initplus

The government actually wants banks to speculate with customer funds. You might think that the safest bank would be one that doesn’t do anything with customer deposits, just sits on them.

Well this “safe bank” actually falls foul of US banking regulation - there are regulatory requirements for minimum levels of “speculation“ as a bank.

Now clearly SVB got in over their heads here with risk. But just sit on all the money isn’t a legal alternative.

People are getting too caught up in the 250k limit.

Most bank failures result in all depositors being made whole - even over the limit. Normally by regulators “hinting” that a larger bank should acquire the failing banks assets whole.

FDIC exists to create faith in the banking system - no to just enforce a rule book on how deposit insurance is structured. The goal is for FDIC to never have to pay out in the first place - FDIC’s existence prevents the bank run from starting.

The obvious conclusion to the 250k limit (splitting deposits between banks) doesn’t really do anything to actually reduce risk in the banking system, or risk to FDIC. Splitting deposits like this is artificial behaviour that doesn’t have any real benefits to overall stability of the economy.

The government wants it to not be a possibility. Bank accounts are meant to look safe.

If government didn’t make it clear they were stepping in here, what do you think would have happened first thing Monday morning? Every small/medium business would have seen the writing on the wall and moved their funds out of small/medium banks and into the handful of “too big to fail” megabanks. Stepping in and making up the cost prevents contagion.

The 250k limit is somewhat of a red herring. If every depositor split their funds up to stay under the 250k limit, it doesn’t actually change the overall risk profile for the FDIC.

What bad behaviour is being enabled here? The people who stand to gain from taking risks here (bank owners) have lost everything and are not being bailed out.

What is the bad behaviour of the depositors? Trusting that their deposits were safe in a bank? This isn’t bad behaviour, the government wants people to think that bank deposits are safe. Safety of bank deposits is really important to the economy.

They are treating it differently because the alternative is a run on every small/medium bank with high numbers of business customers.

It’s not to protect the existing depositors of SVB - but to shore up the stability of the entire banking sector.

Think about this with some numbers: if there are 100 banks, and every business puts 1% of their cash into each bank, the overall risk is the same as if 1% of businesses put 100% of their cash into a random bank out of the 100.

The cost to FDIC if an individual bank fails is the same in both the above scenarios, even though in the first businesses put a lot more effort into spreading out their funds. It looks less like it could have less risk to the FDIC, but really isn’t making any difference.

The way to prevent bank runs is to prevent banks from becoming insolvent in the first place. Making it illegal to take funds out of an insolvent bank is not the answer.

SVB catered primarily to businesses - the last thing we need right now is every business worrying about where their bank deposits are. If they do worry, it may trigger more runs as they all try to move their funds to “safer” banks.

Whether an item can be delivered to a particular address seems like a seperate concern to whether the address is correct/real.

The L3 cache appears again in Windows, so it's clearly there physically on the die. But under the default performance tuning of the deck seems like it's not really being used. Possibly they are turning L3 off/down at runtime?

The missing L3 cache is interesting, I can only assume that it's another sacrifice made for power optimisation. Maybe their perf metrics showed disabling L3 provided a worthwhile power/perf tradeoff?

Right, what parent called "expressiveness" is probably more accurately described as "flexibility". And flexibility is not always a good thing.

But the purity is implicit. Nothing will stop you from refactoring add into an impure function. As a programmer you have to maintain this mentally by remembering which types are pass by value and which are pass by reference.

I feel like this doesn't make sense because mainstream functional languages all provide a way to express imperative code.

In Haskell you can write an entire application using IO/do blocks and it will be completely imperative. In f# you can declare every variable mut.

Functional languages don't prevent you from writing imperative code. They give you the opportunity to choose not to, and enforce it at compile time. The opposite is not true of imperative languages, there the type system cannot help you to enforce that some code is purely functional.

Any programming language represents limitations placed on the underlying expressiveness of the hardware. By this argument raw x86 asm is "more expressive" than any higher level language.

The whole point of a programming language is to provide you with common patterns/designs to simplify reasoning about your program. Even base concepts like a "function" or "struct" is really a limitation on the underlying expressive power.

