
The news only dropped about 5 days ago about the US partnership. Its still a Chinese app. Now the deal with Oracle will have them designing the algo, storing US users data, and doing US moderation. It wasn't this way before.
Nah, the writing is on the wall for a long time and they nearly got shut down several times.
I can’t imagine that the permission to continue operations came without major concessions.
You may wish to remove the ?si=… tracking string from your URL. It might also be worth editing in some context: right now, it's a bare YouTube link (which I don't particularly want to click on). Is this footage? A video essay? A pop song?
No I was referring to the original comment twisting words. They intentionally invoke the horrors of dictators running over protesters to disperse them.
Your comment showed the actual event, thanks. It showed an armored vehicle slowly pushing through protesters to gain access to something.
Not remotely comparable scenes, so I must assume intentional twisting of words. My comment was written sloppily.
I honestly don't understand why you posted the Narcissist's Prayer; however, to clarify: the original comment I responded to referred to another comment that said, "the United States government rolling through protesters in armored vehicles," to which I replied, in part, "it's a video of a WCCO news (local MSP TV station) segment which shows an armored vehicle pushing protesters out of the way."
The person who said I was twisting words said (emphasis mine), "There’s an enormous difference between driving slowly through a crowd of protestors with no injuries versus running over protesters with a tank."
At no point did I nor the comment I responded to use the words, 'running over protesters with a tank'.
I don't think anyone is claiming that the specific famous tank man of Tiananmen got run over by the tank. However, there is plenty of evidence that people got "mowed down by tanks" at Tiananmen Square: https://www.theglobeandmail.com/world/article-why-is-the-arm...
The US has been applying soft power and hard power in South America - to put it euphemistically, as the most recent US intervention was just days ago - for close to a century. The Chinese... haven't.
Why should people in South America give a shit about Tiananmen or Tibet and at the same time not give a shit about the escalating authoritarian grip of the US regime, which is infinitely more relevant to their lives?
How can you say the Chinese "haven't"? They've been using soft power for some time with Venezuela. They've been importing Venezuelan oil. They have been making loans as well. The loans a are a huge part of "soft power". They've also replaced a lot of items impacted by Trump's tariffs from South America.
Things the US could also do if it unsanctioned them.
I threw my computer off the balcony. I look at a web design business. "No fair!" I think to myself, "if only I had a computer I could have a web design business too!"
The US could? The US has used soft/hard influence for a long time. We've crop dusted Colombian fields attempting to eradicate cocaine. We've eliminated leaders in Central and South America. We've influenced elections trying to get specific leaders elected. We've sanctioned the shit out of one little island, we blockaded the island when it was allowing itself to be used/influenced by another government we didn't like. We've allowed US corps to invest and build infrastructure within these countries. We've given them millions/billions in various ways including straight cash injections.
I don't know how much more would need to be done for you to think things are being done.
The U.S. government has not publicly presented any concrete evidence showing that TikTok has actually been used to influence US public opinion in line with CCP policy.
If I was a foreign government I would promote division. For the left promote anti-center truth. For the right, anti-center truth. For the center, anti-wing truth. Recommendation systems do this automatically, they are inherently anti-social. This power needs to be controlled domestically were we can force changes to algorithms if needed.
How about just letting the user choose, instead of foisting your own idea of 'right' on them.
If I was the US blessed feed, let me have it.
If I wasn't the Chinese maintained one, why not.
Or, even better, let me make my own! Or use one from an open source that I, the user, trusts.
Hell, EXPOSE THE ALGORITHMS. The simple fact that we can't see the weights, or measure inputs to outputs, means we are in total control of whomever currently holds the reins, and they can literally play God behind the scenes if they have control over enough eyeballs.
Wasn't there something about the algorithm pushing brainrot to US audiences while Chinese users got more educational/high quality content? Turning Americans stupid might count.
They said "concrete evidence". Have we also considered that US consumers seek out brainrot, so the algorithm gives them what they want? How is that different from any other US-owned social media?
China has media laws that would make much of what appears on any sort of Western media platform illegal, so they're obviously going to get a very different experience in China. From anything that might violate social ethics, to clickbait titles - all illegal in China. They've even cracked down on overly effeminate men - 'girly guns' [1] and a million other things I'm not listing here. Basically Western style social media simply is impossible there.
In any case, entirely Western oriented platforms also push brainrot to Western viewers, so I don't think there's any conspiracy so much as just cultural differences.
China is less interested in turning Americans into carriers of the red banner, and more interested in sowing political discord and instability. Just like Russia was doing in 2016, creating faux Bernie rallies and organizing them across the street from faux Trump rallies.
Sounds like you're in agreement with the parent - outside the US, people see content that reflects poorly on the US, and which is blocked for US citizens
> The painful to answer question is whether the intention is to block the spreading of lies or the spreading of truth?
