ahzhou

It’s for buying political capital, which is as important as the underlying tech. Look at how the US gave away its solar industry to China from of lack of political will.

This is an incredible false equivalence. Your comment here would have been censored in China.

Interesting how Amazon uses “Automated Discounts” to get around “Usual and Customary” prices. Typically, pharmacies only get reimbursed by PBMs (middlemen between insurers, manufacturers, and pharmacies) up to a maximum of of their U&C price so their incentive is to set it really high in order to maximize reimbursement rates.

Amazon’s GoodRX-style discount cards gets around this by basically charging a spread over the reimbursement and kicking back some portion of the spread to the PBMs. By making the price an “Automated Discount” to some number between the U&C price and the reimbursement price, Amazon can charge less without impacting their reimbursement rates while paying the PBMs off.

This is a huge simplification of the US’s broken pharmaceutical system, obviously. One example of the nuances here is that Amazon is probably not able to set the price of the discounted drug.

Some reading: https://www.drugchannels.net/2020/08/how-goodrx-profits-from...

Types are just human and machine readable documentation - orthogonal to debugging IMO.

Great typing, like great documentation or great design, helps with debugging - but that’s not the main goal.

If you’re building B2B software, It’s not that you can’t track but rather that you need to spend time building data retention infrastructure before customers would even consider trialing your solution.

FWIW, I think the discussion here this highlights the biggest problem with the approach. Either it's confirmatory and wasn't a question worth asking in the first place or it's surprising and people need real data to evaluate the response.

I’ve known both. Some kids have a huge chip on their shoulder to carry the family legacy or exceed their parents, while some were happy to coast on their trust fund.

I think the biggest difference is that kids without family money don’t have the option to coast.

If the goal is to pick “future leaders”, Yale succeeded as far as admissions are concerned. I think it’s unreasonable to expect that an institution can predict what decisions a teenager will make in their 60’s.

While true, standardized testing is still more fair than qualitatively judging extracurriculars or applying arbitrary GPA normalization across high schools.

1. Connections talk, but money talks even louder. The companies they buy are also trying to sell for top dollar and rarely court just one buyer.

2. CLOs are definitely problematic, but the current rate of default is vastly overstated. If private credit defaults at the rate people seem to think it does, we'd be looking at a GFC-type situation. I can't say if there's systematic risk in CLOs a la MBS, but any possible risk hasn't materialized yet. The biggest PE firms today rarely default. Blow-ups such as Toys-R-Us are newsworthy for a reason.

See my comment below for more information. I'm not in PE, but I am very familiar with how the sausage is made.

Pensions are very rarely lenders. They may invest in lenders but tend to look at historical performance.

Political appointees are not the ones making investment decisions. At most they are setting very broad themes. For example, political appointees are the ones that say all investments must be carbon-neutral.

> Why is private equity ending up with all these resources? Who is selling to them and why? Why didn't this happen before? It's not like PE is new.

PE as an investment strategy was pioneered by KKR in the 60's, beginning with the acquisition of family-owned businesses facing succession issues. The strategy exists because there's really poor liquidity for private companies. If you're a founder who wants to sell your stake, you either:

    1. Sell to a strategic buyer

    2. Sell to an individual

    3. IPO

    4. Sell to PE

The first three options may not be always viable at a given point in time (aka illiquidity). Strategic buyers aren't always looking to buy, an individual may not have enough capital, and IPOs place a huge amount of reporting burden on a company. PE is the only reasonable exit option for a huge swath of private companies. Take a hypothetical family-owned supermarket chain with 10 locations across 3 cities in the midwest. All the kids are terrible successors. Who's going to buy it?

    1. A strategic buyer, such as Walmart, may not be interested right now.

    2. An individual probably doesn't have enough enough money to buy the whole thing

    3. An IPO saddles the business with reporting requirements that simply don't make sense for a supermarket with only 10 locations

People sell to PE because it's often the only way to exit. Almost all bootstrapped companies belong in this category. At the end of the day, PE funds are simply groups of professional investors that specialize in buying out companies. If they are poor investors or poor operators, they will tank the company. If they are good investors and good operators, they will sell the firm for a profit in the future.

--------

> When PE loads up a firm with supposedly unsustainable billions of debt, someone is on the other side of that transaction, lending the billions. Who does that and why? Are they perpetual suckers, unaware of the decades of experience we have doing this?

Many people offer debt, from private credit funds such as Golub Capital to bank syndicates. The best lenders are not suckers and the best PE funds almost never default. Oftentimes, the PE fund itself is collateral. PE funds that regularly default will have a difficult time finding a lender.

People also misunderstand the purpose of debt. It's simple math:

You want to buy a company for $100M. You can buy the entire thing with cash and sell it again after it doubles in 5 years, netting yourself $100M, giving you a 1x return.

Alternatively, you can pay $20M down and borrow $80M at a 10% interest rate (simple interest for ease of calculation). When you sell it for $200M in 5 years, you net $200M - $20M - $80M * 1.5 = $60M, giving you a 3x return.

Obviously, you've also increased the risk - your business must be capable of paying off an extra $8M a year in interest.

