Wasn't Freeman Dyson at least skeptical of ever bigger and better particle accelerators as reaching diminishing returns on the amount of physics knowledge gleened per dollar spent.

Forget Freeman Dyson. Hasn't anyone taken ECON 101 as a college freshman and remember "diminishing marginal returns." And forget the harm to the environment, isn't the outragious electric bill a sign of more and more resources thrown at something to "scale it up" without considering where the scaling law levels off?

What is the revenue model? Selling what you disclose to the AI?

Or will anything beyond the most brain-dead AI be a big monthly subscription?

Will your employer insist that you not use their paid-for AI for personal use in the way of Cyber Monday that you weren't supposed to use your work Internet to purchase your Christmas presents but people did this anyway?

Or will AI gradually become useless owing to who pays the most coin to train the neural networks a certain way, becoming useless like Web search?

That's not how it works. You can't just buy a tract of land. You have to buy the land, jump through hoops to get permits to build, defeat NIMBY lawsuits, get the local municipality to run services, defeat more NIMBY lawsuits, get new permits from a new municipality administration, then finally break ground on construction.

NIMBYs of all stripes throw up roadblocks during the permitting process and will then sue to get an injunction. If you defeat those lawsuits they'll go after the municipalities suing that permits were approved illegally (which they sometimes are).

After several rounds of legal wrangling you can finally start construction. This ends up with only huge developers being able to build because they can absorb all the pre-construction costs until they sell all the homes. They can also afford the scale to benefit from bulk orders and contracts.

The cost of expanding housing is mostly in the project development rather than labor costs of construction. Even if you had construction robots the construction companies would only offer marginal savings as their bids would just be human contractor - 1%. There would be no reason for them to leave money on the table offering a huge discount off human labor.

The product page mentions macOS Tahoe in several places. That should have been a huge giant obvious clue.

The Heliocentric Theory had a problem with a negative result for the parallax of stars. Think of it as the Michelson-Morley result of its day.

This is why Tycho Brahe rejected the Copernican interpretation, the best observational astronomer of his day--he couldn't measure any shift in stellar positions as the Earth went around the Sun. No one had the imagination that the nearest stars were so freakin' far away.

There are other large horse body parts.

I started using Google Gemini to evaluate applications of transfer students for credit for specified courses at my university.

What a person in my role is supposed to do is email faculty colleagues teaching the course in question, but good luck with that. A faculty member wants such requests to just go away. Even if you know the contents of a course you have taught for years, how do you know that Cow College's offering is anywhere close?

The AI not only has a lot to say on how equivalent a pair of courses, if it is just making stuff up, I don't see how it can be worse than what my colleagues say.

Maybe if I don't call attention to what I am doing, no one will notice the difference. Maybe it is an improvement. A grad student turned in a form and left out which courses they wanted credit for. It came to me but requesting an undergrad-level course, and this topic is in the "wheelhouse" of the student's PhD advisor who approved. I sent it back that my "research" indicates that the course taken at MIT is advanced-graduate level at our Cow College and suggested an advanced grad level course here that according to the AI is a "great match."

I sent this back to the student--is the student going to complain?

It could happen.

Not defending HP's ludicrous rental scheme but if you're wanting to play with local LLMs the mobile chip in the laptop is going to be underpowered compared to the full sized chip in the Hetzner server. Despite the name the chip in the laptop will have far fewer compute cores and a fraction of the VRAM. It will also have more thermal throttling. If you want to play with LLMs you'd be better off just doing cloud instances with inference providers. They're often cheaper than a dedicated monthly server.

Not recognizing each other at the liquor store and where these gadgets are sold?

But Dr. Tongue and Count Floyd made such effective use of 3D in really scary movies?

Is the mix to increase potency or is it for apparent potency.

Don't know about farming, but the homeowner squirts glyphosate and nothing happens whereas Roundup causes the weeds to turn brown right away.

Is this about effectiveness or is it about lack of patience for the glyphosate to kick in.

I hope they have access to enough eggplant.

At some point are they going to explode like those pagers?

1SAP to find them,

1SAP to bring them all and in the darkness bind them.