Jiro

Karma: 5,285

Jiro 2 Feb 2026 17:22 UTC
4 points
0
on: The Smoking Lesion Doesn’t Really Distinguish EDT from CDT
My response to the smoking lesion question is:
1. If the lesion doesn’t affect your ability to reason, then deciding to smoke as the result of a reasoning process isn’t correlated with cancer, even though deciding to smoke in general is, so you should smoke.
2. If the lesion does affect your ability to reason, the question is ill defined because it may not be possible for you to execute some strategies, or only possible to execute them with some probability.
Of course “as the result of a reasoning process” needs to be phrased more precisely (every decision can be said to be the result of some reasoning process). Maybe “when you make some particular deduction, compared to a counterfactual world where you did not make that deduction”.

Jiro 2 Feb 2026 17:08 UTC
2 points
0
in reply to: James Diacoumis’s comment on: Would you kill a vulcan to save a shrimp?
You’re thinking of a p-Vulcan as a slight variation on a human. But the context includes:

Remember, it’s highly likely that shrimps have some form of phenomenal consciousness and experience some form of suffering. Shrimp suffering is bad. Even though we lack the capability to accurately predict its intensity, the shrimp certainly suffers more than the Vulcan would. Vulcans totally lack the capacity to suffer.

A p-Vulcan doesn’t have to be very humanlike. It only needs to be shrimplike, but without the ability to suffer. I do think that many existing video game characters would qualify, by that standard, as similar to a p-Vulcan. Is it wrong to kill those video game characters?

Jiro 2 Feb 2026 2:27 UTC
2 points
0
in reply to: James Diacoumis’s comment on: Would you kill a vulcan to save a shrimp?
Repeat the robots question, except ask the question about video game characters instead. Now, there are some game characters that have very simple patterns of behavior, but there are some that are a lot more complex, even if still describable by a set of algorithms. I’m sure there are characters that can and will beg you not to kill them. Is it wrong to play the video game and sacrifice the video game characters?

Jiro 31 Jan 2026 16:53 UTC
3 points
−1
on: Disjunctive arguments can be a reverse multiple-stage fallacy

Suppose we want to make an event seem likely. If we use the above method but slightly over-estimated the sub-event probabilities and use a large number of sub-events, then the resulting final probability will inevitably be very large. Because people tend to find moderate-range probabilities reasonable, this would be a superficially compelling argument even if it results in a massive over-estimation of the final probability. I propose this is a kind of reverse multiple-stage fallacy.

I suggest that this post is an instance of this fallacy. I gestured at the issue in my response but didn’t have a name for it. A nonexhaustive list with at least 12 listed subevents of “is humanity evil”, each of which is “there’s some chance that this specific thing is humanity being evil”.

Jiro 30 Jan 2026 16:45 UTC
14 points
4
in reply to: Gurkenglas’s comment on: AI found 12 of 12 OpenSSL zero-days (while curl cancelled its bug bounty)

a third-party entrepeneur can offer to back promising reporters.

Libertarians like to suggest this sort of thing a lot, and it usually doesn’t work. If you’re dealing with thousands of dollars, the third party enterpeneur has overhead for such things as lawyers and advertising. It would suck to do it and be arrested for money laundering, or be stuck with bad contracts, or be told that your rejected bugs violate a disparate impact rule. He’d have to make a profit after taxes, and he would have to make a profit compared to other uses of the money, not just compared to doing nothing. And OpenSSL could end the program, change the size of the bounty, or add conditions like “no third party” at any time, leaving the third party having spent on overhead and gaining nothing. (Or should the third party contract with a fourth party for insurance against OpenSSL doing this?)

In fact, they could announce that if they reject your bug, you have official license to sell it to third parties.

The official license won’t exempt OpenSSL from liability for publishing the bug and having party B use it to cause harm to party C—party C didn’t grant any permission.

Jiro 29 Jan 2026 15:36 UTC
5 points
1
in reply to: Gurkenglas’s comment on: AI found 12 of 12 OpenSSL zero-days (while curl cancelled its bug bounty)
That would mean that most human bug reporters wouldn’t report anything because of risk aversion. Also, remember the aphorism about it being difficult to get someone to understand something if their salary depends on them not understanding it—the curl maintainers or their bosses would be highly motivated to reject bugs.

