“Abstractions should be written in blood”

For small enough n, linear search might have been faster.

I don't think glibc malloc makes an optimal set of tradeoffs for any scenario.

Imagine what $100B could have done for medical or energy research.

As a thought experiment, they probably helped the world global happiness more by burning $100B rather than just sitting on a massive cash hoard as so many tech companies enjoy doing. Sitting in their coffers it was doing nothing. At least this way, it went to a whole bunch of individuals, manufacturing startups, etc. And theoretically many of those recipients spent some of it.

I'm not saying it wasn't wasted spend, but velocity of money is a thing and maybe it's better off in the hands of the people who it was spent on instead of sitting in Zuck's war chest.

https://en.wikipedia.org/wiki/Velocity_of_money

Can I just remind people what Facebook the company is responsible for? And I'm not talking about Myanmar and so on and so forth and I'm not talking about them as a VR entity. People miss posts all the time because the algorithm won't show it to them, even from close friends. People had accumulated "Other" inbox messages for years, that they were never told about. And so on. "Helping the world global happiness" was never a priority

I wouldn’t call what Meta is providing to the world “happiness”. Addiction is a lot more apt.

Pakistan spent under $20bn to convert to about 25% solar. $100bn would probably have been enough for the whole country of 250 million people.

Almost nothing. That is a few weeks of global medical R&D spend, spread over more than a decade.

Annual R&D spending of pharmaceutical companies:

Merck - $17.9B

Johnson & Johnson - $17.2B

Roche - $14.6B

AstraZeneca - $13.6B

AbbVie - $12.8B

Bristol Myers Squibb - $11.2B

Eli Lilly - $10.99B

Meta’s losses on Metaverse last year - $19.2B

So, simply redirecting their spending in that division would instantly propel Meta to be the biggest medical researcher in the world. And as a bonus they’d get a real return out of it.

It would cover the entire NIH budget for two years. That isn't nothing.

40 nuclear power plants.

They are being downvoted because Nuland is an utterly insignificant diplomat, but serves well as a dog whistle for people who subscribe to the belief that the Maidan revolution in Ukraine was really some Obama organized coup. This is a story peddled by the Russian government, which of course is where Yanukovych promptly fled after having protestors shot. At the same time Russia was busy staging troops and material for the actual coup they were planning.

Yes, insignificant enough to remark "Yats is our guy" (during an intercepted but confirmed-authentic conversation with another US diplomat that Russia subsequently leaked), only for "Yats" (Arseniy Yatsenyuk) to subsequently become the Ukrainian PM. Coincidence, I'm sure...

ISW is a shill for neoconservatism (just look at their board!) and is funded by US defense contractors. They try to give the appearance of neutrality via technical jargon etc. but are anything but.

In this case the threats all seemed to be delivered in Hebrew, so it's plausible that those involved resided in Israel.

Yes, it seems likely.

I assume the author would have noted in his blog post if the Hebrew used in making the threats came across as non-native or non-fluent.

Machine translation is really good these days, but I still think there are some "tells" for a truly vernacular and local command of a language.

Because translation is so very hard?

Yes, for colloquial speech to be plausible to a native speaker.

I assume the strange English translations of some of the Hebrew messages are because the messages were badly translated to Hebrew. It certainly sounds like the kind of result you'd get out of Google Translate.

“outdated internal wiki” has to be responsible for so many AMZN outages…

Shows how although AI is great, good ol' issues that we had in human-coding times are still persistent and problematic even during the AI-age.

Actually CoEs would make an amazing training corpus for code reviews.

much like google doodle