Something that has been clear to me in using it, aside from direct claims by the authors, is that Claude is itself vibe coded slop. The number of random errors I get from using various parts of the web UI or CC that should work feels high for such a popular product. But they’re so deep in the vibes that I don’t think they can tell when some path in their web UI is broken. I tried to share a public link to a chat and it asked me to login when opening it on another computer. I tried to download a conversation and it threw an error. When I download markdown output the download succeeds but the UI throws an error. I have tried to control the behavior of Claude Code in tmux using documented flags but I can’t seem to get them to work properly. Agent teams don’t clean up their tmux windows, making the view a mess after they run. Claude code is an amazing product that I love and also it is itself vibe coded slop.

And there’s no reason why they couldn’t vibe fix the issues if there was a process to report the bugs. Fixing issues like that could also be something that’s fully automated. Provided there’s a good test suite (not a given).

It looks very different from the outside than it does from the inside. We are all subject to this.

Does it though? I don't see Canadian or Swiss or Slovak propaganda regularly reminding us that their country's leader is the "greatest ever."

Mark Carney's famous speech at Davos was a breath of fresh air compared with anything ever spewed by the deranged current president of the USA. I am so glad I live in the best country in the world with him as prime minister and that we have no propaganda here in Canada. We will do so much better when we enter trade agreement negotiations with that degenerate loser south of the border in the next few months. That guy can't even ties his own shoes because of his cankles, but Mark Carney can tie not only his own shoes but he always wears sensible socks too.

You may have missed propaganda because you missed the propaganda.

I think you will enjoy this: https://youtube.com/shorts/k3nwW40sYkI

Because some nations are leader-oriented and some nations are system-oriented. Ask any European if they support the state system in their country. Or ask any muslim if their branch of Islam is the best.

Almost all countries in the world will have heavy handed propaganda that their way of organizing things are the best and most fair that could ever exist.

If you think NK an US look the same from the outside check your body temp. Source: am outside both.

No. We aren’t all subject to this. MAGA isn’t even logically consistent. You don’t even need fact checking to spot the bullshit.

And miss me with the inevitable both sides response.

Weren't they all anti war a few months ago and anti pedo before that?

Not sure what you're talking about. We were always at war with Iran.

Most inventions are an interpolation of three existing ideas. These systems are very good at that.

My take as well. Furthermore, most innovations come relatively shortly after their technological prerequisites have been met, so that suggests the "novelty space" that humans generally explore is a relatively narrow band around the current frontier. Just as humans can search through this space, so too should machines be capable of it. It's not an infinitely unbounded search which humans are guided through by some manner of mystic soul or other supernatural forces.

Indeed. Every time someone complains that LLMs can't come up with anything new, I'm assaulted with the depressing remembrance that neither do I.

I can't even find a good example of an invention that is not an interpolation.

The inclined plane, the wheel, shall I keep going?

Stand on a fallen log on a hillside and you'll interpolate pretty hard.

I agree completely, but you forget that another option is that the powerful will use these tools to make us suffer and we will be powerless to stop them.

That's the default option. Power seeks only more power, sharing is worthless to it, except as a temporary instrument. And AI is a perfect tool to concentrate even more power in tiny hands.

You would vote. Everyone else would vote for "their party" regardless of policy cause tribalism. Plus politics always follows the same old pattern "AI took your jobs? We generate food etc so cheaply now it may as well be free? Our robots could be building human shelters for free? Those are _small_ problems, if I get in I'll stop those dirty immigrants coming in! I'll make it so that men act like men and women act like women and they both use the correct bathroom! I'll get rid of the gays, too!"

And it's not even just the right leaning that are tricked. The left get thrown a bone now & then on some trivial thing like getting gay marriage (which shouldn't have even been an argument, more of a realisation) meanwhile nothing is done about corporate tax evasion, improving labour laws etc - or they lean too far into their voterbase and allow rampant immigration, welfare handouts without checks & balances etc.

And all of this because managing a large group of human is pretty much impossible. We need to reorganise into smaller groups.

And look at what that led to. Democracy is not a panacea.

The goal we all seek - liberation - is a distant one. That said I’m skeptical that UBI is the right way. UBI assumes and requires an elite ownership class and a powerful state to force them to share their profits. But as we’ve seen, such class members will organize to penetrate the state and contort it for their own ends. Thus any successful UBI will be a compromise or it will be dismantled by the powerful class that owns the economy.

In my mind, only community ownership of the means of production can truly achieve what we desire. Of course with all distant goals, it is hard to see how we get there. And to be clear I do not mean state ownership.

But I am curious, on my basic point of elite capture of the state, does that make sense?

I am struck that TFA’s title says UBI is “the only way to share”, amusing to me since literally directly sharing is another way. I understand we all have spooky ideas of what that means, but think for example of the concept of library economies. You borrow what you need, but you don’t own it nor have the right to destroy it. We share.

> UBI assumes and requires an elite ownership class and a powerful state to force them to share their profits.

It makes no assumption about an elite ownership class at all. It merely assumes profits, and rearranges how those profits are distributed (away from shareholders, towards labor). There is no need for community ownership of the means of production (though that might have some different benefits, along with some different disadvantages).

You need high marginal (or maybe not even marginal) corporate taxes and a committment to the concept of UBI. Who owns the companies, from the perspective of UBI, is immaterial.

Community ownership does not share the productivity in sector A with workers in sector B. UBI does.

Community ownership does share across sectors if the community owns both sectors. Why would it not?

