dr_analog
top 1% of underdog fetishists
No bio...
User ID: 583
Source
napkin math. let me see why I'm so off from the GAO report
UPDATE: I don't quite understand the takeaway from Figure 11. How does interest on the debt growing to 27% of federal "spending" mean we would find the revenue to keep paying it? Isn't it more clear to say that our liabilities would grow such that 27% of them are interest payments alone?
In 2025, we already currently deficit spend $2t to meet our $8t a year in obligations.
What figure 3 of that report makes clear is that the size of our interest payments will grow out of proportion with revenue growth. That's the scary part IMO. It's already happening and I think even higher GDP growth + the other politically impossible things I mentioned in my OP wouldn't help us catch up with it.
dr_analog
top 1% of underdog fetishists
2mo ago
Yes, before, if China approached your country and said you must take your pick: the US or China, you would probably pick the US.
After all of this unhinged clown shit, if the US says you must choose between the US and China, it's a tough call!
dr_analog
top 1% of underdog fetishists
3mo ago
I understand it's somewhat an open secret that Russia's subs are confined to near-Russia and the US actively tracks them and can pre-emptively obliterate them the moment things get hot.
I understand it's somewhat an open secret that Russia's subs are confined to near-Russia and the US actively tracks them and can pre-emptively obliterate them the moment things get hot.
dr_analog
top 1% of underdog fetishists
3mo ago
The theme is marking the occasion of the SOTU by taking inventory of some of the broken promises, incompetence, lies and hypocrisy in just six weeks.
dr_analog
top 1% of underdog fetishists
3mo ago
Because of a treaty that isn't worth the paper it's printed on. You know this stuff.
I don't follow. If it's not worth the paper it's printed on why does the "Because" happen?
She’s doing so because she finds “the process of being pregnant and birthing joyful in and of itself” instead of for money but that doesn’t make it better.
I just mean it is not an immense sacrifice to some people that they might only consider because they wanted to get paid or had some other kind of illegitimate gain from it or wanted to pervert the social order in some way.
Some people, friends or family, see a loving couple that can't reproduce on their own and want to help.
Anyway, are you similarly against giving children up for adoption?
What about the people doing the adopting?
I would make an exception if she were willing to truly fulfil the role of a mother for the child in some sort of weird 3-parent relationship, but my understanding is that this doesn’t usually happen.
As an aside, the legal process is quite explicit that the surrogate has no rights to a relationship with the child. And again, it's controversial. People don't necessarily volunteer to strangers they they have done this.
Yeah I dunno man. After UK NHS told my (then) wife it would be a year wait to see a psychiatrist (to continue her mood meds she was on in the US), we found some Covent Garden psychiatrist that charged £300 for her once monthly sessions and it was all no problem and reimbursed at 100%
I don't think it's normal for a health insurance plan to let you just see any specialist with zero triaging and cover 100%
I really have no idea what their limits were. I went to a private GP to get a health certificate to run the Paris marathon, and it cost £100, and that was reimbursed. A years worth of sports physical therapy was also reimbursed 100%
I could only conclude the cohort was so healthy that even our splurging didn't matter.
So... what's the Elon deficit reduction strategy here? Get like 5-10% of the government to resign, maybe fire another 5-10%, then go to congress and say look we can spend 10-20% less on salaries, go ahead and pass a reduced budget through reconciliation?
This actually sounds not too crazy.
Hopefully losing 10-20% of the work force doesn't cause a corresponding 10-20% reduction in government revenues but... it's kind of hard to see how it would.
EDIT: maybe something's wrong with me but I consider this topic fun and not weighty which is why I posted it here in the Friday thread. Seems like it has created a more typical CW thread discussion. My bad.
dr_analog
top 1% of underdog fetishists
5mo ago
I notice, as of a few hours ago, a new provider called "DeepInfra" appears with similar rates to DeepSeek. Despite the name they don't appear related to DeepSeek.
dr_analog
top 1% of underdog fetishists
5mo ago
Deepseek as a provider is by far the cheapest and fastest with modest but totally usable context length and output limits. The Americans serving this (with potentially superior GPUs) are completely shitting the bed, half their responses just stall in the chain of reasoning and don't get anywhere, despite them being 10x more expensive. They clearly have no idea how to run this model, which is reasonable since it's deepseek's baby. But Americans can all run American models just fine at the exact same price. Claude on google or amazon costs exactly the same. I think in addition to the advantage of knowing how to use their model they have some secret insight into how to use compute efficiently.
Hi again. Any more insight on this? Perhaps it's because they optimized it for Huawei chips or something and everyone else is trying to make it run on Nvidia?
Context: https://x.com/olalatech1/status/1883983102953021487
dr_analog
top 1% of underdog fetishists
5mo ago
Deepseek as a provider is by far the cheapest and fastest with modest but totally usable context length and output limits. The Americans serving this (with potentially superior GPUs) are completely shitting the bed, half their responses just stall in the chain of reasoning and don't get anywhere, despite them being 10x more expensive. They clearly have no idea how to run this model, which is reasonable since it's deepseek's baby.
Are we comparing apples to apples? I understand the models with Llama and Qwen in their names are distills from their native model for compatibility with plugging into existing frameworks, though they might perform like crap.
Whereas I understand the native DeepSeek r1 is a mixture of experts thing that selects a dynamic 37b parameters out of the overall 671b.
dr_analog
top 1% of underdog fetishists
5mo ago
I don't think following a GPT-from-scratch lecture is going to get you there
I wasn't claiming that. Just trying to support the claim that they were more open in the past. I doubt any novel AI technique discovered in the future will even have that.
I'm not convinced that they have any left to make.
Counting out the most absurdly well resourced AI lab with a history of breakthrough success seems fairly bold.