What it should be about is preventing someone else from blocking the spreading of truth.
"Block the spreading of lies" is something authoritarians say when they want to declare any criticism of themselves to be a lie and censor it. You can't block the spreading of "lies" without ordaining someone as the decider of truth and there is nobody you can trust to have that power.
But if we were actually doing what we should then what we would be doing is developing censorship-resistant uncentralized systems rather than fighting over the keys to the censorship apparatus.
When you have a personality disorder like NPD, you'll believe to your core that every criticism of you is a lie.
When you're in an abusive relationship they say intentions don't matter, only impact does. Because victims often focus on the intentions of their abuser and stay in the cycle of abuse.
Let me repeat it, intentions don't matter, only impact does.
We should let people know how bad politicians are. If everyone knows every time a politician is a mass murderer, it might provide an incentive for politicians to stop mass murdering people.
The general problem is that people think based on relativity.
Suppose there are thousands of law enforcement officials in the US, some minority of them are violent offenders and as a result of that some minority of police shootings are murders rather than legitimate self-defense or protection of the innocent, where the number of annual illegitimate police shootings is somewhere between 2 and 999, and the propensity for those people to be prosecuted is lower than it ought to be. Suppose further that China has over a million Uyghurs in concentration camps and is using them as slave labor and subjecting them to forced sterilization.
Is the first one bad? Yes. Is it as bad? Uh, no. But you can present a distorted picture through selective censorship.
Obviously what you want is for neither of them to be censored, but not wanting a foreign power to be the ones who decide what people see is fully legitimate.
> Obviously what you want is for neither of them to be censored, but not wanting a foreign power to be the ones who decide what people see is fully legitimate.
It's less legitimate when you don't want a foreign power to be the ones who decide what people see on their own platforms. The US for example shouldn't dictate what US users see when they visit www.bbc.co.uk
The just US got mad because a Chinese owned/operated social media platform got massively popular and they just wanted the ability to control and censor it.
"Their own platforms" is the flaw. Countries and companies shouldn't "own" the means of mass communication to begin with.
How the feed is filtered should be a fungible commodity that anyone can swap out for themselves or offer to others without sacrificing the network effect, because the network itself shouldn't be owned.
Notice that the US doesn't censor bbc.co.uk, because the web is a decentralized system. But then ordinary people end up on Facebook or TikTok, which isn't.
I mean, you're not wrong, but there isn't very much nuance here.
I think there are a number of things occurring all at once and it's going to lead to the destabilization of most democracies (which China is a big winner if this occurs).
Democracy has never really been as free as the people living in democracies believe. The rich and large media entities have always controlled the vote with much more impact than the actual issues individual voters had.
If you believe this previous statement to be true this leads to a number of issues in the modern world.
One is that previous to now most countries demanded some kind of local media ownership, so the message would be more aligned with someone living in the country rather than some other entity (not perfect, but still better than nothing).
Another is media groups tended to be smaller and more fractured. They may hold conflicting opinions on things.
Which bring us to now, with huge foreign media organizations holding massive sway over gigantic audiences. This isn't just about China over the US, it's just as much about the US over many EU entities. These are potential powers that can change course of the world and they have governments behind them directing them where to go.
Also don't forget the US absolutely loves to control what gets in the media. The right in the US didn't just start brining up socialism and communism yesterday, it's been a control mechanism on what can be published and what you can see for over 100 years.
> I feel like Claude Code is starting to fall over from being entirely written by LLMs.
The degradation is palpable.
I have been using vscode github copilot chat with mostly the claude opus 4.5 model. The underlying code for vscode github copilot chat has turned to shit. It will continuously make mistakes no matter what for 20 minutes. This morning I was researching Claude Code and pricing thinking about switching however this post sounds like it has turned to shit also. I don't mind spending $300-$500 a month for a tool that was a month ago accomplishing in a day what would take me 3-4 days to code. However, the days since the last update have been shit.
Clearly the AI companies can't afford to run these models at profit. Do I buy puts?
> Solar prices in the US are criminal, protecting oil and gas who bought all the politicians.
It would be worth including control of the people who vote for the politicians by direct investment such as when the oil producing Saudis bought the second largest stake in NewCorps which controls FoxNews controlling the content that influences voters. And, less than ethical control using bots on social media by Russia.
A lot of what influences "solar prices in the US" is controlled by foreign oil producing countries like Saudi Arabia and Russia controlling content and media consumed by American voters.
Here in California, they drastically cut back on the price that you get for solar powered electricity from homeowners. It used to be around $0.30/kWh at any time of day and now it's can drop to $0.00-$0.05/kWh during the day when the state is sunny. If you can afford to have a battery installed, the rates are far better as you can either run off the battery when rates are the highest in the evening, or you can export it back to the grid when prices are much higher.
