

It’s got all the worst parts of extreme anti-fraud measures, like freezing your account and holding onto your money just because you did something suspicious like receive money, but without actually protecting anyone from fraud.
It’s got all the worst parts of extreme anti-fraud measures, like freezing your account and holding onto your money just because you did something suspicious like receive money, but without actually protecting anyone from fraud.
Yes. As others in this thread have explained, they’re approaching peak coal and that line is not one that you can extrapolate upwards as a straight line into the future.
I also think it’s not reasonable to compare a developing/emerging economy with hugely increasing total energy requirements, with ones that already got their polluting growth phase out of the way in the 19th-20th centuries, especially when a very significant part of that coal is burned in the service of making consumer products for the latter. It’d be much more reasonable to compare them to India, which oh look, they are doing much better than in both current percentage and growth rate. Whilst it’s true that Africa is doing better in those graphs, they’re also not having nearly as much success in production or growth terms.
So overall, yeah it could be better on paper, but it’s very much treating perfect as the enemy of good and preaching at a country who built as much TWh solar&wind capacity just in the last 12 months of your graph alone, as the USA has over its entire lifetime.
(I was about to draw a few more conclusions from those graphs but noticed they’ve left out a bunch of other energy sources for no obvious reason, possibly mischief, so I can’t compare - the graphs imply that these regions are replacing coal with solar&wind, but without the data for total consumption including gas, nuclear, hydro etc we don’t actually know what the true situation is.)
There’s such a huge gap between what I read about GPT-5 online, versus the overwhelmingly disappointing results I get from it for both coding and general questions.
I’m beginning to think we’re in the end stages of Dead Internet, where basically nothing you see online has any connection to reality.
Your own graph shows the ratio of renewables to coal hugely shooting up in the last 4 years.
How naïve it was of me, to think that the New York Avianca case in 2023 was high profile enough for lawyers to have learnt their lesson, but nope, it’s getting worse each and every month that goes by:
https://www.damiencharlotin.com/hallucinations/
It doesn’t help that the most common outcomes there are “Warning” or a fine in the low thousands. If a legal practice can save $500,000 a year on avoiding doing their own research, and the worse that’s likely to happen is “Warning” or a $2,000 fine, then why would they not?
I’ve found them useful, sometimes, but nothing like a fraction of what the hype would suggest.
They’re not adequate replacements for code reviewers, but getting an AI code review does let me occasionally fix a couple of blunders before I waste another human’s time with them.
I’ve also had the occasional bit of luck with “why am I getting this error” questions, where it saved me 10 minutes of digging through the code myself.
“Create some test data and a smoke test for this feature” is another good timesaver for what would normally be very tedious drudge work.
What I have given up on is “implement a feature that does X” questions, because it invariably creates more work than it saves. Companies selling “type in your app idea and it’ll write the code” solutions are snake-oil salesman.
If you look at all the unicorns of the past few decades, a surprisingly large number of them did it with software that wasn’t in any way technologically advanced, but exploited technology to find loopholes in the kind of industry regulations that were there to stop companies from screwing people over.
PayPal was a way to do banking without registering as a bank. Uber, Doordash and other gig economy apps are exercises in sidestepping employment law. Airbnb, despite its origins as a couchsurfing app, didn’t get huge until professional “hosts” started using it as a way to run apartment hotels without having to meet the expectations or obligations of one.
If you want to build a tech unicorn, all you need to ask yourself is, “how can I make something 5% more convenient and 200% more shit?”.