From my math, once you account for inflation, Silicon Valley (and broader tech industry) compensation has declined about 23-30% since 2000, relative to the national average. I know multiple companies where it's basically been flat since 2012, particularly for support roles - $90-100k then is only making $105-110k now in the same roles (with some of the same people, granted, but also new blood).

Considering how many foreigners from specific countries I see in places like Omaha, NE, I'm pretty sure it's not just a problem in Silicon Valley.

That may be true, but it's irrelevant.

They'll 500 500+ Americans from Division A while hiring 500 Indians to Division B.

And then the next quarter, it's fire 500 from Division B (all Americans, mostly white men), and then hire 500 Indians into Division A - because they had a shortage, don't you know.

How was it not fraud?

H1B is/was intended for worker shortages: for roles you couldn't fill domestically. It was not supposed to decrease the wage of the role, but (in theory) keep it at parity. It was always a ruse, but that was the premise.

If you're firing people and then hiring others at a fraction of the pay, you're literally abusing it. You've applied under false premise. That's fraud.

This 'study' was clearly constructed to create narrative, not to do actual science.

"The study was conducted with 189 people."

And what was the distribution of men and women in that sample? Because that's a very important criteria.

I promise you the results would be significantly different with all-male vs all-female groups, and depend heavily on the attractiveness of the 'male' or 'female' AI persona (assuming it had visuals).

I felt that Duke Nukem 3D was a journey of ripping people's heads off and shitting down their throats ;-)

It's worse. OpenAI train on user conversations and monitor them for keywords, looking to scoop people's private ideas. This has been documented in individual instances, it is also implied in the terms of service, and it is common sense in a world where the primary training materials (books and web conversations) have been exhausted while the system shows no sign of improving on its own.

Luckily their financial ponzi scheme isn't going to last too much longer. Neither will Anthropic's.

D'Oh!.

There's also a novel that states that the lifeblood of an organism can be used to bring eternal life to certain parasitical creatures (who otherwise can't survive in sunlight). It's probably worth washing the transplanted hearts in holy water before inserting them into the recipient.

"Past performance is no guarantee of future returns" applies to any recorded dataset. Modelling a fictional being's mental states is fun, but stating that it is useful is premature.

The observed fictional being might be always sleepy in your dataset (because it's easier to record him when he's not moving much). And identifying the being's favourite colour (orange) may be irrelevant and wasteful in your model.

If you observe yourself in the mirror, and see your reflection eating spaghetti, is it a useful framework to pretend that the spaghetti is real? If you're still hungry, will you reach through the glass to steal the plate from the other you? And should you first make a plan how to defend the loot, in case the reflection becomes angry and punches you in the face?

Python isn't a glue language. It's the lingua franca of most programmers in the workforce at this point in history.

Try hiring some expert C++ programmers, or even competent Java programmers. They're in the minority. What you get instead is barely competent grads who only know Python. What do you do with those grads after you've hired them?

1) teach the grads a strongly typed industrial strength language so they can contribute.... eventually

2) develop your new projects in Python because the grads can be productive straightaway, and the long term issues with Python can be pushed off to a future manager.

Which brings me to my point: Python should be frozen as is. No more development. Tinkering with it just because some people want to contribute and get visibility in open source is not a good cost/benefit outcome for humanity.

Freeze Python. It's the modern BASIC. Move all the existing developers to a new, unrelated programming language project. Give them free reign to do whatever they want there.

That's not the important metric. It only tells you the percentage of users, not the percentage of people who play games on Linux (or buy games).

For instance, I have a Steam account. I've got the client installed on my phone and Mac. I don't really play games anymore. I'm sure this counts towards something.

What's meaningful to those who might target a platform is what percentage of those people buy games. It might just be that Linux users are the only ones who play dungeon crawlers, for instance - which could be meaningful information.

I'm still impressed that the Teams web client can't edit a message with an emoji in it without messing up the keyboard controls and often corrupting the message and losing data. It's had this bug for years. It's literally a messaging app where you can't reliably send messages.

I call article shenanigans.

Element is a UK company. Which is not part of the EU. Why should the EU choose to use a non-European company, headquartered in a non-European country, with a non-European, non-compliant legal system?

The UK is already barred from offering unrestricted financial services in the EU, I expect communications services to be an area of concern.

Brexit is Brexit. The UK is out, and simply cannot be trusted with European communications backbones.

I may be willing to believe that there's been a tender process and their product is currently being evaluated, but on privacy grounds alone I doubt their product passes the bar unless they reincorporate in an EU country with 51% EU ownership, all other things being equal.