This has been my experience. Asking the AI agent to code something but missing some detail that I would have expected another developer to follow up on. The agent doesn't ask the follow up question to clear the ambiguous part and just assumes the answer. It will create something off base enough that i have to spend several hours correcting because it can write the code so fast i cannot possibly type at that rate. When its all wrong it completely wasted my time.

This changed my philosophy on how to interact with the AI agents. I now spend most of my time working with the models build up documentation on the project goals and design. After I'm happy with the documentation, I ask it to implement what was documented, and from there its much better at the writing the resulting code. However, no matter the model it still needs quite a bit of hand holding to get the finished details in place. It might be frustrating, but if I can get the models to smash out 20k lines of code in a week and I'm reasonably happy with the results, I feel like its a win.

I admit I don't interact much but I do read regularly.
Does that mean I might as well not exist?

https://www.reuters.com/articl...

China definitely retaliated, and chose the targets politically.

Culpability is 100% what China is trying to avoid. It could be as simple as not acting fast enough (basic denial) up to and including the world wide problem WAS CAUSED by THEM (lab accident). It doesn't matter which side of the scale they are on its only that there is any culpability at all with China that they want to avoid.

The sadist part is that which ever side of this it is, it undermines trust in the world community. Their (China's) typical response is normally: "The world didn't react fast enough" and the finger pointing begins.

Thanks for taking any ownership, really...

That is a really interesting read. I'm aware that they are completely different laser designs though. The ultra powerful ones compress the light into the shortest amount of time possible. With that they can do particle physics sorta similar to the large colliders, just in less space and somewhat less energy. These navy ones need continuous beams. Both are very awesome tech.

It makes me think that for any of the claims in the article to be plausible, they must be exploiting some new material science effect much like the doped optical materials in lasers that amplify or compress the beams.

Lasers have gotten crazy powerful https://www.popsci.com/researc... however these are limited to labs. Heat up all the air in between the laser and the target causes all sorts of problems, so not many projects still attempting to use lasers for military purposes.

Also, the US Navy rail gun seems to be progressing well https://www.popularmechanics.c... not sure how long till they have it mounted on a ship though (if not already).

https://www.worldometers.info/... Most susceptible seems to be 50yr+.

I'm glad I'm not the only one who looks at the UID when reading comments.

I completely agree, total mass to orbit would be a better measurement. I did some brief searches but I wasn't able to find anything. However, between the Apollo program and the space shuttle that would easily put USA on top of the total mass to orbit.

More recently though, the Falcon 9 and the Atlas V are both very capable launch platforms. While China is catching up to them, most of there launches are on much smaller rockets with a little more than half the lifting capacity.

I assume you are not familiar with the phrase Tyranny of the majority

The EC was specifically meant to prevent such a problem. Another thing that most people don't understand is the USA is actually a collection of small countries that have agreed to a federal standard of laws. The laws not specifically described by the federal standards are left to the states.

This leads to the leader of each state is being selected by the people who inhabit it (usually called the governor), and the leader of the collection of countries is elected by the states (in the USA's case the president).

If the popular vote was allowed to selected the president, the most populaces states would forever dictate who was the president, and the remainder of the country would be forever under their rule. Which is what they were trying to prevent when designing the EC.

I was thinking exactly the same thing and was wondering if the network engineers were just too dumb and disabled default features to prevent this kind of problem...Then I read the FCC report. It not clear if the equipment that CentryLink uses are actually routers or just glorified switches. Also it describes how they have a intentional network loop setup for 'redundancy' but no indication of any routing or switching protocols running to prevent loops. So its probably a combination of inexperienced network engineers and equipment from the lowest bidder where the software and its defaults are not very good.

And similar to this concept truck by MAN: http://blackandgold.com/ee/507... Which seems to pre-date the Nikola design by 4 years.

Netflix was amenable to having their movies play on big screens in France, but a law in the country requires movies to not appear in home platforms for 36 months after their theatrical release.

Where does a law like this originate? So if it is shown is a theater, at all, it automatically cannot be shown on any other medium for 3 years? I can only see this as a law to fuck content creators over because Theater operators have more say with the elected than the electorate. Maybe this is a sign that most of the 'would be' democracies are oligarchies as well :(