Writing headlines is a selection process. You write about all the useless but cool stuff while ignoring all the boring but important stuff.

Improving Li-ion by 1% doesn’t make headlines, but that sort of stuff has been going on in the background for a few decades already. That’s why current batteries are so useful and widespread.

Lab prototypes are sexy, even if they’re 50 years away from becoming commercially viable. Sure, these things can charge fast, or hold a huge capacity, but they also tend to die after 10 cycles. Fixing that is going to take a long time, just like it did for Li-ion batteries.

Thanks for taking the time to explain this. Turns out there’s always more nuance to these things.

The blockchain solves one problem: trust. Do we really have that problem in normal energy markets?

People who buy and sell energy would need to trust each other and the middle man in between them. If they have trust issues, using a blockchain could make sense. As far as I can tell, the current system doesn’t suffer from a lack of trust, so what would the blockchain do in this case?

Edit: Turns out, it would be more accurate to say that it redistributes trust away from a central authority… well at least in most cases it does. There are still situations when even a blockchain has a central authority you need to trust. It’s not guaranteed to solve all trust issues, but in the best case scenario it can solve some of them.

And if that works out, you could also use the same method to shape the political atmosphere on the platform. Who knows, you might also disarm the next arab spring before the first demonstrations even happens.

I’m pretty sure governments are very interested in seeing the results you get from using LLM bots like this.

Yeah well… What can I say. Meta knows their users.

It’s for driving engagement. Basically, meta wants to make sure their platforms look appealing to the companies that use them for advertising purposes. More eyeballs, more ads, more money.

Don’t feed the trolls.

Yes there are, but would it be possible to replace them with ML and get more accurate predictions?

Yeah, that’s pretty impressive. I wonder if you could apply the same philosophy in other areas too. Instead of training the model with data produced in a simulation, you could just feed it real world data instead. Like, if you gave a bunch of stress-strain data to a model, could you make better predictions about the behavior of physical structures, such as bridges and towers.

About 4 years ago, this video showed that a ML model can be used to cut costs on physics simulations. It’s about time we did that with weather too.

As a part of grid balancing, we are already doing that to some extent. For the most part, the idea is that you can increase or decrease the load if you see the frequency of the grid beginning to drift off target. These types of frequency containment reserves can usually react very quickly, which means that most industrial processes don’t qualify.

However, since the duck curve is fairly predictable, we could (and should) extend this idea to slower processes too, such as the ones you mentioned. I don’t know if that sort of power reserve is actually being implemented, but it certainly would make a lot of sense.

It’s just that most industries prefer to operate 24/7. Having your reverse osmosis, electorlysis, electrowinning, arc furnace etc. running only during sunny hours is nice for the employees but bad for business. The investors of such factories prefer to see profits sooner rather than later, and restricting operating hours isn’t helping.

Cheaper electricity would obviously result in lower operating expenses, so I can definitely see some potential in this idea. You would just need to find some environmentally minded investors. They would also need to tolerate the risk that comes with a fluctuating power supply, which could be a tall order.

If the fluctuations of the local energy market are dominated by solar power, that means more work during the day and none during the night. If there’s lots of wind in the mix too, that could mean lots of night shifts during windy seasons and none during others, which isn’t great for the employees.

Exactly. Grid energy storage doesn’t have to be light or small. It’s not going anywhere, and you can build such facilities in remote locations.

Who cares if it weighs as much as a factory and takes the same space. You could go with molten calcium, redox flow batteries or even wilder technologies.

Have you’ve ever felt bad for buying cheap electronics or plastic products, because they aren’t good for the environment or the people working at the factories? Well, this article gives you a digital version of the same feeling.

It is possible that new battery chemistries or compressed air storage may prove cheap enough to use for long term storage.

There are plenty of options to choose from, but only few are actually industrial grade at the moment. So many promising ones are still in pilot stage, and I’m really looking forward to seeing which ones actually prove to be viable.

Traditional lithium based batteries clearly aren’t it, but LFP looks ok though.

How about investing in grid energy storage, to cope with intermittent production?

If you merely abandon your account, the change can be seen in the MAU statistics. If you delete it instead, you’ll make sure you won’t slip in the future. Either way, investors care about the active users, because the total number of accounts isn’t a very useful number to them.

Power demand would have to drop significantly for that idea to work. I don’t think semiconductors are even capable of delivering enough computation with only 2 mW. Maybe a completely different sort of technology could pull it off, but currently there’s nothing like that in the horizon, so who knows if we’ll ever get human powered devices. Maybe some tiny computers with hardly any processing power could be a realistic application.

Elon is so good with these predictions and promises, that there’s website for tracking them.

Oh, let’s add some financial motivation to hanging out in social media all day long. What could go wrong. Literally nothing. It’s just physically impossible.

It must have been a slow day at the news office. Literally nothing special to write about, so this article was slapped together.