The best I can think of is some kind of escalating fee where you get to report X number of invalid bugs for free and only after that do you have to pay to get the maintainers to look at your bugs.

Jiro 28 Jan 2026 17:36 UTC
11 points
8
on: Is the Gell-Mann effect overrated?
I suggest looking at, not articles about scientific topics, but articles that are about something else but where the author invokes scientific topics.

Jiro 27 Jan 2026 15:08 UTC
2 points
2
in reply to: Shankar Sivarajan’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
It’s true that a lot of places aren’t very scientific, but as long as your results were anti-Trump (and given the problems with this idea they pretty much will be), you’ll be able to post them even in places that aren’t very scientific, since they’ll care more about the result than the method.

Jiro 27 Jan 2026 4:16 UTC
1 point
0
in reply to: speck1447’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”

Anti-Obama sentiment was not partisan?

It is literally partisan, but it was nowhere near the same degree of being partisan that’s involved here.

I wonder if you’ve thought about the bad incentives this behavior produces?

The incentives that being able to respond to political rants create are that such rants become less useful (since they don’t go unopposed) and people would be less likely to make them. I find these incentives acceptable.

Like, if Satan himself were president of the United States and was killing a million people per day, eliciting celebration from his supporters, would you still think discussion was not justified on the grounds that it’s political?

Thousands of people are being killed in Sudan—certainly more than Trump has killed—and we’ve managed to go without LW being full of discussion about it. So yes.

Jiro 26 Jan 2026 23:55 UTC
−1 points
0
in reply to: speck1447’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”

but is only capable of being realized as broad anti-Trump sentiment?

Partisanship in the US could be something other than anti-Trump sentiment. There’s no logical necessity for it to be that, after all. It just isn’t actually separate from anti-Trump sentiment. (Outside the lizardman constant.)

Surely you understand the difference between an argument being difficult to challenge for social reasons and that argument being convincing, right?

I have no idea which argument you’re referring to.

if people shouldn’t be allowed to argue that Trump is genuinely exceptionally bad regardless of its truth value, then you’re also not allowed to argue that those people are overreacting regardless of it’s truth value.

First of all, you are demanding that you can attack all you want, but nobody gets to defend. No. This is like the difference between initiation of force and self-defense. If you’re going to argue that Trump is uniquely bad to the point where norms can be violated to tell everyone how bad he is, then everyone else gets to say that you are overreacting. You started it.

Second, my point is “you shouldn’t post about it here regardless of whether you’re overreacting.” It doesn’t matter how genuinely bad Trump is; you (and the OPs) shouldn’t be posting about him either way.

Jiro 26 Jan 2026 7:41 UTC
0 points
0
in reply to: speck1447’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”

Have you noticed this pattern on other issues here?

I haven’t noticed this pattern on other issues outside LW, so I don’t particularly expect to notice it within LW. Or rather, I have noticed the pattern on other issues, but all these issues involve, in practice, the same set of political enemies, so it isn’t really an independent set of things that people think are the worst ever.

Elsewhere you’ve claimed that this isn’t a hypothesis we should allow ourselves to even consider no matter how true it is

Sorry, I have no idea what you’re talking about. Unless you’re blurring the distinction between “post here” and “consider”.

here it seems like you’re arguing that it’s basically wrong in the sense that increased partisanship in general rather than exceptional circumstances are responsible.

Of course it’s increased partisanship. The last time a non-Trump Republican was president was January 2009. That was before most of the rise of social media. The era of social media has made Trump a target of partisanship in a way that was just impossible before now, not that nobody tried. And Trump is only a year into his current term, and for obvious reasons, the start of his term is going to lead to a lot more anti-Trump partisanship.

And if you’re asking specifically about why things got worse in January 2026, see above: I don’t believe they have.

I think I’d basically have to lie to someone to convince them that Trump is actually completely fine and totally in line with what they should expect, to the extent that breaking communication norms to mention his mistakes is escalatory.

