To summarize report, there's not anything you can do to avoid old age dementia because its causes are not understood.
The biggest attendant factors are vision or hearing loss. Note that these factors are merely attendant, not stated as causative.
There are no significant pharma treatments.
The "surprise" finding was that the examination regime produced improved cognitive "scores" in both the mediated group and the control group. And these improvements persisted until death. In other words, the study is a complete mess.
Takeaway: the mind is going to go along with the body.
The Googlez say that Steve Jobs rejected the use of on Apple keyboards as a "overuse" of the brand, which led the use of ⌘ (place of interest, St. John's arms, Bowen knot) for the "command" key.
Generally typefaces are not subject to copyright.
The proposition that Apple can impose its discretion on others' use of unicode is unreasonable by the intention, definition and nature of the code.
I find that the more thought I give this, implications of such constructs expand far beyond this particular use case.
For example, consider other other brands which are a re-use a character, X being the most notable.
At what specific points in digital systems does a number become regarded as an IP asset? Whose choice is it to regard a number as IP (when is a number just a number)?
The questions continue to unfold with further thought.
When the author of a programming how-to on using magic beans to create apps feels he must inform the student that prompting "fix my app" can work better with some situation context, then even the pretense "software engineering" is dead.
What's even more dumbfounding is that the magic beans may actually just "fix the app".
Why doesn't this kind of article strike terror in the heart of the industry?
It's a canary-in-a-coal-mine and the bird has died.
Light from panel is diffuse; it can't project rays, so will not look like daylight.
Perspective of view always totally wrong.
As screens age they'll drift or fail and not match each other, looking very fake.
No actual window is a channel for the view outside, so once placed, no one need ever further decide the view. The simple possibility of changing channels is destructive to the idea of a window.
But other than totally destroying the idea of reality, fake windows will be very "realistic."
Beehive offered as any analogy for a new media paradigm of otherwise unspecified structure
> A hexagon is the most efficient shape in nature because it maximizes area while minimizing perimeter.
The hexagon is not the most efficient shape for area to perimeter, it's the circle.
Honeycomb is not hexagonal, it merely appears that way by virtue of packing uniformly sized cylinders.
As an empty honeycomb analogy is the only idea offered for a new "Popular Front" media— noting a thousand words submitted to summarize some approach— and that the claim of a hexagon as proffered is simply wrong, this article reads like someone announcing yet another way to milk some millionaire rubes from their money in a vein like The Lincoln Project, which are a forlorn sector of right-wing ideologues, alienated due to the rise of MAGA, but still reflexively classist and Republican.
Maybe Substack can't figure out any other way to grow revenue than become expressly political propagandists?
The idea of a languages being separable to the brain is a misunderstanding: The root of language is a common biological trait that works irrespective of the cultural situation.
Every major AI service claims be-all end-all capability, yet cognoscenti run merry-go-rounds for burgeoning specialist offerings.
Richard Dawkins is hoping for a Dating Game of AI GFs:
BachelorAItte #1, if a man told you he'd love to prompt you at your backdoor, would you A: Invite him to continue with a compliment? B: Refer him to Claude? C: Beg him to reset you?
Wish I could run my life like this: I'm adding value by blowing other people's money and acting like I'm rich.
"Ok so here's the deal, you front my incredible lavish lifestyle and eventually I'm gonna run into managers who I will talk into giving me money that belongs to yet other people, and then I'll pay you back with a percentage and the managers will take a commission and those yet other people will ride out the market volatility. The P&Ls can be steered to make this work for you and me with the support of the network. All we need is to get those managers on board. Trust me!"
And we've seen this before.
2008: Really Smart Guys banks get help of ratings services to gin up managers who tap out the network, until the RSG banks blow a derivatives gasket and Fed has to construct several-many-trillions to make them whole (a lamb was sacrificed to the god of socialism) while the little guys at the edge of the network got pushed off a volatility cliff. "Past performance is not a guarantee of future returns"
"Systemic failure; no one is responsible; RSG bonuses will be paid, carry on"
Reset button gets pushed.
Clown on NBC who told everyone to buy-buy-buy the bull at the moment it croaked in 2008 keeps his job telling everyone to buy the bear.
20 years later: Fantastical new instruments even more perverse than derivatives are on the hoof and RSGs enjoy insane largesse while nation is hollowed out.
RSGs are sometimes bothered by a vague sense of shame and guilt that their lives are bullshit and they have lots of stuff but no community— it's troubling— but too unspecific to merit more than an angsty blog post on how SF just doesn't feel right anymore.
With luck the SA fault will rip a 9.0+ soon and all that angst will roll into the ocean.
From the paper:
//
"Nobody is going to run 100 meters in five seconds, no matter how much is invested in training and machines. The same can be said about using the brain. The human mind is no different now from what it was five thousand years ago. And when it comes to mathematics, you must realize that this is the human mind at an extreme limit of its capacity."
//
Previously on HN:
Shannon Got AI This Far. Kolmogorov Shows Where It Stops
Look at Activity Monitor report when it's overloaded to see what's busy.
MacOS and apps have many kinds of background activity which present large transient loads, for example Spotlight indexing and photo library facial scanning.
Do you have a browser or other app infected with a crypto-miner or other malware?
If you create a new user and sign in using only that new user, do you see the same overload?
Are you running an unsupported Mac for Tahoe via OCLP? Thar be performance hazards. Search Macrumors "unsupported" forums.
There's a far greater hazard than AI boosting itself:
As models are under direct control of totalitarian silos, ideology can be baked into every tendril of knowledge represented.
A service such as Gemini may be a societal ideological risk on a greater scale than the notoriously fascist influence of radio/TV in the 1920s and 30s.
"Guardrails" are direct lines of injection of ideology, and
"hallucinations" can serve to mask.
Why should anyone trust singular corporate control points for doctrine on the totality of knowledge? It seems like a crazy risk.
News is already utterly polluted with doctrinaire bias.
Politics are clearly designed to keep the electorate away from decision-making, and open doors to privileged silos of control.
AI can own the populace coming and going as the models inject doctrine into the core of knowledge and scour the consumer for measures doctrinal uptake and alignment. Will information become totalized with no possibility of escape?
The biggest attendant factors are vision or hearing loss. Note that these factors are merely attendant, not stated as causative.
There are no significant pharma treatments.
The "surprise" finding was that the examination regime produced improved cognitive "scores" in both the mediated group and the control group. And these improvements persisted until death. In other words, the study is a complete mess.
Takeaway: the mind is going to go along with the body.