Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jstanley
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
33 ms
·
631.
▲
by
jstanley
1y ago
But you are inside the universe. The mathematical structure of the universe can contain all of your perceptions and to you they will "feel real". Indeed, to you they are real. If you had a perfectly accurate universe simulatio
632.
▲
by
jstanley
1y ago
> In other words, your property is more valuable, to you. That's fine, but then it becomes a property tax rather than a land value tax.
633.
▲
by
jstanley
1y ago
> What are you suggesting as an alternative? No mandatory acceptance of offers, land valuation by fiat.
634.
▲
by
jstanley
1y ago
I agree with everything you said up to "Nothing in those definitions is capable of expressing the concept of having perceptions (consciousness)". Do you think the universe is not computable? If you think the universe is computable
635.
▲
by
jstanley
1y ago
> it's the thing receiving the output not just an evaluation of a mathematical function. how do you know it's not just an evaluation of a mathematical function?
636.
▲
by
jstanley
1y ago
I'm not saying all algorithms are conscious. I'm saying I don't think it's obvious that algorithms can not be conscious.
637.
▲
by
jstanley
1y ago
I also read comments like that and wonder if other people aren't conscious. I don't know how anyone who experiences consciousness could be confused about what it means to be conscious, or (in other threads, not this one) could arg
638.
▲
by
jstanley
1y ago
You can't tell if another being is conscious without being on the inside of it.
639.
▲
by
jstanley
1y ago
I'm getting: > Error code: SSL_ERROR_ACCESS_DENIED_ALERT from Firefox, which I don't recall ever seeing before.
640.
▲
by
jstanley
1y ago
The point of the system where you have to accept an offer at your supposed valuation is that it forces accurate valuations. If you now say that we need to provide two different valuations (land + structure) but we're only taxed on one,
641.
▲
by
jstanley
1y ago
I think you may be confused about how LLMs work. Editing the history you send to the API works perfectly fine. You don't need the LLM to specifically "re-evaluate" the history from the point you edited. In a way it is "r
642.
▲
by
jstanley
1y ago
But just because my land is worth £100k doesn't mean I should be forced to sell the house that is on it for only £100k! If you must have that sort of system, there probably ought to be an obligation on the buyer to first pay to have th
643.
▲
by
jstanley
1y ago
They work differently, so the failure modes are different.
644.
▲
by
jstanley
1y ago
By that standard, humans aren't generally intelligent because you're still not economically redundant?
645.
▲
by
jstanley
1y ago
You also need to simulate load on the engine even in neutral. With no load, the speed genuinely would rise to red line at a rate controlled by throttle position.
646.
▲
The British Sundial Society
(sundialsoc.org.uk)
1 points
by
jstanley
1y ago
|
0 comments
647.
▲
by
jstanley
1y ago
For your specific case: maybe 1 in 10000. Or 1 in 5000 because "11230" would fit just as well. So about 13 bits of improbability. But your hypothesis took a lot more than 13 bits to encode! It would be better compressed to say &qu
648.
▲
by
jstanley
1y ago
Why is that a bad answer?
649.
▲
by
jstanley
1y ago
If you think writing more words will be more persuasive, just... write more words? The test already incentivises being persuasive! If writing more words would do that, and the incentivised human persuaders don't write more words and th
650.
▲
by
jstanley
1y ago
You don't want to ask AI to directly manipulate 360 distinct data points for the same reason you wouldn't ask a human to do that. Get it to write a script to process the data instead.
651.
▲
by
jstanley
1y ago
It sounds to me like it's the same sort of thing as protobuf.
652.
▲
by
jstanley
1y ago
Nah, screen's main use case is as an ad-hoc method to daemonise random scripts.
653.
▲
by
jstanley
1y ago
You mean after saying something stupid or wrong.
654.
▲
by
jstanley
1y ago
Have you tried Go?
655.
▲
Orienting Toward Wizard Power
(lesswrong.com)
6 points
by
jstanley
1y ago
|
2 comments
656.
▲
by
jstanley
1y ago
It took me a while to understand this comment, because I imagined that scaling up a golf ball would involve creating new atoms, but what you said only makes sense if you are scaling up the individual atoms. What you're saying is that t
657.
▲
by
jstanley
1y ago
In "docs like code" the first noun is plural. Your examples were pluralising other words.
658.
▲
by
jstanley
1y ago
"law schools enter test"
659.
▲
by
jstanley
1y ago
I think you are misunderstanding them. I also frequently cite ChatGPT, as a way to accurately convey my source, not as a way to claim it as authoritative.
660.
▲
by
jstanley
1y ago
Ah, using the YouTube app is a blunder! If you use the web version in Firefox you get to block ads.
More ›