Your insinuation that the UK will implement a Ukraine war draft in order to advance a totalitarian agenda is weird, unnecessarily conspiratorial, and dilutes your point. I encourage you to make one point and to make it well.

Thats right. At the end of the day, PE investors need to make their returns and blow-ups like Toys-R-Us did not make good returns. That said, there are systematic problems with the incentives involved.

1. PE investors tend to be VERY financially savvy, but sell to less skilled investors. If they see that they have the chance to sell one of their assets at a great price, they don't have any issue with anyone holding the bag. There were at least a few IPOs/SPACs that left the (relatively less-savvy) public holding the bag.

2. PE investors tend to be be finance-minded, not operations-focused. That means that their planned optimizations think about the financial health of a company, not the "real" health. Culture can suffer because of this, for example.

3. PE is very interest rate dependent, and low interest rates / bad investors can (and probably did) make the tech bubble worse.

It’s basically the acquisition (buying out) a company using debt-financing (leverage). The typical plan is usually to use business revenue to finance the interest payments, optimize the business via cost-cuts or roll-ups, and flip it for a profit in 5-7 years. Like most things, PE can be helpful or destructive depending on the execution and the exact strategy for flipping.

At its best, it’s bringing in experienced operators and maturing a company into something that is stable, before selling it to an acquirer or IPO.

At its worst, it’s saddling a weak business with debt, hiring terrible execs, and making unsustainable cost cuts to stave off an implosion.

The strategy matters a lot. Some funds specialize in “distressed assets”, for example and are very good at carving up dying company and selling it for parts.

It’s incredibly hard (read: impossible) to make top-down dogfood pushes across a massive company like Meta. You have three options:

1) Create an team to internally sell it

2) Make it an ultimatum, and fire people if they don’t agree

3) Hope your marketing is good enough that you get grassroots adoption

Option 1 is a waste of money - your best sales folks should be making money, not doing internal processes. Option 2 is terrible for moral and retention. Option 3 is ineffective. I’m sure Zuck would have loved to have the whole company on the metaverse if he could.

It’s like saying that the root of western culture lies in Greek philosophy or Judeo-Christian traditions. It’s not totally inaccurate, but the realities of life and modernity means that it is an oversimplification of something that is complex and living.

It is useful context, however, just as studying Socrates is useful context for understanding Western thought.

That's up to you. I personally find it fun to meet people and I prefer being near nice smells. It doesn't mean that a rose perfume makes me uncontrollable.

Most perfumes aren't even naturally sexy smells. It's not like people are pumping pheromones in the air.

This would make sense if we were lizards, but it turns out we're humans with agency. The way that you describe it, the only way men can control primal instincts of having orgies on the streets is to make everyone wear a burka.

Everyone likes looking attractive. Just because a guy works out doesn't means that he's ready for quickie in the bathroom.

I don’t think it’s right for the hiring team to decide if an individual will “get too much attention”.

IME, women are all too familiar with the challenges of working in a male-dominated workplace, but ultimately just want to pursue a career they enjoy. The “right” decision is to respect that drive and not to infantilize them by making decisions on their behalf.

Media is horribly neutered because of censorship, but cultural exports like food are seeing increasing traction.

You’re probably over-planning.

Scale back and just plan for a release in 2-4 weeks. Ignore everything else outside of what you can get done in that time frame. Drop the consideration of things like the domain and the name. Fight the urge to play with fancy scaffolding that eats productivity. Once you’ve launched, you’ll have a foundation to iterate on.

Release early, release often.

Yeah, Google needs to make sure that their research doesn’t go the way of Xerox PARC.

Not exactly - there’s a bunch of non-trivial stuff you need to worry about when both local and remote states represent sources of truth. Things like CRDTs and event sourcing make it easier, but there’s still more complexity than dealing with only one source of truth. In an online-first experience, any local state being held is (for the most part) an optimization or is intentionally ephemeral. You can just toss it away at a performance penalty if it ever gets too hairy.

In an offline-first experience, you need to be very careful that you treat everything like a source of truth. You also need to deal with schema migrations and business logic migrations, since you need those to partially live on the client.

I’ve always considered traditional food to be based on generation. If your grandparents grew up eating it, it’s traditional.

Many foods comes from cultural cross-pollination.

> Doctors don't use unions. They instead artificially limit supply by erecting massive barriers to enter their workforce. This keeps them in demand. Software contrasts this with an open-door policy. On top of that, all white-collar labor jobs barring software involve serious licensing requirements as further gatekeeping. Some doctors do, however, collectively negotiate with hospital administration. The payment / profit sharing schemes can be very different depending on the hospital and what a particular group wanted.

You can get totally socialized profits at one place and totally performance-based at another.

Advances seem to make some of the hacks less necessary (such as CoT behaving better with newlines), but precisely specifying what you want will always be useful.

While true, I think the quantity is abnormal in terms of human history, since meat used to be considered a luxury item.

We use ngrok for testing OAuth2 handshakes on localhost.

> Consumer deposits have an average account size of less than $200,000

Whoa! FRB has wealthy customers. US average cash in savings accounts is $41,600.

And this is pure cash too - not total net worth.