That's exactly it. It's not like all these data centers are only going to teach models or perform inference. The goal is renters' economy.

Popular media likes to take this very simplistic view and turns the nations' economies into competition, forgetting that economies should work for the people, not vice versa. If it was a competition, then eventually someone would have to win (what is the definition of winning is unclear to me either) which so far doesn't seem possible. What actually matters is what a country does with the economy it has. Whether it translates into better lives, stability, and opportunity for its citizens. By that measure, "4th largest" tells us almost nothing. The more useful question isn't "how big" but "for whom".

I have seen it commonly cited (although haven't bothered to check the actual sources, mostly because I believe they should be taken with a grain of salt) that developers spend somewhere in the ballpark of 50-60% of their time doing "coding work" - that is writing the code, thinking about the solution in terms of technical aspects, reading code reviews. The rest are meetings, coordination, administrative tasks, being blocked or whatever else you can think of. Even if, wishfully thinking, the value of the act of "coding work" fell by 90%, the act of doing "software engineering" work would still cost no less than 50% of what it currently does. Headlines like these are alarmist and have little substance.

Edit: I also feel stumped why so many people give in to this hype that LLMs are good at coding when they can't even do seemingly simple tasks of plain English language summarization accurately as evidenced in https://www.youtube.com/watch?v=MrwJgDHJJoE. If the AI summarizes the code in its own context incorrectly then it will not be able to write it correctly either.

It's relevant because it shows models haven't improved as much as the companies delivering them would like you to believe no matter what mode (or code) they work under. Developers are quickly transforming from code writers to code readers and the good ones feel demoralized knowing that they could do it better themselves but are instead forced to read gibberish produced by a machine in the volume of dozens of lines per second. Moreover, when they are reviewing that gibberish and it doesn't make sense, even if they provide arguments, that same gibberish-producing machine can, in a matter of seconds, write counter-arguments that look convincing but lack any kind of substance for those who understand and try to read it.

Edit: I am saying it as a developer who is using LLMs for coding, so I feel that I can constructively criticize them. Also, sometimes the code actually works when I put enough effort to describe what I expect; I guess I could just write the code myself but the problem is that I don't know which way it will result in a quicker delivery.

With the pace of inflation we have been witnessing over the past years $1T has become unimpressive. Let's talk percentages. And if somebody wants to talk about absolute numbers, they should talk not only about negatives, but positives too, as in how much the stock market has gained before losing that $1T.

Fiat currencies are becoming meme tokens at this point. I think we are in a hyperinflation era and consumer price growth will simply have to catch-up to the stock market growth rate because wealth isn't being created right now, it's just being inflated.

All of those who are investing in stock market and thinking they are becoming rich might just realize that those were the paper gains when they will still not be able to afford anything with all the big numbers in their investment accounts.

Too many services is a good problem to have in a regular market. The problem is the accessibility and fragmentation of content. If all services had the same content, they would have to compete on the features, price and performance, which would be a healthy form of competition.

Yes, need proper standardisation. we are now in the era when there’s fragmentation in a tape audio players, with dozen of different incompatible cassette tape formats and musicians having lock-in contracts with one format each. To listen to all music you would have to buy each player on the market. Only piracy allows you to buy just one tape player and have copies of all other tapes to play.

Not talking about this specifically but I am very skeptical towards biographic books because they make the most despicable individuals look how they want to be viewed.

People overestimate computer vision and other AI capabilities. A few days ago I had a relative struggle to open some fancy cosmetics box. Another relative decided it would be helpful to photograph it and send it to ChatGPT for help. Coincidentally, not that long ago I also heard someone telling about the recent AI advances on the radio and talking how they send photos of things and ChatGPT helps them figure out how to use them instead of reading those boring manuals.

I guess that's what marketing does to the people. But also it doesn't take many failures and broken expectations until those people decide that it's not worth the effort and stop using these tools entirely.

I once reported this kind of interview scam repository with the full backstory and explanation why I was reporting it and Github's support asked for a proof that it was a scam. As if I was supposed to do the detective's work. I just wrote back to them that they can do whatever they want with it as I've done my part.