Immutability/purity does solve real problems for programmers: it means that you only have to reason about mutability at the margins of your application. I think people who don't have experience with functional style don't realise how helpful this is. I choose to use functional languages because it makes me a better developer - I am able to write code that I am not diligent/smart/disciplined enough to write in an imperative style.

F# is a great language, but it somewhat waters down its strengths to bolt it to a giant hulking lump of C++ full of mutable state. Interacting with an impure api means throwing out many of the advantages of (mostly) pure programming.

American Virologists feared that their field would be blamed for the pandemic, and used motivated reasoning to convince themselves that the field could not be responsible for the pandemic. And the US was gripped by Trumpism, so anything in any way connected to Trump automatically became verboten.

Some of the leading voices in opposition to the lab leak hypothesis such as Peter Daszak had close academic connections (collaboration, grants etc.) with the BSL4 lab in Wuhan.

The circumstances that make the theory unfalsifiable (lack of transparency/controls in Chinese biological research) are the issue here. Due to the lack of transparency we will likely never have definitive proof for the origin one way or the other.

At a minimum the US government should not be funding any biological research in places without adequate transparency/controls.

Even outside the context of refactoring/extending an existing codebase the code examples seem bad, look at his example of a prime generator from the bottom of the article. Think it’s more a case of “your own sh*t doesn’t smell”.

You could overload operators like + and / so they evaluate to an AST representing the calculation rather than the actual calculated value.

Then you need to do some analysis on your AST to try and restructure it in a way that preserves as much precision as possible.

You don’t really need Haskell for this though, in theory it will work in any language (but more ergonomically if you have operator overloading).

The correct solution is to use an isochrone rather than a basic radius search. Or to simply sort all the shores within the max search radius using a vehicle routing library/api.

Exactly. This seems analogous to Apple demanding 30% of the bank fees I pay because I use my bank's app.

Sure, but this is a way that sports betting is less like gambling and more like legitimate risk management for some businesses.

I have a hard time taking this seriously, it's the objections of one individual who seems to be seeing ulterior motives behind WASI... I don't think their objections are really as strong as they make out. Big red "In violation" labels are a bit much.

WASI is trying to be a cross language API for use by multiple languages that target WASM. But the author wants this API to more closely follow existing JS conventions? Sure that's useful to JS/Web, but it's not useful for anyone else trying to target WASM with a language that's not JS. And if you want to target Web API's with JS, can't you already just write native JS?

Bikeshedding over string encoding for example and objecting to WASI's promotion of UTF-8 because it doesn't match the encoding used by JS.

There is no real market participant who could benefit from hedging in sports betting. Only sports betting operators themselves, and the teams (who are legally forbidden from participating in betting).

But in a commodity market there are real participants whose primary business model is production, being able to hedge against future price swings provides real value to these businesses.

The alternative isn't SBF keeping the funds. It's the funds going through the US bankruptcy proceedings alongside the rest of FTX's assets.

SBF chose to operate out of the Bahamas because of their lax legal structure. The fact that he claimed for days this was a hack makes this look like a failed attempt at buying his freedom. I have much more faith in the integrity of the US bankruptcy process than whatever nonsense is going to happen in the Bahamas...

I believe the Bahaman incorporated entity is "FTX Digital Markets Ltd" https://bfsb-bahamas.com/providers/name/ftx-digital-markets-...

This is the entity (alongside the physical presence of SBF) by which the Bahamas are arguing they have jurisdiction over the FTX bankruptcy proceedings.

FTX chose to file for bankruptcy in America. There is a jurisdictional dispute between the US and the Bahamian court over where the bankruptcy proceedings should take place.

In my view FTX chose to incorporate in the Bahama's for a reason, and that reason wasn't the transparency and integrity of the Bahamian legal system.

Europe doesn't have a large scale crypto mining industry. Energy prices aren't competitive enough there for it to be viable.

Europe's hashrate could disappear overnight and barely anyone would notice.

>Satya: Thx for the chat. Will stay in touch. And will for sure follow-up on Teams feedback!

I wish I had a direct line to the CEO with my own Microsoft teams complaints...