millimeterman

    Start with Praxeology (the methodology of Austrian economics).

"Start with a fringe ideology that explicitly rejects empirical evidence" is not exactly a good start.

    The simple axiom that all preferences are subjective instantly eliminates a lot of the foundation of micro (and macro) economics as it is typically taught today.

No? If anything, econ 101 classes are basically Austrian economics classes due to how many idealized simplifying assumptions are made. And subjective preference curves are the entire basis of the supply/demand graphs that econ 101 students are bombarded with.

I half agree. While I respect Pope's artistic vision and the work that went into the aesthetic, I personally found it headache-inducing to actually look at. Especially with how much of the game is about gleaning clues from the environment, I simply could not enjoy it and dropped the game pretty quickly.

Void definitely exists in ML.

  —- Haskell
  data Void

  (* Standard ML *)
  datatype void = Void of void

The government is using tax money to fund the creation of consumer products, while consumers vote with their wallets that they don't want those products. Doesn't sound very democratic to me.

That kind of "stability" is more accurately described as stagnation. It may protect workers in a narrow sense, but it also disincentivizes the kind of productive risk-taking that leads to business creation, innovation and overall societal increases in wealth. This type of worker protection scheme is what people tend to reach for because it's the first thing they think of and it sounds sensible, but it can really drag on society in the long-term.

A much more effective approach would be to make loss of employment not be a big deal, e.g. through stronger social safety nets. That way you ensure people's lives are financially stable but without introducing friction to efficient allocation of resources. Ideally, losing your job would be no more than a mild inconvenience - like having to buy a different brand of milk because your usual is out of stock. That ideal isn't realistically achievable, but we can at least try to get closer.

I may have misunderstood your earlier comment as a claim that minimum size laws are beneficial and should be maintained.

If I understand correctly now, your argument is that minimum size laws are not currently a bottleneck for building denser apartments? If so that’s a valid point wrt what changes should be prioritized, but I think it’s fine to point out that minimum sizes also need to go.

Minimum unit sizes necessarily increase the price of housing and lower supply - there's no free lunch. Small apartments suck, sure, but you have to ask yourself whether it's a good idea to ban them and thus force everyone to pay more. Personally, I don't think small apartments have enough negative externalities to justify such regulation.

That strikes me as a fairly modern thing. I imagine that weevils and/or rocks in flour was a common issue just a few generations ago (or, in many countries, to this day).

In the very long term, yes a large plot of land in a densely-constructed city would be extremely valuable. In the short and medium-term though, the downward price pressure of increased supply outweighs the upward price pressure of amenity effects.[1][2][3]

Very few NIMBYs will openly admit to wanting to maintain their property values because it's a much less noble sounding cause than "preserving the character of the neighborhood" or whatever.

1. https://research.upjohn.org/cgi/viewcontent.cgi?article=1334...

2. https://blocksandlots.com/wp-content/uploads/2020/02/Do-New-...

3. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3867764

> We have had local regulations and NIMBY groups forestall new construction/denser housing, but the state pushed through legislation to forcibly override those local blockers. Incentives should now be aligned for builders to build, but local homeowners will continue to fight against it because their incentives are not aligned.

California has only made a few tiny steps toward allowing construction. They are good steps, don't get me wrong, but there is still a very long way to go before California can even begin to catch up on its decades of under-building.

California's permission of converting single-family zoned land into multifamily housing is in name only. It can only be done by owner occupiers - not by developers who want to buy land and build on it. Which basically never happens.

The strong scientific consensus is that saturated fats are worse for you than unsaturated fats.

The idea that saturated fats are better than unsaturated fats is a small minority opinion. It's all the rage with wellness influencer types, but it is currently supported by far less evidence than the consensus.

I suspect the community will start creating lower precision/quantized versions of the model very quickly. LLaMa 30b quantized to 4 bits is runnable on a 3090/4090.

> I'm happy to hear about your self control.

Actually, my self control is really quite bad. I'm terrible at doing difficult or unpleasant things and am very prone to low-effort, self-destructive hedonism. But as a result I have a very intense fear of addiction because I know I would struggle to pull myself back from it - I've never drank alcohol or done any recreational drugs for that exact reason. I was very hesitant about pursuing potentially lifelong medication, I requested the smallest possible starting dose from my doctor, and I don't intend to ever increase it beyond what I currently take (10mg/day).

On a philosophical level I still don't _like_ that I'm taking medication, but the benefits are significant enough that I tolerate it. And perhaps ironically, ADHD medicine actually helps me push past my lazy hedonism and do things that are necessary but dull.

(I would note that it doesn't make me _want_ to do boring things, nor do I think it's supposed to if dosed correctly. Even on medication, I'd still rather play video games than do work. But it at least makes it possible for me to sit down and work anyway. Whereas before, every day was an exercise in staring futilely at the screen while begging my brain to do literally anything.)

> I wish I were effective as a person naturally, I'm sure you can relate.

Very much so. I really wish my productivity wasn't dependent on medicine that

a) I need to remember to even take

b) Has annoying negative side effects

c) Is a controlled substance, with all the hassle that brings

But unfortunately, I guess that wasn't meant to be.

> I've encountered several people here claiming that stimulants don't get you high and that they aren't addictive.

Just because stimulants _can_ get you high and be addictive doesn't mean they will when taken appropriately. The only time I felt any kind of high from my ADHD medicine was the first day I took it. And I quite easily skip taking it on weekends, with the only side effect being that I browse the internet all day and struggle to do anything productive.