The price is signaling that additional solar power production during the day isn't very useful; and additonal solar power production in the early evening when demand is high and the sun isn't shining and you need a battery system to have already been accumulating energy during the day is useful, albeit more expensive and complicated to build and run.
With falling battery prices this should be an addressable problem. Soak up the locally generated excess energy and sell it later in the day when the need is there. Electrical arbitrage seems like a win/win solution for the utilities their customers.
The first round of people paid way more for their solar panels though, and those higher prices helped bootstrap the industry. Should people who paid much less for panels get the same reward? I'm having trouble getting outraged about this, it seems to be incentives working exactly as they should.
I agree, and maybe my "ladder pull" comment comes off as too negative. Most early solar buyers were either in it for environmental reasons or for a modest return on investment. I don't think many were expecting a windfall.
Residential solar is completely counter-productive right now in california. Just take a look at the CAISO price maps during the day when the sun is shining. There's so much power they are paying people to consume it. It's a negative force for grid stability. Getting paid for making the grid less stable is ridiculous. Until there is widespread battery storage or massively improved transmission and distribution systems grid-tied residential solar is a solution in search of a problem.
I agree wholeheartedly, and the technocrats are complicit with the GOP here.
It's funny how “free markets” keep producing the most expensive solar prices in the developed world. Don't get me started on Healthcare (I just moved back to the U.S. a couple years ago after 18 years in Canada, what a cluster*ck).
Oil and gas buy politicians, foreign oil money buys media influence, and social-media bots keep voters angry at the wrong targets.
Saudi capital helps shape the messaging, Russia helps amplify the noise, and Americans get stuck paying more for clean energy while being told it’s patriotic.
If even a Democratically-led California is doing this, how can you point fingers at just the GOP? It's endemic to the system, and not restricted to just one party.
Republicans are always trying to increase and protect oil subsidizes, cheer "drill baby drill", and have social media tools peddling their bs who are funded by foreign influence campaigns. The Democats may not be perfect, but they are much more likely to cut oil subsidies or at least subsidize solar and other renewables to balance. This is especially true as the next generation takes over by primarying the tepid fossils currently in office. Meanwhile, dear leader Dumpty likes to suggest windmills cause cancer because he doesn't like what they look like on the horizon of his golf dumps. Saying this is a "both sides" problem is laughable.
Yup. 46.8 GW photovoltaic in California and 22.8 GW photovoltaic in Texas. Over twice as much PV in California even though they have higher standards for new construction in general than Texas does.
The US doesn't have a free market in either health care or electricity generation. An actual free market in solar power would probably result in more or less what we are seeing with the actual highly regulated market in electricity, namely extremely cheap prices for additonal solar energy in the middle of the day when the sun is shining, higher prices for additonal solar energy in the evening when demand is high and the sun has gone down, and some fixed cost to pay for physical electric grid infrastructure that needs maintenance regardless of whether it is being used at any particular moment.
Oil and gas don't buy polticians more than any other industry does, but voters do get particularly angry at politicians when the price they pay for energy suddenly spikes.
The list of the oil producers listed and omitted on a given forum in these contexts is always interesting. On HN it is often SA or Russia, and almost never Qatar or Iran.
My favorite example of this is RFD TV [0], a tv channel dedicated to farmers.
The name came from Rural Free Delivery, which was a program by the USPS to ensure farmers in rural areas received mail because it was not profitable for private carriers to deliver in remote and rural areas.
Around 2010, realizing that the message of a government program was good for farmers they completely white washed and removed the concept of free [something] from big government being good from all branding, websites, and tv programs.
I worked with Mastra for three months and it is awesome. Thank you for making a great product.
One thing to consider is that it felt clunky working with workflows and branching logic with non LLM agents. I have a strong preference for using rules based logic and heuristics first. That way, if I do need to bring in the big gun LLM models, I already have the context engineering solved. To me, an agent means anything with agency. After a couple weeks of frustration, I started using my own custom branching workflows.
One reason to use rules, they are free and 10,000x faster, with an LLM agent fallback if validation rules were not passing. Instead of running an LLM agent to solve a problem every single time, I can have the LLM write the rules once. The whole thing got messy.
Otherwise, Mastra is best in class for working with TypeScript.
I learned that every step that can be solved reasonably without an LLM, should be solved without an LLM. Reliability, cost, performance, etc.
I try to transfer as much work as I can out of LLMs and into deterministic steps. This includes most of the “orchestration” layer which is usually deterministic by nature.
Sprinkle a little bit of AI in the right places and you’ll get something that appears genuinely intelligent. Rely too much on AI and it’s dumb as fuck.
Make their tasks very small and simple (ideally, one step), give them only the context and tools that they need and nothing else, and provide them with feedback when they inevitably mess up (ideally, deterministically), and hope for the best.
Time to start a Drupal consulting firm again.
reply