akavi

Has anyone gone from not "getting" the value of algebraic effects to getting it? And if so, what was the tutorial/process by which you achieved enlightenment?

It's only not tiny in terms of geographic area.

Population wise and urban area wise, it's miniscule.

How’d GPT-3/3.5-turbo do?

The difference is unlike self-driving and crypto, LLMs are providing value to people _today_.

In my personal life, GPT4 is a patient interlocutor to ask about nerdy topics that are annoying to google (eg, yesterday I asked it "What's the homologue of the caudofemoralis in mammals?", and a long convo about the subtleties of when it is and isn't ok to use "gè" as the generic classifier in Mandarin.)

Professionally, it's great for things like "How do I recursively do a search and replace `import "foo" from "bar"` to `import "baz" from "buzz"`, or "Pull out the names of functions defined in this chunk of scala code". This is without tighter integrations like Copilot or the ones linked to above.

I've got to wonder, how does a second player in the LLM space even get on the board?

Like, this feels a lot like when the iPhone jumped out to grab the lion share of mobile. But the switching costs was much smaller (end users could just go out and buy an Android phone), and network effects much weaker (synergy with iTunes and the famous blue bubbles... and that's about it). Here it feels like a lot of the value is embedded in the business relationships OpenAI's building up, which seem _much_ more difficult to dislodge, even if others catch up from a capabilities perspective.

Both "smart people have smart children" and 90% of this forum being smarter than their parents could be true if "being on this forum" was selected on intelligence.

I don't have the time to work through the math concretely, but intuitively, imagine that the intelligence of parents is normally distributed, and the intelligence of their children is a tighter normal distribution around the parents' intelligence. Set the cutoff for "participating on HN" to to some threshold. Given the right set of parameters, most of the "weight" in the aggregate child distribution above the threshold will come from the long tails of the child distributions of parents in the fat part of the parent distribution.

Sorry, that was rude of me. I know it's a ton of work, and am endlessly appreciative of the work you (I assume?) and others have been doing to push this forward.

Now even Java has pattern matching, what's Javascript's excuse?

Literally the only feature preventing Typescript from being my perfect language.

I don't know if you need to invoke ulterior motives to explain startups jumping on the biggest change to the tech landscape since the iPhone.

The danger is that the opportunity academia is giving you is something more like "you’ll be given the chance to try to understand what they were doing 5 years ago".

48% discount from peak valuation (95 G$ in Mar 2021). Ouch.

I disagree that Spanish's /o̞/ is closer to General American's /ɔ/ than to its /ou/.

But that's fundamentally a question of perception that people are going to disagree about

My reading is that the "assessment" (tax) will be borne by all banks, not just the bank that fails.

My gut says that the incidence will fall primarily on deposit holders (likely in the form of marginally lower interest rates), and not significantly on bank equity holders, but I suspect it'd take an econ phd to fully parse that out.

Because if there was a run on JP Morgan, the US government would absolutely "bail out" depositors.

Consolidating increases the risk of a bank run, decreases the risk to the individual depositor.

To save people the math, that's 260 k$ in cash compensation (or ~mid FAANG L6 tier).

I am confident in saying there were not many making more than that.

You can get loans backed by the shares themselves. Ie, the lender gets x% of the shares, rather than being paid back a specific dollar amount at a later time.

You're right, it's not necessarily true relative to past levels of consumption. But there's a limit to economic growth, and the choice of "paygo" vs "investment" probably only has marginal effects on the rate of economic growth, so the point stands.

"Non-producing" people are retirees (as well as children). People who don't produce market goods/services that make up the vast bulk of our material needs.

PAYGO vs investment earnings is purely a financial construct.

The central issue is that physical goods and services need to be produced around the time that they're consumed, and at a given point in time, an economy has a fixed capacity to generate these goods and services.

If you have a non-producing population, and a producing population, then the only way for the non-producing population to consume a given quantity of goods and services is for the producing population to not consume those goods and services. Whether you call the reduction in consumption ability of the working population "taxes" or "investments" doesn't change this fundamental fact, except insofar as it alters certain behavioral incentives around deferring consumption and legal frameworks around property rights.

(This isn't 100% of the story, since you can "lay claim" to production of working people overseas via international investment, but is accurate in broad strokes.)

Sounds like most cholinergics then?

How does it differ from racetams/uridine/huperzine in your experience, if you've tried any of those?

Wikipedia suggests that "ευ" was pronounced /eu̯/ in ancient greek[0]. My suspicion is that "ευ ζην" is an Ancient Greek "fixed phrase" which is pronounced with Modern Greek phonology.

[0]https://en.wikipedia.org/wiki/Ancient_Greek_phonology#Diphth...

Where are you getting an /f/ sound from?

Edit: Ah, it seems the digraph "ευ" is pronounced /ef/ in Modern Greek, despite (by convention) being transliterated "eu".

Naively, the category seems to be "high growth tech". So that includes both the SV megacorps and startups (which are also concentrated in the Bay Area).

Are companies not allowed to stop paying wages to striking employees? Or is it that tuition remission is viewed as distinct from wages legally?

Is Dan Price associated with effective altruism?

Dustin Moskovitz's Good Ventures also seems to focus on more "prosaic" issues like medical research/direct cash transfers/animal welfare.

I'm not sure I'd expect people with a real need for a medication to address executive dysfunction to be the set of people who persist in acquiring something when it becomes difficult to do.

Freight forwarders merely subcontract to the underlying carriers, who, theoretically, are pretty commoditized. In this model, the differentiating costs for a freight forwarder is in the labor required to organize the various subcontracts, which is what Flexport attempts to drive down via tech.

That said, I can't really vouch for the soundness of this argument one way or another. It's been a long time since I worked there, and my primary interest when I did was more in understanding the industry at a conceptual level than in litigating the viability of the business model.

G => SI prefix for 10^9

YC-funded freight forwarding company. Basic thesis is by using tech, you can dramatically lower the labor costs of forwarding, and since freight forwarding has huge returns to scale, the benefits compound. They raised 2 G$ from 2019-2022, and their CEO used that warchest to expand counter-cyclically through the pandemic. TBD if either of the two theses pan out.

Freight forwarding itself is the industry of coordinating the logistics of freight shipment, ie subcontracting with physical movers of goods (ocean carriers/railroads/trucking companies) and dealing with customs/legal requirements involved in the movement of goods

(Disclaimer: former FP employee and current shareholder)

Depends on the reason you need to request gender, but for in-app pronoun purposes, I'd do:

  [ ] male
  [ ] female
  [ ] non-binary
  [text box for other]

And use "they" pronouns for the last two options.

(Ideally you'd allow the user to specify the pronouns themselves, but even among people who use non-conventional pronouns, getting them to correctly enter the subject/object/possessive cases of their choice is heavily hit or miss)