Also, you haven't really answered the point. You may be able to get this established. But how do you keep it established? How do you keep the elite ownership class from dismantling it? (Based on historically observed behavior, the default assumption is that they will try.) If you don't have a plan that accounts for that, you don't really have a workable plan.

I didn't say that tackling the elite class wasn't important.

But saying that the existence of an elite class implies regulatory capture is a step beyond that.

Regulatory capture is absolutely a problem. While one could advocate for eliminating the elite class (e.g. wealth taxes, confiscation, execution ... as you wish), I'd probably go for tightly controlled political donations & spending, combined with a strong anti-corruption culture (which has been severely damaged by, ahem, recent administrations).

I don't think anybody but those that are really close to the halls of power and have sufficient capital to engage in large scale lobbying is going to be able to achieve regulatory capture. So I suspect there is significant, maybe even perfect over lap between the groups that could achieve regulatory capture and the ones that actually do, and that outside of that group it is pointless to even try. You can get into the club by lucky accident, you stay in the club through regulatory capture.

Regulatory capture requires that laws (or regulations) are drafted that favor your interests. The only ways I am aware of for that to happen are:

(a) sufficient political donations/bribes to get lawmakers to draft suitable language themselves (or via their staff)

(b) a combination of political donations and a worldview on the part of lawmakers in which it is "just normal" for those affected by regulations to draft them, such that you yourself are able to draft the legislation.

There are levels of government where neither of these require incredible levels of wealth, I suspect.

Both could be stopped by limiting political donations and a political culture in which "the chemical industry writes its own rules" is self-evidently corrupt and/or non-sensical.

True, but the USA has institutionalized the power of money in politics to the point that this is now a reality: what would be called outright bribery elsewhere is called campaign donations, there are lobbyists who get to write the laws that favor their paymasters and in fact it has been argued that 'money is speech' (it doesn't get much more bizarre than that to me). What Musk did during the last elections would get you jail in some 3rd world countries, you know, where they take voting serious.

Whether any of these require incredible levels of wealth or not is moot, I think. The reason for that is that it only matters when 'lesser levels of wealth' come up against 'greater levels of wealth' and the latter will always win that confrontation.

If you can manage to believe in a system of 1 person 1 vote for just a bit longer, or maybe even 3 poor people 1 vote, then I think there is still plenty of space for "lesser levels of wealth" to overcome "greater levels of wealth". There are simpler more of us than there are of them.

I hope you're right, but I fear you are not. I guess we'll know by November this year.

I think you make perfect sense. And given that one has to take a cynical take to the writings on UBI knowing the voice of the elite establishment will overwhelm any grassroots thinking since it is actually supported by a financial sponsor to ensure the message is received the world over, unlike say you and I blogging to 7 people. So one can expect most points on it in the public discourse to be biased and in favor of elite-benefitting outcomes. Indeed, when you consider most topics in the media, no one has unique perspectives, they regurgitate the same couple perspectives everyone else does, which are probably crafted by PR firms.

The more I think about UBI though the more I come to the sad conclusion that you can't eliminate money. We can't just give everyone everything they could want for free; I don't think the planet can sustain everyone living like elon musk. So there has to be some forcing factor, some hand on top of the cookie jar that tells your monkey brain that just wants to be high, fat, and orgasming all day that it needs to endure a delayed reward or face some physical exertion to keep the party going. And for better or worse, that forcing factor is by making you have to do something to get credits, that you return for your portion of the produced abundance. This mechanism is able to tolerate the fact that someone might work on efforts not directly tied to any one thing and reap a generalized benefit from all the diversity of that which is produced. Whatever comes next, has to get through that hurdle and the more I think about money the more I find reasons why it is actually a great tool for this sort of distribution of resources and incentivizing labor.

I guess the challenge is that there is a lot of lopsided compensation, where people like say elon musk are paid handsomely even though there is realistically only so much a single human can do. When Elon musk does something that say moves billions in real world dollars on his decision, again this isn't because he is a unique superhuman, just that we have set up power on this planet such where one single person can make a decision to move billions in real world dollars, and if it wasn't musk it would have been someone else in that seat because the seat exists at all. So much power shouldn't be accumulated in one position, because again, we aren't superhumans.

Yep I have an autonomous task where it has been running for 8 hours now and counting. It compacts context all the time. I’m pretty skeptical of the quality in long sessions like this so I have to run a follow on session to critically examine everything that was done. Long context will be great for this.

Are those long unsupervised sessions useful? In the sense, do they produce useful code or do you throw most of it away?

I get very useful code from long sessions. It’s all about having a framework of clear documentation, a clear multi-step plan including validation against docs and critical code reviews, acceptance criteria, and closed-loop debugging (it can launch/restsart the app, control it, and monitor logs)

I am heavily involved in developing those, and then routinely let opus run overnight and have either flawless or nearly flawless product in the morning.

I’m telling it to use red/green tdd [1] and it will write test that don’t fail and then says “ah the issue is already fixed” and then move on. You really have to watch it very closely. I’m having a huge problem with bad tests in my system despite a “governance model” that I always refer it to which requires red/green tdd.

[1] https://simonwillison.net/guides/agentic-engineering-pattern...

Strong disagree. There absolutely are “more humane” methods but they are still not humane. Forced separation of mothers from their children (standard practice) and killing the animals before their natural death are two inhumane features of even “free range” practices.

There is no humane way to kill a creature against its will.

To be clear, free range represents a major improvement over the unconscionable horrors of factory farms, but it is not flatly “humane” without qualification.

Beautiful story, thank you for sharing.