Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
allears
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
13 ms
·
271.
▲
by
allears
1y ago
All the chatter about subscriber numbers distracts from the fact that they're burning way too much capital to ever make a profit.
272.
▲
by
allears
1y ago
If test fails, resume execution at next line.
273.
▲
by
allears
1y ago
Otherwise known as $240 a year. For a browser? When there's so many ways to access AI for free?
274.
▲
by
allears
1y ago
A misspelling in a headline like this doesn't make me want to click...
275.
▲
by
allears
1y ago
Generally, I tend to eat natural foods. I have for decades. They just taste better. Dyes are mostly used in processed foods, because otherwise they would look unappealing next to fresh natural food. And for a very good reason. All that is t
276.
▲
by
allears
1y ago
I agree, no sarcasm intended. It's bland and uninformative, with no evidence of actual thought behind it.
277.
▲
by
allears
1y ago
So what would a 'hallucination' be in this context? An order for a half ton of toilet paper? If I know ChatGPT gets things wrong, why would I trust it to shop for me?
278.
▲
by
allears
1y ago
Article headline says "ultrasonic" (frequencies too high for human perception, such as bat chirps) but article describes very low-frequency sound, sometimes called "infrasonic." Strange error for a supposedly tech-orient
279.
▲
by
allears
1y ago
The kids are all right
280.
▲
by
allears
1y ago
Nope. I don't want my puzzles to dumb themselves down for me. The whole point of a puzzle for me is the challenge. If they're too easy they're no fun.
281.
▲
by
allears
1y ago
That dog won't hunt. Competing with China may be important if you're an executive, but if you're a worker, not so much. If you want some real incentive for workers, try sharing some profit with them. And by some, I mean a lot
282.
▲
by
allears
1y ago
Proving beyond a doubt that the purpose of the tickets is not to provide a clear path for the streetsweepers
283.
▲
by
allears
1y ago
This, unfortunately, is a political issue, and currently, facts have little bearing on the outcome.
284.
▲
by
allears
1y ago
An organization of truly advanced thinkers. This is the perspective on AI that has been sorely missing.
285.
▲
by
allears
1y ago
I've been using the Aura frame. Seems to tick off all the boxes you mentioned, and it's worked seamlessly for me so far.
286.
▲
by
allears
1y ago
Not sure what effect that's supposed to cause -- I stared at the cross in the middle while a bunch of faces flashed on either side. Most (all?) of them looked distorted, but I didn't look directly at them (as instructed) so I'
287.
▲
by
allears
1y ago
So-called "hallucinations" aren't a threat to AI, and they're not hallucinations. That's a marketing term meant to anthropomorphize a statistical machine. They're a mathematical certainty. AI software is simply
288.
▲
by
allears
1y ago
I have the same model, and similar complaints. I reduced the transparency (Settings-Accessibility) and that helped, but the needless animations and other cutesy touches are annoying and distracting. I turned off the call screening, because
289.
▲
by
allears
1y ago
This was a whole lot more fun back in the 60s with Timothy Leary.
290.
▲
by
allears
1y ago
It's the job of an AI company's lawyers to make sure nobody is responsible. It's a tough job, and no doubt well paid.
291.
▲
by
allears
1y ago
Quietly acknowledging the elephant in the room, while loudly proclaiming an elephant-free zone, isn't an admission. It's just more misdirection. "Mathematically?" As in, don't worry your pretty little head about it?
292.
▲
by
allears
1y ago
Cuz it's easier to get a table?
293.
▲
by
allears
1y ago
"Only for the thoughts you want to share?" Article doesn't specify how that mechanism works. I have enough problems with blurting out dumb things aloud without thinking, I doubt I'd have much luck controlling subvocaliza
294.
▲
by
allears
1y ago
I've always avoided it myself. Lack of privacy, with new, additional lack of privacy!
295.
▲
by
allears
1y ago
Pop science indeed. Nothing new here. The Turing Test was the product of a much earlier era. Our machines today can easily fake a conversation, but there's been little progress in defining what intelligence is, let alone consciousness.
296.
▲
by
allears
1y ago
YouTube influencer and professional skeptic. Her shtick is to 'debunk' various aspects of the scientific establishment. There's usually some truth mixed into her rants, but it's hard to tell how much and where.
297.
▲
by
allears
1y ago
Bravo! I thought I was about to read a review of yet another motivational app, but I couldn't agree with you more. Competition and striving for goals are highly overrated in our society, IMHO. Simple enjoyment of a pleasurable activity
298.
▲
by
allears
1y ago
You're right and I bit. I'll try to hold back next time. And there will be a next time.
299.
▲
by
allears
1y ago
He's right, but since he's Gary Marcus, he's totally biased, so I don't bother to read his rants.
300.
▲
by
allears
1y ago
I've been seeing endless posts on HN about this product -- the sheer number of posts says "scam" to me
More ›