Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
TimByte
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
91.
▲
by
TimByte
7mo ago
Europe finally realizing that payments infrastructure is geopolitical infrastructure
92.
▲
by
TimByte
7mo ago
In practice I mostly ignore ads too but it feels like an ecosystem-level tragedy
93.
▲
by
TimByte
7mo ago
I think a lot of people who grew up on the early web have that reflex
94.
▲
by
TimByte
7mo ago
The depressing part is that this is probably working just well enough financially
95.
▲
by
TimByte
8mo ago
HP Prime emulator still wins for actually solving equations
96.
▲
by
TimByte
8mo ago
Yeah, that would've been the cleanest experiment
97.
▲
by
TimByte
8mo ago
The scary part isn't "LLMs doing sums." It's that the same deterministic model, same weights, same prompt, same OS, produces different floating-point tensors on different devices
98.
▲
by
TimByte
8mo ago
From a debugging point of view, the author's conclusion was still completely reasonable given the evidence they had
99.
▲
by
TimByte
8mo ago
The real lesson here isn't even about Apple. It's about debugging culture
100.
▲
by
TimByte
8mo ago
And once a system starts making probabilistic guesses about who you are, the burden flips onto the user to disprove it
101.
▲
by
TimByte
8mo ago
Yep, there's a real trust gap when it comes to handing over biometric data
102.
▲
by
TimByte
8mo ago
The selfie re-verification escape hatch helps, but it also quietly normalizes "prove you're an adult to get full functionality," which is a pretty big shift from how most software has worked historically
103.
▲
by
TimByte
8mo ago
AI can help with synthesis once those insights exist, but it doesn't naturally occupy that liminal space between groups, or sense the cultural and organizational gaps
104.
▲
by
TimByte
8mo ago
A good tech writer knows why something matters in context: who is using this under time pressure, what they're afraid of breaking, what happens if they get it wrong
105.
▲
by
TimByte
8mo ago
In my experience, great tech writers quietly function as a kind of usability radar. They're often the first people to notice that a workflow is confusing
106.
▲
by
TimByte
8mo ago
The hard part is the slow, human work of noticing confusion, earning trust, asking the right follow-up questions, and realizing that what users say they need and what they actually struggle with are often different things
107.
▲
by
TimByte
8mo ago
The failure mode isn't just hallucinations, it's the absence of judgment: what not to document, what to warn about, what's still unstable, what users will actually misunderstand
108.
▲
by
TimByte
8mo ago
Just people choosing to be slightly more attentive than they strictly need to be
109.
▲
by
TimByte
8mo ago
Now interruption is the default state, and attention is already fully saturated before another human even enters the picture
110.
▲
by
TimByte
8mo ago
The hitchhiking framing is powerful because it strips things down to trust and openness
111.
▲
by
TimByte
9mo ago
LLMs didn't get good because text is flashy; they got good because text is dense with intention
112.
▲
by
TimByte
9mo ago
Yet you don't need special tools, schemas, or viewers to get some understanding out of it
113.
▲
by
TimByte
9mo ago
I think some of the binary tooling exists less because engineers hate strings and more because humans aren't the primary consumers anymore
114.
▲
by
TimByte
9mo ago
Where I keep coming back, though, is that text still seems to be the backbone that lets those richer systems scale and persist
115.
▲
by
TimByte
9mo ago
Text feels boring only until you notice how much work it quietly does better than everything else... And pragmatically: text respects time and bandwidth
116.
▲
by
TimByte
9mo ago
Text has always been treated differently than images or video, partly for historical reasons and partly because regulating it runs straight into classic First Amendment landmines
117.
▲
by
TimByte
9mo ago
My guess is that Utah\Louisiana will either pause, copy-paste the same approach and hit the same wall
118.
▲
by
TimByte
9mo ago
Soo the strongest form of privacy protection isn't better storage or better policies, it's simply not creating the data in the first place
119.
▲
by
TimByte
9mo ago
Age gates at the App Store level aren't a narrow restriction, they're a universal checkpoint
120.
▲
by
TimByte
9mo ago
What also gets glossed over is the privacy tradeoff: to "protect minors," you end up collecting more sensitive data about everyone, including adults downloading trivial apps
More ›