I guess the obvious difference is that one of these is falsifiable - work can be done to figure out if the 'letter of indulgence' was attached to any actual action that had an effect in the real world, as is the case in this article, and appropriate action taken.

Arguing about the effectiveness of carbon offsetting schemes vs whether or not individual schemes are fraudulent are two different things. On the surface this sounds like straight-up fraud - the Greenpeace article implies that it is legal, which would be stupid, and hopefully the laws would be modified to make sure something is actually being done.

Otherwise, I quite like the analogy!

Now inconceivably vast amounts of highly detailed and personal data on citizens will only go to that one safe place, American corporations. Well, I guess also by extension, the American government. But what are the chances anyone would come to regret that!

Paying to export power seems weird. Usually you're paid to export it? Unless neighbouring states also have a surplus but are more readily able to use cheap/negative rates?

So slave labour is OK in every other product that is imported tariff free into the USA, just not cars?

It's not about "slave labour"., to whatever extent that actually exists. It's a useful thing to shout about but nobody has ever cared about this as long as it means cheap goods.

It's about protecting the car manufacturers. And this is fine, actually! Just be up front about it and say "these tariffs exist and you can't get cheap Chinese cars because we are trying to preserve the American auto makers".

Yeh, I am saying that this should be expressly forbidden. "Only companies with no debt can do buybacks" might be too simplistic but it's probably a better starting point than what we have now.

I think the problem is when buybacks happen with money from debt

It can run a lot of native apps. You can run a Linux mode on it and install stuff via apt. I don't quite understand how it all works - it's some mutant container/virtualisation thing - but it /works/.

It's Linux on the desktop. It's great for SMEs and has tons of enterprise features. Anecdata, but more than half our SME is.ok Chromebook because it's a lightweight, cheap browser focus desktop environment, and we can set up full Linux Firefox on it if needed (we did for a while to run a few websites that needed user side certs, which weren't supported in Chrome at the time, but are now).

It's the most underrated OS out there I think. People are going to blink and it's going to be everywhere - even more than it is now. (I say this as someone very critical of Google for all the usual reasons.)

Hey do you pay for a blue checkmark by any chance? Seems right up your alley

Thanks for the very comprehensive reply! I didn't realise it was so far along. I am in Australia and I don't think it's available with that level of functionality yet here. I am still pretty wary of computer vision only solutions but it sounds like they are getting something that works better than I would have guessed.

Is the feature called FSD but it's basically just lane assist stuff? Or is there more to what they're calling FSD? I was under the impression there isn't a comprehensive set of "self driving" features available just yet even for those that bought FSD as a package.

So like a battery, but the energy is wasted on garbage and heat

Unless I'm missing something these are universities spending a few tens of thousands on some marketing stuff. This is peanuts for a marketing spend for a large org. Like a rounding error in terms of what they'd spend annually. What am I missing?

Where did you get 431? A quick search tells me that is a pre 2022 number and the 2023 number is closer to 380

Microsoft are all in on the LLM AI push. Stunning amounts of money being spent and a huge commitment across their entire product base.

I get it, the upsides look shiny and impressive. But the downsides are still so real - the failures modes of this AI are so catastrophic and so subtle that it doesn't feel like it will take too many high profile disasters for this bubble to pop massively. I would be a nervous investor.

To be fair if they actually got to the pint they were overcharging customer, they did a lot better than many other tech companies, who would have had to blow venture capital dollars to achieve the same thing.