It's been around for quite a while. It's seems to be a replacement for C and C++. Some see it as an alternative to Rust, but it doesn't have the same safety features. I would love to find a language that's a replacement for C++ that can be used to migrated existing code bases to, but none of them are, really. That's partly C++'s fault. It's very hard to interoperate with C++ code and libraries from other languages, without layers of wrappers (PySide comes to mind for Qt). Even interop between different C++ compilers is difficult. So for now I stick with C++.

New Glenn booster can also glide a bit during descent, not unlike the Starship. The strakes on the rocket create a tiny bit of lift, so it has a more flexible landing envelope than the mostly ballistic descent of the Falcon 9. I was very impressed by what New Glenn did on its first successful landing.

Wouldn't doubt it at all.

Maybe I'm just a weirdo but I am very annoyed at them for trying to take away the option of local-only accounts. Why do I need to let them be a third party to everything I do starting with logging in? Nobody asked for this.

"It's not like SpaceX did not have any missteps on their path to creating reusable boosters."

They weren't really missteps. It was part of their design philosophy. Build it enough to get past a "goal" (say, get past the launch tower) and test. If it doesn't meet the goal, ID the failure, redesign and test again. Once it reaches that "goal", create a new "goal" (sat, reach 20,000 ft). Repeat until it's reliable.

While this involves a lot of explosions, the actual time it takes to get a workable and reliable rocket was dramatically reduced.

Looks less like a failure on China's program and more like China learning from Musk.

Or even just good old fashioned human-driven competition. A 60x price increase will draw new competitors. I'm not saying this is easy and I could do it in a weekend or a year, but still, this is not rocket science.

OpenAI got out ahead but really, how do you beat google at this?

Technology-wise, they've had top researchers all along. Want more? Just hire them, not hard when you have infinite money.

And google has access to everything. They serve about 1/3 of the population on earth every day. Not just search but webmail, texts, maps, word processor, TV (youtube), transportation (Waymo) everything.

Google is on almost everybody else's webpages too, through Google Ads.

There isn't much about your digital life google doesn't know about, and almost every potentially productive use of AI can be deployed to billions through their own services.

The other launch site cannot reach the space station with either soyuz or progress.

Every week I read about yet another GOP-controlled state legislature redrawing the boundaries to eliminate districts that might vote for a party other than the GOP in time for the midterms. Indiana is the latest state to do this. I don't see AI overcoming this. It's truly interesting that this widespread, coordinated effort to eliminate democratic seats with a map is hardly a blip in the news media. They are talking about the dems taking back congressional power in the midterms but frankly it's not going to happen.

Several democrat states are responding in kind now, although in their cases they are moving forward with a veneer of legitimacy through referendums. The GOP states could certainly do this but for whatever reason they choose not to. Something about democracy doesn't sit well with them even when it favors them. Still with democrat states gerrymandering away a few GOP seats, it's not enough to counter what these other GOP states are doing in my opinion.

Sadly the cancer of gerrymandering is working its way north. It's already been done here in Alberta to try to weaken the power of the cities vs the conservative rural base.

Right, what the free market wants to do is levelize our standard of living with our low-cost competitors, or import all the chips from them (with the security and supply risks that entails).

Simply shaming Intel for seeking government handouts does not solve our problem - how to maintain a domestic industry including internal competition rather than government choosing the winners and subsidizing incompetence.

From here in my comfortable chair it's hard to judge how bad the situation is, vs. to what extent it might be a form of protest by somebody who just doesn't like self-driving cars. There has been vandalism and harassment of a few types, from setting them on fire to calling dozens of them to the same place at the same time to cause gridlock. In San Francisco there was a huge flap because a waymo ran over a cat.

OK, it could be argued the government is the problem in the first place, since laws are a big part of why production here is economically nonviable. The problem is how specifically to solve that? Each law is there for a reason. It's easy to dismiss regulation broadly but harder in each given case.

If the US as a whole were a good place for this, a happy market solution would be for Intel to be eaten alive by another American competitor until either regains its competency or goes away. But surely you can see the national security risks of the more likely outcome - our supply depending on potential adversaries, including all the chips in critical infrastructure and defense hardware.

That is the basic problem, they don't. Mass-producing semiconductors in the USA without subsidies is not economically viable. The CHIPS act (or equivalently, tariffs) is an effort to tip the financial scales in favor of maintaining domestic production, for national security (so we can't be "cut off" by other nations). But such a scheme will not work if it is not executed in a financially predictable manner.

The question is whether the ads will be separate, like banner ads (albeit probably informed by your discussions) or woven into the responses themselves, whcih would be bad bad bad.

That's just using AI as a better search engine. Nobody is feeling compelled to go with its suggestions.