Part of that solution was that the US and Europe outsourced their manufacturing to Asia. So the emissions are counted as Asia's, but they are for producing goods for US and European companies that will be shipped back to those markets.

China in particular has increased their coal usage, but they've also substantially increased their renewable energy. The solar capacity added just this year in Asia is more than the total solar output of North America. China's mix of energy is cleaner now than before, but total energy usage has skyrocketed leading to greater emissions.

The problem is solvable, but countries need to care about solving it. Industry has helped to ensure the political will isn't there. The fact that we need every country on board makes it politically difficult. We could all learn from one another and do things to solve it, but will we?

Subject says it all

So there will be yet another rewrite of Outlook that makes it even worse. Maybe they can make it feel even less like a Windows application than it already does.

The human body generates more bio-electricity than a 120-volt battery and over 25,000 BTUs of body heat. Combined with a form of fusion, the machines have found all the energy they would ever need.

It can't get warmer if there is no thermometer to read.

It is the intense wealth those few people have that affords them that enormous power, so of course inequity is at the root of it.

You don't get an oligarchy if your economy doesn't allow for an obscene concentration of wealth. You certainly don't fix the problems caused by the oligarchy by continuing along a path where the wealth transfer continues to get historically bad year after year.

Exactly. We're heating the oceans rapidly, as well as changing their chemical composition. We're making it more acidic and adding toxins. We're doing it on a timeline that is much faster than evolution operates at.

Since coral is a fundamental part of ocean life, the consequences can be a lot worse than just the coral dying out. It could devastate most ocean life. Some life on land relies on ocean life, so the impacts would be planet wide. A comparable situation that is easier for us land dwellers to understand would be if we lost all forests by the end of the century.

A rational reaction to this science would be to do everything in our power to stop this from happening. If the study is right and we do nothing, we've already put things on a path towards eradicating most life on the planet. Whereas if we take urgent action and the study turns out to be wrong and the coral would've been fine either way, we will have "wasted" a lot of effort to save the coral but we will also have given ourselves a less polluted planet with a safer climate.

USB-C is nice when everything is brand new. Compact size, no worries about getting the direction right, etc. But then it gets dirty or wears out and the plug doesn't stay firmly in place.

Whereas USB-A just keeps on ticking, even though they typically get more usage since every time you plug something in it seems to take 2-3 attempts to get the orientation right.

Why are so many websites trying so hard to replace functionality that already works with a half-baked AI?

They're spending money on development time to create this feature, which then requires them to spend money on data centres to power this AI going forward. In the end they get a worse website that costs more to operate. How can this possibly be good business?

Agreed. Unfortunately the bloated video was can be monetised, whereas the simple instructions don't make much money. So the economics of the internet have created an intensive to deliver information as inefficiently as possible.

Obligatory William Shatner

A few from "the other side" came over just to troll, then quit when they were blocked rather than getting rewarded with engagement.

Thankfully Bluesky isn't a private social media company that just wants to juice its engagement, which only leads to a gradual enshitification of the product (and society's discourse in the process).

That is both its biggest strength and its biggest weakness. There have been many versions over time with many ways to do things. It has almost infinite flexibility which means quite a bit of complexity. There are quite a few ways to not just shoot your own foot, but blow it completely off.

You can start a new project from scratch and if you stick to a specific subset of C++, you can create safe code. That relies on someone knowing what subset they should be using, which could be difficult for newcomers to the language. If you're working in an existing code base, there could be a minefield already in the code.

A language like rust has the advantage of starting from scratch, so there isn't an old unsafe way to do things that they need to remain compatible with. They can make the defaults safe so that it is harder to get into unsafe territory. Of course they have the same issue as C++ in that they need to work with C code, which then throws some of that safety out the window.

At times this strategy does work for them. Teams has made major headway thanks to it suddenly appearing as part of Office and even auto-starting on many Windows machines. They were successful at cutting into Slack and Zoom's businesses by forcing Teams onto the whole corporate world. Why use something else when Teams is already there. They're hoping to do the same with AI.

However, this Copilot stuff seems extra forced and comes off as desperate. They've re-branded their entire office suite to be "Copilot", plastered Copilot icons into most of their apps, and even put Copilot buttons onto new computers. They cranked up the hype way too high, and now they're basically saying "we've spent so much money on this; please use it!".

From a branding perspective, this reminds me of when they were hyping .NET. They put the name on everything including things it had nothing to do with. The end result was that people were confused about what .NET was. If people are confused about what the product is and why they should use it, they'll probably just stay away from it.

Emacs has a doctor feature based on ELIZA, which dates back to the 1960s.

Maybe if we throw a few more billions of dollars at it, modern AI systems will be able to do everything that emacs can.