aqouta
Friends:
User ID: 75
aqouta
15hr ago
I take the critique of the American model fully, I've never liked it very much and have seen it as mostly downstream of other things I don't like, namely a refusal to just administer IQ tests and deal with the disparate impact. But when I look at whatever europe has had as an alternative I don't really see anything better. If I wanted to judge a system by its fruits I'd look to the east, which unfortunately also has some notable problems with how its elites are selected. I think perhaps the best method would be to leave it up to capitalism to incentivize refinement and reward merit with power, a very libertarian direction of thought.
At work we're stuck with VS code/github copilot which is not ideal but allows customizing/spawnimg subagents and tool calls which is the big requirement. Agents themselves aren't a big deal, they're essentially just custom prompts, they become important for being created as sub agents to isolate tasks to narrow contexts. There's some customization to your own env you should do but you can just start by asking Claude/gpt to spin up some basic ones that you can tweak over time. Basically any time I notice it get stuck or need to guess and check a bunch I have it makes some tweaks to the documentation. I'm cooking up a method to automate that process. Where I'd spend my time first is making some guide files for your code base. Putting together some easily greppable documents outlining our database schema greatly improved its ability to interact with it.
Today was our biweekly demo day. One of our engineers showed their work in putting together a harness for aiding in translating a winform app to a rest site + web front end. They built an orchestrator, an array of custom agents designed to handle the specifics of our environment, custom skills for understanding and interacting with our db ect ect, dozens of files. And as they walked through their prompting it became clear to me that they never actually invoke the orchestrator so they were in fact just using the vanilla agent.
It's obvious to me that we're like a couple more releases from all this work not being necessary, the future tools will simply as a matter of course customize themselves to the environments they're exposed to. But there is as of now an art form to getting really good results from current models. I'll say the most important concept is something like optimizing for "context density", you have ~1million tokens to work with but every marginal token long before that degrades performance while relevant context improves it. So you need to balance it out, using sub agents to offload discreet tasks and provide maximally dense reports. I have oracle agents that simply returns true or false, a single line, or a full report depending on what is asked of it. Of course this works even better if you can have the thing go through and summarize your code ahead of time with the intention of minimizing the guess and check nature of looking through your repo, every wrong search pollutes context. Or you can just poorly write out an ambiguous two lines, throw the vanilla nerfed agent in the deep end, have it go through 5 iterations of "compacting context"(i.e. throwing away important bits of information because you've hit the hard limit) and get back a sub par response then laugh about how dumb these hype monkeys are, maybe they're just so terrible at coding that this half assed demonstration is impressive to them?
Whatever dude, enjoy 6 more months of ignorance before it's impossible deny, you'll deny it anyways, not my problem, I tried to warn you.
aqouta
1d ago
It's not known for sure, the labs do not say, but it's very likely that sonnet 4.6 is a distillation of opus 4.6, essentially training a weaker model on the signal of the stronger model. What is definitely known is that sonnet 4.6 is considerably weaker than opus 4.6, given opus tokens cost 5x as much as sonnet tokens this should be clear. Sonnet is great for rote work, my admin agent that handles git and jira calls uses sonnet, but I'd never use it for core work.
"You're not using the latest model, just one more model and we'll reach AGI"-bros officially in shambles after this one.
...
This was with the latest version of Claude Sonnet. We don't have access to the latest version of Opus
Come on dude, you can't be serious.
I honestly have tired of retreading this conversation, the skeptics are immune to evidence. Even the people on my team who are enthusiastic about AI are configuring and using it poorly. The difference between prepping the environments, having agents go through and document the code base before running an orchestrated agent with dedicated subagents and just yolo throwing an agent at it - and I half expect you didn't even run the damn thing in agent mode - with a half cocked vague prompt practically hoping it fails so that you can own the ai bros is massive. You don't want it to work so it won't. I know this isn't convincing to you, I've tried being convincing to you guys, even after you're out of a job you won't be convinced. It's pointless.
edit: slightly less run on sentence(slightly)
aqouta
5d ago
If we were talking about tactics or even grander strategy this argument makes sense, but war aims? No, we're supposed to be consulted about the war aims. You can have congress agree and declare war based on confidential plans but our representatives get to see the plans and agree to them, not one man.
aqouta
5d ago
And, if you wanna go for right makes right morality, then surely you'll be perfectly fine if one day the shoe is on the other foot and the Palestinians achieve military supremacy?
