ericd
ericd on Hacker News
And in some, like ours, it takes 5.
Yeah, a lot of HVAC guys don’t install anything but the big chonky outdated single/dual stage domestic brands. I read this comment as he got one built in with the house, and unfortunately, since they tend not to be understood by buyers and therefore command any premium in the sale process, there’s not much of an incentive to install anything but the crappiest pump on the market. This happened to us, the last homeowners wanted to replace their old furnace right before selling, so they installed the lowest efficiency heat pump that’s legal to sell. It’s trash in cold weather. I DIY installed a cold weather rated mini split myself later, it’s incredible in cold weather, and sips power.
I went down a big rabbit hole when I installed it a couple years ago and made a little calculator to compare heat pumps based on the weather in your area and the pump temperature-efficiency curves: https://heatpushers.com
It needs some revisiting, though, I’m guessing half the links are broken by now.
There’s a challenge in that different heat pumps have wildly different performance characteristics, and they’re not widely understood by people. Some can’t heat effectively below 47 degrees F, others happily keep pumping out lots at -13 degrees F.
> > EVs are 4 - 5X efficient
> Not even close
Cybercab is supposedly >200 mpge.
> How are you planning on powering all your electric appliances? And before you say solar, you are aware that the sun only shines during daytime, while peak energy usage happens in the morning and late afternoon.
You’re aware that levelized power source price comparisons quote “solar+storage”, not just solar, right? And that that price is now cheaper per MWh than combined cycle Nat gas?
Battery prices and solar prices have come down so rapidly that anyone operating on knowledge from even a few years ago are going to have a very inaccurate understanding of the landscape.
Another bad one I've heard, buying thinly traded penny stocks as a way to exfiltrate money from brokerage accounts without wiring.
Anyone have any thoughts on what happens when banks/brokerages start getting hit in earnest? Think it's prudent to split single brokerage portfolios across multiple? Or do they have enough speed bumps in their procedures that they're not quite so vulnerable to major losses?
The specifics are tough, but the rough shape isn't impossible. But it's hard, which is part of why most people try not to do it publicly with much confidence, lest they look like they're not worth listening to on the topic. Like, why are we listening to his opinions on this when he so clearly isn't an expert, and so many of his past opinions on it have been so far off?
And I think a big part of why you shouldn't listen to him is that he's said repeatedly that he doesn't really use generative AI outside of occasional experiments - he hasn't seen its potential enough to bother using it himself beyond some initial tests. Which is fine, but I think it also means that he likely doesn't have a good intuition of the likely future demand, which is pretty key in figuring out whether debt loads are likely to be sustainable or not. Investing is fundamentally about trying to predict the future and putting your money into the things that will likely be successful there. And he's demonstrably been terrible at predicting the future of this stuff.
Go read "Some Zitron predictions" here:
https://danluu.com/zitron/His track record of prediction on this topic is stunningly bad, and he's made a job out of being counter to the AI hype, so he seems likely unable to honestly update his beliefs to adapt to changing evidence without jeopardizing the attention people pay to him.
Except the GPUs last a lot longer than 3 years - A100s are 6 years old at this point and still in heavy use.
Oh yeah, for sure, they're nice. Just mean that it's also just a real utilitarian benefit for those without technical aptitude.
Yep, for someone with ~zero technical aptitude, I'd happily pay the premium for decreasing the number of things that can go wrong. Things getting unplugged accidentally account for a lot of tech support.
I'd say Snow Leopard was peak.
I think the only real solution to this is to switch to an OS with a backing org that has optimization for power users as a first-class goal. I've been liking Omarchy and NixOS for that. I say this as a Mac user since 1989.
It's a nice thought, but the strategic implications make it very hairy, and international trust isn't going in the right direction for that. But within large countries like the US with very varied climates, it seems like we really should.
Maybe you're thinking of B Corp? That's just a certification from some group. Public Benefit Corporation, on the other hand, changes directors' fiduciary responsibilities. For example, earlier this year, the Delaware Court of Chancery ruled that PBC directors are not obligated to maximize shareholder value when selling control of their company.
I highly recommend Eric Ries' new book Incorruptible if you want to learn more about this stuff. In there he tells the story of the Vectura Group, a pharma company that was developing treatments for asthma and other respiratory diseases. It was looking at a sale, and it had a couple offers, the one from Phillip Morris slightly edged out the other, and because of fiduciary duty, even though many involved viscerally hated the idea of going to Phillip Morris, as the cause of many of the ailments they sought to cure, they sold to them.
Yeah, a copy of office I bought bricking myself, even while constantly keeping office autoupdate running was my last straw with them. Fortunately, LibreOffice seems to be mostly good enough now.
Good thing LLMs are really good at translation tasks?
Right, I didn't get into this to make money, I got into this to make things. This stuff is amazing for that.
After it bulldozes the problem, can you not then use the LLM to help you understand the solution, bit by bit?
>The 0.004% richest Amreicans hold the majority of US wealth
Nah, it’s more like the top 5% hold half. The top 0.1% are at ~13%.
Also fyi, ours run it on their Raspi 400 via dosbox, they love it too. The DOS games library is full of educational-ish gems (they really like Lemmings and The Settlers, too).
Hm so you're saying all the SaaS runners should put "Lifetime, paid monthly" in small print under their price? Bam, goodbye churn.
In nvfp4, it's about 300 gigs once you offload n-grams, 491 without offloading, you can run it pretty well on 4x DGX Sparks, which last I checked was about $20k. So, it's definitely runnable.
Or you can just use any of the neoclouds' shared hosting. The thing for them to be freaked out is that these models are getting good enough very quickly, and all the shared hosting providers can run them for a tiny fraction of what the frontier model companies charge.
I think those proposals are usually scoped to science/tech PhDs, I guess to avoid that scenario, but I only have a headline-level knowledge of the bills. I don't know if it's that insane at most labs? But my first/second hand knowledge is a decade out of date at this point.
What does it mean to "break a lot of PhD programs"? Research is competitive, and hard work. Yeah, there are easier options, but in the long run, the countries with the people willing to work their butts off academically are going to outperform ours if we can't get their best to come here and work their butts off here.
>Clipping a green card to every PhD is probably political non starter. As American, I'd be very leery of such a scheme.
There is actually bipartisan support for trying to make it easier for PhDs to stay - both parties have repeatedly introduced legislation to help with that - but it seems to keep getting derailed by arguments about other aspects of immigration reform.
To follow up on this, US PhD programs are not generally packed with US citizens, they're frequently the cream skimmed from the rest of the world. For many subspecialties, there's just not going to be enough US citizens to do all the cutting edge work that could/should be done.
We should clip a green card to every PhD we grant.
You could theoretically run these (slowly) if they were open weight. ~$10k of DDR4 is enough to hold them. The data itself costs ~0 to replicate.
This seems to imply that training will ever be done? But yeah, I think the idea is that the appetite for thinking-on-tap will be enormous.
Even for smaller models, I think they’ve found that training an enormous, inefficient model and then distilling it internally to something much more efficient to serve is the way to go.