Funnily enough the one thing agentic AI of current generation is definitionally not good for is in fact tasks that are rare or unique in nature.

I mean it is better than Japanese status quo by some measure.

bribe Trump for a done deal.

Win9x and Win2k (and the other NT descendants) are fundamentally different operating systems. In general, NT had a much more robust kernel, so system panics were and remain mainly hardware issues, or, particularly in the old days, dodgy drivers (which is just another form of hardware issue). I've seen plenty of panics on *nix systems and Windows systems, and I'd say probably 90-95% were all hardware failures, mainly RAM, but on a few occasions something wrong with the CPU itself or with other critical hardware like storage device hardware. There were quite a few very iffy IDE cards back in the day.

The other category of failure, various kinds of memory overruns, have all but disappeared now as memory management, both on the silicon and in kernels, have radically improved. So I'd say these are pretty much extinct, except maybe in some very edge cases, where I'd argue someone is disabling protections or breaking rules to eke out some imagined extra benefit.

It might actually be. Consider this thing:

https://vxtwitter.com/TaylorOg...

Now consider this but in glasses. In case of phone vs glasses as an interface for modern smartphone, glasses make little sense. Phone is good enough.

But when it comes to this sort of an agentic device, glasses actually provide additional value that phone doesn't. You can just look at things and ask the agent to do with what you're observing. Doing this with the phone is far less intuitive.

Thank you, I enjoy slaying religious idols when they're counter-productive to my species' well being.

One of them being that High Priesthood Of Climate Science is beyond critique and that observing reality and concluding that they're wrong is Heresy to Faith of Science.

As in reality, process of observing things and concluding that hypotheses raised by other people are wrong has a name: science, practiced by scientists. Note the lack of capitalization. It's a descriptor, not a name of religion people like you worship: Science, led by High Priesthood of Scientists, the only people with divine right to observe and interpret reality.

Globally. Planet greening is so massive, it's observable from space.

"Nature" isn't a goddess. It doesn't "keep up". It is.

See, this is the problem with faithful. They find observation of reality to be an insufficient standard of proof.

This problem is universal in all religions, as religious circuitry in humans is specifically intended to be able to rally the tribe against overwhelming odds. I.e. "you cannot reason people out of positions they didn't reason themselves into".

Yes, that is the argument of the climate zealot side. "But none of it matters because planet is going to die. We need to stop raping Mother Gaia and start worshiping her".

These are the same people who predicted in 2000 that we'd have a "global starvation event by 2020", as well as that Pacific island nations would be underwater and cease to exist by this time. And that were talking about "global boiling" in 2020.

Turns out reality doesn't bend that way. Regardless of strength of faith of the zealots.

Yes, because Russians are key partners when it comes to operating the ISS. They own and control about half of it after all.

It's in the name. International Space Station. Not American Space Station.

I would guess the point is that eventually games are available for more than just Windows. Because Microsoft has been trying for better part of decade to make their own "games store/app store" to compete with Valve. Down to making editions of OS where you couldn't even install software from outside Microsoft's own store. They're really trying to lock down the OS without losing most of the customers.

And while it has been failing miserably so far, the writing is on the wall and Valve would be smart to push software companies off Windows into "works on everything" paradigm.

Funny how if you look at the change log of proton, it's full of fixes for the "tricked-out crazy-straw-coded games". Which would suggest they're running anything but flawlessly.

https://github.com/ValveSoftwa...

Reminder: most of these papers cite numbers by which growth will be reduced by the end of the century. Projected total growth in the same time frame is several hundred percent. This is a factor that slightly reduces that number.

"Climate sceptics" generally argue that this sets a ceiling on cost of efforts to mitigate climate change, because otherwise cure is worse than the disease.

Again, the best evidence of someone arguing in bad faith is their pointed refusal to look up a basic fact. Examples:

Pictures of planet from space on "is earth round" argument.
Basic mathematics on "does 2+2=4" discussion.
Basic medical ethics on "do we halt studies and administer the cure if during testing we find overwhelming evidence that cure works".

You are pointedly refusing to do this. And then complaining about the dumbest grammar communist level of red herring I've seen on internet in a month. It's hard not to observe this weird behavior and not conclude you're acting in bad faith.

Do you know who else "does a catastrophically worse job than a human in specific conditions"?

Other humans.

Because they're also self-learning neural networks, and share the problem of "each distinct neural networks develops its distinct quirks as it learns" with current AI.