KolibriFly

6 ms·

61.
by KolibriFly 6mo ago
If a device is uncomfortable in daily use, modifying it to fit your body actually makes a lot of sense
62.
by KolibriFly 7mo ago
So maybe the framing is: LLMs are good at mapping the landscape, but not at discovering new continents
63.
by KolibriFly 7mo ago
Yes, not all "writing" is actually thinking, and a lot of what we call writing at work is really just ritualized context transfer
64.
by KolibriFly 7mo ago
A lot of the value is in discovering the holes in your own reasoning while trying to make it legible to someone else
65.
by KolibriFly 7mo ago
Today it's mostly "all-or-nothing" at the panel level, but under the hood there's already a lot of cleverness trying to approximate the behavior you're describing
66.
by KolibriFly 7mo ago
What LG is pitching here is basically bringing that 1Hz floor capability to large laptop panels
67.
by KolibriFly 7mo ago
What these variable refresh panels are trying to do is kind of the "best of both worlds"
68.
by KolibriFly 7mo ago
What's interesting about these newer 1Hz claims is that they're basically trying to sidestep the exact problems you mention
69.
by KolibriFly 7mo ago
Sure dropping toward 1Hz could be huge. But the moment you scroll, watch video, or even have subtle UI animations, you're back in higher refresh territory
70.
by KolibriFly 7mo ago
That's a really useful historical reminder
71.
by KolibriFly 7mo ago
Yet I'm not sure it comes from more localized decision-making. It might actually come from making the rules clearer and less discretionary
72.
by KolibriFly 7mo ago
Austin's example shows what happens when a city actually follows through and allows density where it makes sense
73.
by KolibriFly 7mo ago
Yet Austin didn't just "build more housing" in a vacuum, they also changed how and where you can build
74.
by KolibriFly 7mo ago
That feels like the real takeaway: most housing shortages are caused by a pile of small constraints, so you need to remove a pile of small constraints
75.
by KolibriFly 8mo ago
Momo is. Yes, she is ahahah
76.
by KolibriFly 8mo ago
That episode aged way better than expected
77.
by KolibriFly 8mo ago
Also makes me wonder if we'll start seeing tools designed specifically to be LLM-legible
78.
by KolibriFly 8mo ago
The human built the system, the AI did the implementation and the dog provided the "intent" even if that intent was just treat seeking randomness. It turns software creation into something that looks less like writing and more lik
79.
by KolibriFly 8mo ago
Overall it feels less like "dog makes games" and more like a glimpse of what happens when you wrap modern models in the right scaffolding
80.
by KolibriFly 8mo ago
To be fair, that's basically a variation of techniques that have existed long before Bluetooth
81.
by KolibriFly 8mo ago
Even when they claim it's "anonymous," the value is in aggregate behavioral patterns: dwell time, repeat visits, path through the space, etc.
82.
by KolibriFly 8mo ago
Bluetooth, Wi-Fi, even things like tire pressure sensors... they were designed primarily for convenience and interoperability, not adversarial environments. Now we're retrofitting privacy onto systems that were never really built with
83.
by KolibriFly 8mo ago
I get the frustration behind that sentiment, but I'm not convinced it's as simple as "scrolling cooked brains"
84.
by KolibriFly 8mo ago
The Commission isn't (at least from what's published) saying "pagination must look like X" or "buttons must be Y pixels tall." They're saying platforms have to mitigate systemic risks tied to their design
85.
by KolibriFly 8mo ago
The tricky part is defining harm without sliding into paternalism
86.
by KolibriFly 8mo ago
The Commission seems to be saying: not without guardrails
87.
by KolibriFly 8mo ago
Infinite scroll itself isn't inherently harmful, it's a pagination mechanism. The harm comes from recommender systems tuned for engagement over wellbeing
88.
by KolibriFly 8mo ago
This era might be one where we have to earn the next clue much more slowly
89.
by KolibriFly 8mo ago
So, if the answer were obvious or quick, it wouldn't be worth building machines that take decades to design
90.
by KolibriFly 8mo ago
This feels less like a story about particle physics "failing" and more like a story about a field running out of easy leverage

More ›