What country map are you talking about? One can't average the whole of Italy due to it effectively being two countries stitched together. The differences are very big.
Yes. That's why I was asking. The map in the post OP links to just treats Italy as a whole unit.
Lots of studies show Sicily having an average IQ of about 90 with northern Italy having slightly above 100, with the GDP per capita difference being over 100%
Okay. But that's explained by upwardly mobile Sicilians leaving after WW2.
I'm not even disputing it necessarily. Cousin marriage is high in that culture, for example.
It would be interesting to see how diaspora Sicilians test.
Have you forgotten what they taught you, though? I find with a lot of non-fiction I lose episodic memory of the book but I retain semantic memory of much of its theses
Let's take Guns, Germs and Steel for example. It's a scholarly tome, as far as books go, and I read every single page of it. But I don't think I was better off for it than reading the Wikipedia article about it.
If the book was an ultimately awesome information resource, it would present the thesis and allow me to interrogate it further in whatever parts I needed more detail or evidence about. Instead it just went on in exhaustive detail from start to finish. This wish isn't a novelty, I spent hours upon hours reading it and it felt completely unnecessary!
(And despite the monumental effort he put into writing it and justifying it, his thesis is probably wrong)
dr_analog
top 1% of underdog fetishists
5mo ago
Hmmm, fair points. Need more info. OP did any of the trained people get raises?
I run 6 days a week, lift weights 3 days a week and intermittent fast from 6pm to 10am. I don't eat "processed" foods either. I've still been gaining weight and DEXA scans confirm that it's actually fat that's being gained.
I've been struggling with obesity since I was a teenager[1]. I had a foothold on it in my 30s in that I was simply overweight and not obese but now it feels out of my control completely. Until...
Personally, I think I just enjoy food more than other people. God I love food. I love eating. It is basically the only thing I really enjoy in life. When I bite into a home made migas taco, the melted cheese, the crunch of the fried tortilla strips*, the creamy avocado, it makes my whole brain light up. I am salivating just writing this.
Have you considered... Semaglutide!? I went on it recently and the effect is pretty interesting. Mostly it's a lot more psychological than "physical", for me, so far. I can feel hungry and be in the kitchen, and feel like snacking, but all of the snacks seem like too much work to get out and eat. So I don't.
Cue meme where person with ADHD does Adderall and they're bewildered at their insane focus and energy and cry that this must be how normal people feel all of the time. But non-ironically.
- And no it did not magically get fucking better living in Europe. The food in America is not uniquely poisonous.
You are very biased in favor of defending the system.
The irony in facing accusations like this is that I irrationally refuse to sign up for any company like UH or Cigna or whatever because I find them too triggering to work with and instead use some low-cost possibly-a-scam health ministry and just pay out of pocket the rest of the time.
I'm also a rare person who has experienced health care in multiple US states, including extreme Cadillac insurance and also Medicaid, and also "socialized medicine" in countries like the UK and Italy. At the end of this I'm generally burned out and annoyed by simplistic rationalizations and explanations.
The system sucks. Everywhere. In the sense that it's run by humans and have to deal with impossible demands and mis-aligned incentives. I don't think simple-minded analyses move the needle in a helpful direction. In fact they're usually wrong and sometimes just get people killed!
Ah yes more censorship of problems and calling everyone bad names and conspiracy theorists
I was not calling for censorship.
dr_analog
top 1% of underdog fetishists
6mo ago
It doesn't make financial sense, but they just do it anyway because anyone with actual business acumen wants to shoot themselves dealing with medical horseshit and avoids it.
And if they do have business acume, someone shoots them! Heyoo!
dr_analog
top 1% of underdog fetishists
6mo ago
This is true, and I'd be interested to see how claim denial rates line up with a given FY cycle. They could be just vastly incompetent, making all of their customers hate them for no reason by being unable to predict claim demand, even with the vast swaths of data they have.
I am very curious about this as well.
Though comparing claims denial rates between insurance companies isn't useful without more context? It's true Kaiser has a denial rate of 7%, but aren't they famously (though not exclusively) an HMO? 7% seems low, if you ignore the fact that (pulling this out of my ass) 99% of medical providers are not allowed.
dr_analog
top 1% of underdog fetishists
6mo ago
Thanks for the reply!
In general, reading about this is fascinating to me. It sounds like an arms race and like providers can get an edge if they have research and analytics firms (or departments) staying on top of this stuff and helping them route through each company's bureaucracy.
Another common thing that happens is that insurance companies will randomly deny things. If I bother to schedule an appeal they will usaully decide to cover, but they know we are busy so if they randomly deny a good number of things will be dropped. Especially cheap drugs - sometimes it's easier to send the patient to Walmart and cash pay than fight the insurance company. I have a limited amount of time. They abuse this. When they do decide to fight your "peer to peer" review is generally with someone in another specialty who retired 40 years ago and has no idea what the actual standard of care is.
This sounds like a class action lawsuit waiting to happen so I'm surprised they do it, but maybe I'm naive about the wheels of justice.
Now everybody does this stuff but somehow United is appreciably worse.
Are they better or worse than Medicaid?
Few providers in my area take Medicaid, and the ones that do have very long waits to see. I understand it's because they have pitifully low reimbursements but also have high claims denial rates.
dr_analog
top 1% of underdog fetishists
6mo ago
Permutation City
I really loved this book. Aside from being generally interesting it felt like the guy predicted AWS in 1994, at the least.
Thanks, good info. I'm going to try ordering some things and see how they go.
Oh jeez I was just planning to do the contours of our cargo bike with strips but after all, why not? Why shouldn't I paper every square inch of the box with these sheets?
More options
Context Copy link