Call of hypocrisy go both ways, if this comes to pass surely you likewise would be outraged if the Palestinians expelled the jews or had a legal system that treated them differently to the Palestinians?
aqouta
5d ago
It’s the threat against Israel
I think this is a factor, but certainly not the only factor or even the biggest factor. Western nations have been getting involved in the sandbox for centuries before Israel came into existence. It's a place with oil resources and a travel route for a preposterous amount of commerce and a geography that lends itself to sectarian conflict. If israel was established in brazil or wherever else they were considering I think we'd still be seeing conflict in that region and I expect the US as hegemon would be involved.
aqouta
7d ago
He has now blocked me on X : /
I don't know how to put this in a way that won't be interpreted, incorrectly, as condescension but I think he's going through a rough time and hope he can get through to the other side alright. He wrote some posts that I cherish and arguing with him in long form helped me think particularly about the AI race with China even if I don't quite agree with him on the conclusion.
aqouta
7d ago
It's Twitter that has fried his brain. Like many he has succumbed to the pattern where he spends his time quote tweeting the most deranged examples of his outgroup which the algorithm recognizes and then serves him more of. He's built a machine that serves him almost nothing but idiocy that he can happily dunk on. He's built up a heuristic where he pattern matches any argument made to something stupid he's seen on twitter posted by someone who he didn't need to use any intelligence to dismantle; his wit and charity have atrophied from disuse.
aqouta
9d ago
It's my belief that if you grant these three things then you have to conclude that the labor theory of value is false.
- Risk isn't labor
- labor theory of value only counts labor input as value
- Risk is an unavoidable factor in the production of things society values
aqouta
10d ago
Sure, I think there were some downsides that weren't adequately accounted for. Although there really isn't any law of the universe that says we need to allow sports betting in particular. I'm just saying the rationalists were right on the upside promise, it's played out just about as expected.
aqouta
10d ago
There are many products that have a risk component that we as a society value. No rational agent would ever work on something with a risk element if that wasn't represented in the final valuation of the product. Therefore a labor only theory of value that doesn't include risk doesn't actually track what society values.
aqouta
11d ago
What if they're not interchangeable? From your model it would never make sense to work on an product that has any chance of failure, and a lot of products intrinsically do.
aqouta
12d ago
If two products both take one hour to make from the same inputs, but one is destroyed half the time and the other one isn't then is one more valuable than the other?
aqouta
12d ago
they'd provide market signals about how likely major events are to happen.
They do this successfully though, and I don't think rationalists ever really claimed there would never be frivolous markets, that just isn't the part they really cared about. There are relatively high liquidity markets for a lot of topics that I find more credible than any expert claims.
aqouta
13d ago
We have more outright nazi sympathizers on this site than people who could be accurately described as supporting the extermination of the Palestinians. This is ridiculous hyperbole.
aqouta
13d ago
I went off Israel in a big way after Oct 7 when the biggest contingent of pro-Israelis on this site started just outright saying, 'look, it's time to exterminate the Palestinians now'
This did not happen, I was there.
aqouta
13d ago
Trying to salvage LTV in any form just seems like a mistake. It's a lens that exist to glorify the worker and stoke resentment against the wealthy, it falls apart when you put any pressure on it. It can't explain scarcity as value, as put in one if its earliest critiques "Pearls are not fetched from the bottom of the sea because men have dived for them, but men dive for them because they fetch a high price." and I don't think your energy theory of value would survive much better. I think it's just cleaner to say that everyone has their own theory of value and there is no universal theory, which is a source of great surplus from the market.
aqouta
14d ago
This is a generalized argument against all large claims, that is to say it proves way way too much. Some claims are true, some claims are false, you need to actually address each on their merits.
aqouta
20d ago
Every dead child has a father in the fedaykin or a cousin in the deep desert resistance or a neighbor who was neutral until yesterday. The Fremen do not need to match the Sardaukar in firepower. They need the Empire to keep making their argument for them.
I'd imagine the Fremen made a few enemies with their own even less targeting bombings. This is not a sustainable path to emancipation, just the spread of chaos, a festering wound that longs to kill its host for daring to apply stinging ointment.
aqouta
21d ago
Right, having some women who are friends is definitely useful in a number of ways. Maybe I misread the OP and they weren't suggesting it as a main strategy, but there really is a strain of thought among women that a man who would be interested in dating but not becoming friends is some kind of contradiction. "why would you date someone you wouldn't want to make your friend?". It makes a kind of sense from the perspective of the selective sex but it just isn't workable.
aqouta
22d ago
One thing that should also be added here is that you have to be comfortable genuinely being friends with these women (not just a friends to get in your pants kind of deal.), and be comfortable with the possibility that it wont go in a romantic direction. Even if it doesn't go that way, you made a connection that's valuable in its own right, and you may be able to date other women she is in proximity with.
The proximity with this is important, I met my wife by getting set up by a friend of hers that I met on a dating app but didn't hit it off with. But the idea that single men who are serious about making a partner should settle for a friendship with women they meet is just an absurd delusion some women harbor who haven't ever seriously thought of the logistics of single men dating. I'd need to have maintained literally hundreds of female friends by the time I met my wife for this to have been a plausible strategy, it just doesn't really work. I'm sorry but if it doesn't work out you can't expect him to stay friends with you, it just doesn't scale. It's not personal, it's just that forming a strong attachment, getting stuck in the friend zone as it were, and then getting rejected in the end eviscerates a portion of your soul each time.
Iran isn't Syria. It's really not clear that an Iran war causes Iranian refugee waves. Iran is a big place with a more middle income economy, so far we've been seeing internal migration to more rural areas
More options
Context Copy link