Wiki Contributions


The AGI needs to be honest

You have asked the bot to summarize the proof for you in natural-language, but you know that the bot can easily trick you into accepting the proof.

I didn't fully get this point. Sure, there definitely are gigantic computer-generated proofs out there, the most famous being the proof of Four Color Theorem (which some mathematicians in the 80s rejected precisely because it was infeasible to check by hand); I accept that our bot could produce a truly huge blob.

On the other hand, optimizing a bogus proof for human validation sounds even more difficult than just search for a correct proof. To search for correct proofs you just need to know math; to search for plausible bogus proof you need to know math and how human mathematicians typically validate proofs. Are we just assuming to work with AGI here?

Without a phone for 10 days

My experience is quite similar; never bought a smartphone and use just a Nokia phone with voice/SMS only. It's not exactly like not having a phone at all, but in a typical day I don't receive more than one message/call, and more often than not the phone remains completely silent for the whole day. I've watched friends managing their WhatsApp chats (and similar time-sucking services) and I'm still very much scared by the perspective of constantly being pinged for random reasons. With my old-fashioned phone I have to pay some cents for every message sent. I could change my tariff plan at any time, but so far I've choose to stick with the old tariff, because it's a very strong incentive to send only messages that actually matter.

That said, I usually spend several hours a day working at my laptop and don't travel very often. Even with a smartphone at hand, I would still prefer to work from my laptop if available (much bigger screen, no weight in my hand etc). Also, my colleagues are accustomed to emails and my parents are sort of smartphone-skepticals themselves; putting all toghether, my social pressure for getting into smartphones is pretty low. But it's definitely possible to live without and I encourage to give it a try.

Is Legal Reasoning Fundamentally Irrational?

The legal system must above all be predictable, and you'll have a very hard time being predictable if your legal rules are written like "citizens should do X unless they have good reasons to do otherwise". At the very least, those good reasons to do otherwise should be elicited explicitly (but they can, and do, change over time).

Take speed limits, for example. Probably we can all agree that driving at 100 mph in city centre is a terrible idea, so the law forbids it in basically every nation (elicited exception: you're a police officer during an high speed chase). But the actual speed limits probably won't let you drive at 100 mph anywhere, not even if you are the only driver along a giant straight road in Nebraska with perfect weather. If you still respect the official speed limits in these conditions, probably is because you don't want to get a ticket from the police (or because you don't trust too much your own driving abilities, but let's just ignore this case for now). Legal positivism applied to speed limits is John Nestor, and I would guess that even the large majority of legal officers don't drive like Nestor, or he wouldn't have one-third of his Wikipedia entry dedicated to this. Even in this case, I think that letting an occasional Nestor cause traffic jam is still a better idea than just writing down "drive at whatever speed you judge appropriate".

On a more general level, I think that a simple fear to get punishment from the upper level of the bureaucracy should be enough to justify the simplest form of legal positivism. Stick to the letter of the law, and no one could formally accuse you of any infraction.

What can we learn from traditional societies?

Diamond discusses the benefits of preserving different languages (most are becoming quickly extinct), such as some evidence that they can protect against Alzheimer, that they are cultural treasures, etc. but I remained unconvinced by these arguments. In fact, I absolutely love that there are many languages in the world and I hope that this continues to be the case, but I cannot find a better argument for it than a simple “I like languages” (which I think is also Diamond’s real reason and the rest is rationalization).

The "cultural treasures" part may have a point. Many forms of poetry (and witty puns) make sense only in their native language, and translations lead to dull results at best. For example, I'm absolutely not familiar with Haiku, but I suppose they must be a nightmare to translate. 

And if we talk about the huge, revered rhymed poems from the past centuries, ensuring that someone will be always around to read the originals seems quite a no-brainer to me. If the whole world switched to English only, we'll lose forever the ability to read the Divine Comedy (and many other works) taking pleasure directly from the carefully selected rhymes. I acknowledge that some of these translations are actually quite good, and especially this one by Charles Tomlinson, who did his best to preserve the original metrics, but I'm still unconvinced that a single language would be fine for this (or maybe I'm just particularly enthusiastic about strictly metrical poetry).

2021 Darwin Game - Human Garbage Dump

So far this is the only biome with non-invincible winners. I suspect that having just one large primary source of forage (like previous biomes) pushes too much towards "maximal-efficient invincible foragers eating this". Now I'm really curious to see how it went in Grassland-Rainforest-Temperate Forest.

2021 Darwin Game - Ocean

I agree, most winners so far are Armor 10 + Antivenom (the cheapest way to become invincible).

2021 Darwin Game - Contestants

I meant "restrict to both designs having Weapons + Armor >= 10" (which I admit may be a moot point since Weapons + Armor > 10 is out of the Pareto frontier anyway).

Weapons 6 + Armor 8 is obviously strictly worse than Weapons 6 + Armor 4 (cost less and do the same), but it's even worse than Weapons 9 + Armor 4 (same cost and do more). And Weapons 9 + Armor 4 is strictly worse than Weapons 10 alone.

2021 Darwin Game - Contestants

Right, maybe it's not always better. But if we restrict to "invincible" predators (Weapons + Armor >= 10), it is, since no one can prey upon you anyway. I've edited the previous comment.

2021 Darwin Game - Contestants

Speaking of useless adaptations, Weapons + Armor is only useful up to 10. If your Weapons + Armor exceeds 10 the extra armor is useless. 40 submissions (7%) had Weapons + Armor more than 10. I think the idea here was to create big exciting powerful monsters.

Actually, is even worse than this. Weapons 5 + Armor 5 seems fine, but Weapons 8 + Armor 1 is strictly better (assuming you wanted a successful predator). Basically, any Weapons + Armor >= 10 combination is dominated by another Weapons + Armor >= 10 combination having +3 Weapons and -4 Armor: same cost and same protection, but with more effective weapons.

That said, I've submitted a Wurm with Weapons 6 + Armor 6 anyway, because my inner Dungeon Master kept saying "It's a sodding Wurm, it should have higher Armor Class!". Weapons 9 + Armor 2 on a giant desert predator seemed just wrong (can someone please provide real-world examples of such "low-armored" deadly predators?).

Petrov Day 2021: Mutually Assured Destruction?

Ok, then I publicly declare to be quite impressed.

(I'll treat this as further evidence that inferential distances tend to be longer than expected)

Load More