To be honest, I don't actually like being on stimulants. Methylphenidate made me feel like I was constantly in low-grade fight-or-flight mode and dextroamphetamine (which I'm on now) gives me mild emotional blunting. Plus both cause slight dry mouth, appetite suppression that I have to forcibly eat through, and insomnia if I take them too late in the day. Not to mention the fundamental hassle of taking pills every day within narrow bands of time, something I regularly procrastinate and/or forget. (Also, I have tried taking higher/double doses as you suggested earlier when I was figuring out my optimal dose. They just made me feel jittery and on edge - not exactly pleasant.)

If I had no ambitions or responsibilities, I would be more than happy to throw my medicine away and sit around on my computer all day every day. Unfortunately, I do have ambitions and responsibilities and medication helps mitigate my crippling, career-threatening inability to get things done. So I'll take that tradeoff.

Stimulant addiction is certainly real, but addiction recovery groups are obviously going to self-select for people that have/had a problem. The reality is that there are millions of people who take ADHD medicine at therapeutic doses (which should _not_ induce a high, FWIW) for years without developing an addiction or needing dose increases.

Knowing the potential risks of stimulants is good, but unduly scaring people or over-restricting access for the majority who use them properly is not helpful.

Talk is cheap - actions are what reveal true preferences. The fact that Teams is so widely used despite it being slow garbage indicates that people actually don't care (enough to overcome whatever benefits they see in Teams).

My uninformed speculation is that this might be an artifact of the embedding layer (e.g. CLIP), not the image training data.

Presumably, the dataset for training the embedding layer (which is trained separately and then fixed) is not stripped of copyrighted content. So it will have learned the "Pikachu" is related to words such as "yellow" and "rat".

Therefore, even if the image dataset didn't have a single picture of Pikachu, the image generator will still likely produce a yellow rat. Just one that doesn't actually resemble Pikachu.

In my experience ADHD medicine doesn't actually induce weight loss, despite weight loss being a desirable thing for me.

It suppresses your appetite, sure, but it doesn't shield you from the effects of not eating all day. I often force myself to eat something for lunch or I'll just end up tired and annoyed in the afternoon/evening (edit: I don't eat breakfast).

Plus, once the medicine wears off at night (or on weekends when I skip taking it), I usually get ravenously hungry and easily make up any caloric deficit.

Sorry, I imagine you posted this before my edit. I don't think this marketing video contradicts my point at all. For Microsoft to commit a large amount of dev time to rewriting Teams and improving its performance, there must have been an actual business reason. That's not something you do because some nerds on HN think your app is too slow.

Presumably, Teams was so unbelievably slow that it crossed over to being an actual problem for customers and not just a mild nuisance. So Microsoft improved performance just enough to stop affecting sales. Is the performance _good_? No, it's still comically slow. But there's simply no incentive to improve the performance further.

I know of mid-size organizations where everyone up to the very top complains at length about Teams being slow and buggy. But they still use it, because it's cheap and doesn't actively hinder work getting done. I don't think it's simply a matter of bean counters being out of touch.

I think the boring but true answer is simply that no one cares.

Users don't care because Teams still basically functions and the performance problems are at most a minor inconvenience. As a result, organizations don't care because Teams is cheap and does everything they need it to. As a result, Microsoft doesn't care because dedicating dev resources to something which neither saves them money nor attracts more customers is a waste.

Microsoft employs many smart people and I'm sure Teams' performance could be improved by orders of magnitude if they wanted to. But why would they want to?

(And to reinforce my point, when the performance _did_ get so bad that people apparently actually started caring, Microsoft not only improved it but made a video advertising it. It's just that there's no incentive to optimize any further than is necessary.)

I wonder what "economically conservative" means here. Usually I would expect someone economically conservative to be against government intervention and prefer market-oriented solutions to problems, but the people you're talking about tend to want the opposite.

It's not, but I didn't say it was. Plus the reasons that "the left" opposes housing are generally quite different than the reasons "the right" opposes housing.

Directly? No. But indirectly? Yes. The rampancy of NIMBY, PHIMBY, and "anti-gentrification" attitudes leads to decreased construction, increased housing prices, and the enrichment of (existing) landlords at the expense of tenants.

By what mechanism would (2) make housing cheaper?

Not when US immigration policy is so restrictive and hostile. We are actively preventing our own prosperity in the pursuit of protectionism.

It's worth noting that while Houston does not have zoning in the strict sense (e.g. assigning land to be commercial only or single family home only), it does still have quite a few harmful restrictions on land use such as parking minimums and minimum lot sizes. These regulations enforce low-density construction and car-dependence, even in the absence of exclusive zoning.

As a result, while Houston has succeeded in keeping prices under control, it should be not looked at as an example of housing and land use perfection. It's also critically important to not look at Houston and get the impression that lax zoning creates suburban sprawl (something which California really doesn't need more of). Lax land-use regulation is, in fact, an excellent tool to decrease sprawl and increase density.

I'm disappointed they didn't push parameter count higher, but I suppose they want to maintain the ability to run on older/lower end consumer GPUs. Unfortunately it severely limits how high-quality the output can be.

What exactly do you think is degrowth? Allowing construction of dense, walkable cities would likely dramatically _increase_ economic growth in the US.