svara

10 ms·

121.
by svara 10mo ago
Should be on the lookout for major upcoming domestic news they're trying to bury.
122.
by svara 10mo ago
https://en.wikipedia.org/wiki/Vibecession
123.
by svara 10mo ago
Yeah, I was thinking about this wrong. I don't think it works. Edit: I think the strategy does work, but a little differently: if you withhold a block and someone else finds one while you do so, you can still publish yours and win a ra
124.
by svara 10mo ago
You can determine statistically whether you have found a block relatively early, and conversely whether other miners are unlikely to find one soon. So you can get a head start on the next block from the likely new head block you've fou
125.
by svara 10mo ago
That's a distinction without a difference. Worker ants have high individual fitness if their colony successfully reproduces because they pass their genes forward. In evolutionary theory this is made clear by using the term "inclus
126.
by svara 10mo ago
> This sentence assumes a certain degree of shared prosperity. I think this is increasingly an illusion. IMO, Social media tends to create filter bubbles which create illusions of shared prosperity I think it's exactly the other way
127.
by svara 10mo ago
But it does? What do you think it optimizes other than individual fitness? I think I understand the GP pretty well. Cheating, or defection in the language of evolutionary theory, is subject to frequency based selection, meaning it is strong
128.
by svara 10mo ago
Cooperation has been "invented" in evolution many times independently and is long term stable in many species. If your comment was true that fact wouldn't exist. We may consider the world we live in today competitive, but at
129.
by svara 10mo ago
Yes that was my point.
130.
by svara 10mo ago
As far as DB goes, I'm pretty sure it's mostly an issue of systemic technical and consequently social collapse. The system runs beyond its limits and consequently the culture collapses because the people inside learn they have no
131.
by svara 10mo ago
It's a big, systemically failing organization running way beyond its capacity. Failures rippling through and compounding in a tightly coupled rail network. If they weren't able to announce the train would stop at one station, why
132.
by svara 10mo ago
But mapping raw values to screen pixel brightness already entails an implicit transform, so arguably there is no such thing as an unprocessed photo (that you can look at). Conversely the output of standard transforms applied to a raw Bayer
133.
by svara 10mo ago
It seems to differ extremely by account or region? I have to say AI summaries work pretty well for me on Google and my overall subjective sense is the results have been getting better the past 1-2 years. Here's what I get for your exac
134.
by svara 10mo ago
It reminds me a bit of a very small scale version of a robotic harbor, which very much exists in many places across the world already. I once toured the one in Hamburg, which is highly recommended if you ever get the chance. Nothing else ha
135.
by svara 10mo ago
You picked an example to support your conclusion in mentioning QA jobs which typically don't require a PhD. There still very much are other jobs that do require a PhD so I don't see what the point is there. More fundamentally this
136.
by svara 10mo ago
He's making it sound that way, although he might plausibly deny that by claiming he just doesn't want to speculate publicly. Either way he must have known people would read it like you did when he wrote that; so we can safely assu
137.
by svara 10mo ago
The cynical view suffers from availability bias - it's easy for us to think of someone who sticks out through bad behavior, but somehow gets away with it, precisely because it is not normal. (1) But if you look at long timescales, it&#
138.
by svara 10mo ago
That can happen, sure, but the cancer's strategy is not a winning one - it dies along with the host. Again, I'm not arguing for some naive Panglossian view. Things can get pretty bad transiently. I just take exception at the cynic
139.
by svara 10mo ago
Evolution works on the level of the reproducing organism, i.e. the individual. Google group selection if you'd like to go down a deep rabbit hole but the upshot is, if pro-social behavior did not confer and individual advantage, the in
140.
by svara 10mo ago
> you can assume they'll tend towards evil. An unnecessarily cynical take. What this is implying is that, in the absence of any morals, evil provides a selective advantage. And yet, pro-social behavior has evolved many times indepen
141.
by svara 10mo ago
Might make sense to scale the load by following electricity supply/prices though? Staying that as a genuine question since I'm not sure how the math works out at that scale, you have to weigh that against hardware depreciation of
142.
by svara 10mo ago
Having people work on things that are at the limit of human understanding is an essential part of a modern educational system. For every professional string theorist, you get hundreds of people who were brought up in an academic system that
143.
by svara 10mo ago
A part of it is reproducing incorrect information in the training data as well. One area that I've found to be a great example of this is sports science. Depending on how you ask, you can get a response lifted from scientific literatur
144.
by svara 10mo ago
A lot of words to call me stupid ;) You seem to have put me in some convenient mental box of yours, I don't know which one.
145.
by svara 10mo ago
Take a look at the new experimental AI mode in Google scholar, it's going in the right direction. It might be true that a fundamental solution to this issue is not possible without a major breakthrough, but I'm sure you can get pr
146.
by svara 10mo ago
That's right - it does seem to have to do with trying to be helpful. One demo of this that reliably works for me: Write a draft of something and ask the LLM to find the errors. Correct the errors, repeat. It will never stop finding a l
147.
by svara 10mo ago
Yes, they'll probably not go away, but it's got to be possible to handle them better. Gemini (the app) has a "mitigation" feature where it tries to to Google searches to support its statements. That doesn't currentl
148.
by svara 10mo ago
In my experience, the best models are already nearly as good as you can be for a large fraction of what I personally use them for, which is basically as a more efficient search engine. The thing that would now make the biggest difference is
149.
by svara 11mo ago
... but not in deep learning or am I missing something important here?
150.
by svara 11mo ago
They sucked compared to Google.

More ›