ursAxZA
Outside weight-class or aesthetics-driven sports, it’s hard to imagine any scenario where a GLP-1 analog creates a net advantage.
In endurance disciplines the binding constraint is almost always fuel throughput: if an athlete can’t take in and process enough calories, recovery and performance fall apart. Anything that suppresses appetite or slows gastric motility is basically disqualifying.
You can already see how narrow that margin is in the sheer amount of gels, bars, and mixes riders consume during long sessions. From that angle, GLP-1 simply doesn’t occupy the same decision space as substances that expand performance capacity or recovery bandwidth.
Dial-up was slow, but at least the internet still felt human.
Fiber gave us speed, not soul.
Sometimes I miss yelling “Corp Por” into the TUBE — back when the screen wasn’t a window, but a passage.
Wouldn’t using something like this trigger anti-doping concerns if an athlete took it?
In sports, manipulating appetite or insulin pathways sets off red flags immediately.
It’s interesting to see the food industry treat the same biological mechanism mainly as a market trend rather than a medical one.
Most people are never choosing between Being and Time and an HN thread. But if they were forced to choose, we already know which one would dominate sheer engagement.
That doesn’t mean HN replaces philosophy — it just means that attention has its own economics. And any medium that captures attention will inevitably show qualities (good and bad) that heavyweight works simply can’t compete with.
My Ryzen agrees — the fans just spun up like it’s hitting 10,000 rpm.
That’s why I offered one possible implementation as a hypothesis, not as a law of nature.
If you have a better non-ideal, real-world design in mind, I’d be interested to hear it — it makes the discussion much easier.
It made me wonder where a future goes when it keeps trying to define both barbarism and normalcy.
As a small tribute in return, three films came to mind:
Bicentennial Man,
Gattaca,
Fight Club.
I’ve always preferred Ivan the Fool — choosing to live, rather than living inside a definition.
A layered mix — county-level public utilities, some private operators, and some hybrid/municipal entities — might be closer to a resilient structure.
Not clean or elegant, but fault-tolerant.
At that point the debate isn’t about pricing — it’s about survival of the system.
I could outline the full methodology behind this view, but that would turn the thread into a private seminar — and that’s not what comment sections are for.
If the floor is gentle, people can actually reduce usage without feeling punished for doing the right thing.
At the moment the baseline tier feels… maybe a “C-rating” version of what a real baseline could be?
If we treat baseline access as a kind of ‘civilization tax,’ the pricing shouldn’t feel punitive for low-usage households.
If a household uses less than the monthly per-capita average, why not cap that baseline at something like $10?
Yes — that gap would need to be subsidized, probably through taxes. But that’s already how grid maintenance works: we socialize the fixed costs while pretending rates are purely volumetric.(and I might be overstating this slightly).
Right now we punish low-usage consumers and reward structural inefficiency. A baseline tier would at least make the incentives coherent.
If the issue has existed for years, retroactively jumping straight to fines feels less like regulation and more like the government exploiting its timing advantage.
If any one of these fails, the meaning collapses.
Algorithms tend to optimize us toward well-being as “well-done”: predictable, consistent, uniformly cooked. Safe, measurable, repeatable.
But human experience is closer to “rare”: uneven, risky, asymmetric, and still alive. The parts that matter most are often the ones that don’t fit cleanly into metrics.
If everything becomes optimized, nothing remains interesting. And more importantly, we risk replacing well-being with the monitoring of well-being.
When a life is constantly optimized, scored, nudged, and corrected, it gradually stops being a life that is actually experienced.
Benchmarks measure competence inside a predefined problem space, but real scientific and engineering work isn’t bounded — it keeps changing underneath you.
At some point we don’t just need a system that knows how to solve problems in theory; we need one that can actually do something with that ability.
The equivalent of making the coffee when we want coffee, not just getting a perfect score on a coffee-theory exam.
The discussion has been about CoT in LLMs, so I’ve been referring to the model in isolation from the start.
Here’s how I currently understand the structure of the thread (apologies if I’ve misread anything):
“Is CoT actually thinking?” (my earlier comment)
→ “Yes, it is thinking.”
→ “It might be thinking.”
→ “Under that analogy, self-training on its own CoT should work — but empirically it doesn’t.”
→ “Maybe it would work if you add external memory with human or automated filtering?”
without an external supervisor, whatever gets written into that memory is still the model’s own self-generated output — which brings us back to the original problem.
Starlink is often treated as the reference point not because it is perfect or fully resilient, but because there is no second network at a similar scale that could realistically serve as a failover today.
If we imagine a hypothetical future where three mature operators exist, then yes — absent coordinated political or geopolitical action, at least one network might remain online.
However, even that surviving operator would not necessarily provide full coverage of the affected region. Global redundancy is extremely hard in practice, because maintaining continuous, worldwide LEO coverage is not free — it requires massive capex and opex, ground stations, regulatory permissions, and local political approval.
True worldwide failover remains more of a theoretical construct than an operational reality.
Neither option is risk-free; the failure modes simply differ.
A government can shut you off for political reasons, a corporation can shut you off for contractual or geopolitical ones.
As long as the system assumes centralized stewardship for safety or reliability, someone will inevitably hold the switch — the only variable is who.
Humans can refine internal models from their own verbalised thoughts; LLMs cannot.
Self-generated text is not an input-strengthening signal for current architectures.
Training on a model’s own outputs produces distributional drift and mode collapse, not refinement.
Equating CoT with “inner speech” implicitly assumes a safe self-training loop that today’s systems simply don’t have.
CoT is a prompted, supervised artifact — not an introspective substrate.
It only proves success under one set of constraints — not the ability to survive when the landscape itself shifts.
Of course, once jamming enters the picture, even that lifeline disappears.
As far as I understand it, it’s a generated narration conditioned by the prompt, not direct access to internal reasoning.
When a society becomes fully efficient, people start craving the slow, the physical, the intentional.
Still, if we ever end up with rows of these giant “balloons,” the landscape might look unexpectedly futuristic.