diziet_sma

I couldn't find any evidence Jane Street is a customer, only that Jane street is an investor. Do you have a source for that?

Very cool! I would imagine this helps LLMs catch errors while refactoring code.

How hard is it to migrate existing pandas/Polars code to Typol?

Most importantly, how did you come up with the name?

That wikipedia page mentions SpaceX reached a deal to lease all of Colossus 1 to Anthropic: https://en.wikipedia.org/wiki/Colossus_(supercomputer)#cite_...

So will SpaceX lease Colossus 2 to Google? If both Colossus datacenters are rented, will xAI have any compute?

> Python essentially bet on UTF-32 (with space-saving optimisations)

How so? Python3 strings are unicode and all the encoding/decoding functions default to utf-8. In practice this means all the python I write is utf-8 compatible unicode and I don't ever have to think about it.

Universal search on Google Pixels has solved a lot of the text selection problems on Android for me, with the exception being selecting text which requires scrolling.

I think self-driving is what makes Waymo premium. A Waymo is a better experience because the Waymo is safer and smoother driver than any human could ever be.

I took the parent post to mean that a few large firms have emerged as clear winners of the speed game, and most other companies compete on (relatively) longer time scales now.

It's not the same thing as multidimensional arrays though. You can have multidimensional arrays without rank polymorphism. Rank polymorphism makes working with multidimensional arrays much easier because you can write one function which works over input arrays with different shapes.

Yes we do. The OP's point is that _if_ there is collusion, building housing will help solve the collusion problem _and_ solve the housing shortage.

Spending effort on theoretical collusion which may or may not be happening is a diversion from the real problem, which is lack of housing supply.

Or perhaps cooking? Cooking can really spike AQI if ventilation is suboptimal.

Another potential explanation is cleaning products.

I think these are both far more likely than perfume, as there is a much stronger link between AQI and lung cancer than perfume and lung cancer (if there is any at all).

There have been many attempts to make a Python JIT that is compatible with CPython to various levels of success. However the larger reason is that the gains from removing the GIL far exceeds the gains from a JIT.

If you're writing performance sensitive python code, the "hot" code is likely already in a C-extension, such as Numpy. So there is negligible benefit to running the code with a JIT.

How long do we expect it to take dynamic derivations to get stabilized?

The GitHub RFC says it's only 80% complete

  fd --help | rg link -C 10

Set -L --follow to descend into symlinked directories. The default is to not.

These aren't financial exchanges, they're a sports betting and an expense management system.

I share OPs skepticism. Market makers invest in microwave towers, FPGAs, etc. I would be surprised if sqlite backed by NVME is on the other end of all that specialized hardware.

Order matching is a single threaded thing though. I would be curious if anyone knows how electronic trading systems are actually implemented.

.... I mean in the case of more efficient cars supply of oil explicitly did not change.

There are meaningful differences in terms of pricing strategies, anticipating demand, etc.

Yes, but for subtly different reasons. In the case of induced demand, the supply curve shifts to the right (eg number of highway lanes increases) so we move to the right on demand curve (or quantity increases, eg more people drive). On the Jevon's paradox case, efficiency increases (eg fuel efficiency increases), so the _demand_ curve shifts (eg more people people drive).

That isn't even a remotely realistic propaganda threat, while tick tock arguably is.

> We’ve seen this pattern in speech recognition, computer chess, and computer vision.

I think the bitter lesson is true for AI researchers, but OP overstates it's relevance to actual products. For example the best chess engines are very much still very much specific to chess. They incorporate more neural networks now, but they are still quite specific.

Intel products have been losing x86 to AMD in servers, desktop, and laptops. x86 itself is losing market share to ARM in the server and laptop space. Intel has no foothold in the fastest growing chip space (GPUs).

Intel foundry has fallen behind TSMC to the point the latest Intel laptop chip (Lunar Lake) was manufactured using TSMC. Recent attempts to attract external client s to fab using Intel Foundry have not met expectations.

Given all this, I think quite a lot of hand-wringing is warranted.

If I didn't know the original quote I would have thought the same. It would be clearer to use L̶A̶R̶R̶Y̶ Megan Ellison

A significant amount of zoning is about what type of housing you can build where, how much of it you can build, and how you have to build it.

For example, in Berkeley California, which was one of the first cities to implement "exclusionary zoning" has roughly the following zoning restrictions:

* R-1: One home per lot or estate, only. Bans apartments in 49% of the city.

* R-1A: One home per lot or estate, unless the parcel exceeds 2,400 SF which allows for an additional home.

* R-2: Two homes on one parcel, only.

* R-2A: One home per every 1,650 square feet on a parcel. A typical residential parcel in Berkeley is about 5,300 square ft thus commonly three homes maximum.

[source](https://darrellowens.substack.com/p/berkeleys-upzoning-would...)

This does help, but the hard part of memory leaks is leaks across module/api boundaries. You can write a memory safe module, I can write a memory safe program, but when my program starts calling your program there's nothing validating I'm maintaining the invariants required to avoid memory leaks.

Hopefully the invariants are documented, but relying on documentation to help programmers avoid memory leaks is far from foolproof.

A similar thing happened to me, but one day my keepass file was corrupted instead of 0 bytes. Quickly switched to 1Password after that.

> Nobody Asked For Them

After riding a mountain bike with disk brakes, I couldn't wait to get a road bike with disk brakes.

> Then Why Did We Get Them?

Because they're better. Full stop. I can stop faster with less effort.

> Rim Brakes Are Better For Most Riders Most Of The Time

I don't need any brakes at all most of the time either, but when I need them I really need them.