No, what you should say is “it doesn’t matter whether Trump is totally fine.” You shouldn’t be breaking norms because someone is really not-fine. That allows one-sided norm breaking where the more you can work yourself into a frenzy, the more you have permission to do things that your opponents have no permission to do back unless they defect too and say that your side is the worst ever. The incentives created by “if it’s really bad, you can break norms” are their own kind of bad.

Jiro 26 Jan 2026 3:39 UTC
3 points
0
in reply to: TsviBT’s comment on: Strong, bipartisan leadership for resistance to Trump.
“It’s hard to empathize with them because it’s hard to empathize with most people who don’t agree with me” has rather different implications than “it’s hard to empathize with this group in particular”.

Jiro 26 Jan 2026 3:28 UTC
1 point
−6
in reply to: speck1447’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
I don’t think “why now” really needs an answer, because it’s not just now; we’ve seen it before. Also, since posts must appear in whole numbers, just normal fluctuation can take the number of posts from 0 to 1 or 2, so even if it was just now, the increase would not require an explanation unless we started seeing dozens of them.

And over the longer run, the overall acceptability of violating norms because one’s enemies are really bad has been going up.

Jiro 25 Jan 2026 16:56 UTC
2 points
0
in reply to: David Johnston’s comment on: Strong, bipartisan leadership for resistance to Trump.
If you are not MAGA, claiming to be MAGA for this purpose is concern trolling.

Jiro 25 Jan 2026 16:01 UTC
2 points
0
in reply to: Elizabeth’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”

Given that, how do you propose to distinguish the worlds where the president is genuinely much worse than average, vs. one where they’re completely precedented or following a trend line?

What you do is you say “given that, we are not doing politics here”.

But if you have to violate that rule, at least you could try to avoid biased tools and procedures.

Jiro 25 Jan 2026 15:59 UTC
4 points
2
in reply to: Archimedes’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
I’m not sure there is a measure that’s not cherry-picked. Deciding how good a president’s actions are inherently means deciding how bad particular measures are, how to trade them off, and which ones he’s even responsible for. It’s also worse if you’re measuring reactions to him, which several of your measures do; the easiest way for someone to be better on the measure “reactions to X” is not to do X.

Jiro 25 Jan 2026 4:02 UTC
3 points
4
in reply to: Archimedes’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
The problem with this idea is that the metrics are cherry-picked in order to make Trump look bad. And even then, what they actually measure is how much Trump is in conflict with the rest of the government, not how much we should be scared of him.

Jiro 25 Jan 2026 3:41 UTC
−2 points
−21
on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
AIs are known to have political bias in favor of the left. Using AIs to get information about Trump won’t produce fair results. Also, AIs are trained on the Internet and the media, which bakes media bias into the AI’s results. The AI will inherently trust the media’s side of things, and will rarely tell you that the media blew up an unimportant incident or left things out.

In some recent posts, some people have been like “Wait why is there suddenly this abrupt series of partisan LW posts that are taking for granted there is a problem here that is worth violating the LW avoid-most-mainstream-politics norm?”.

Like TV shows that point out their own plot holes, saying “look, my post is violating the rules” doesn’t excuse violating the rules.

There’s an abrupt series of partisan posts because partisans like to think their enemies are the worst people ever—so bad that they can violate all the norms they want in order to get their enemies, including norms about avoiding politics. And when you stand to gain from thinking your enemies are the worst people ever, there’s a lot of motivated reasoning going around.

Jiro 22 Jan 2026 5:06 UTC
0 points
0
in reply to: marisa’s comment on: Why I Transitioned: A Response
I don’t see why you’d want to die, come back for a few months knowing nothing, without any existing friends and family, being subject to arbitrary social pressures to see how you’d react, and finally leave the city only to be dissected.

Jiro 22 Jan 2026 4:51 UTC
3 points
0
on: Why I Transitioned: A Response
This is way off topic, but my reaction to Haibane Renmei was “there are aliens kidnapping dying people, wiping their minds, and locking them in a walled city for some sort of psychological experiments”. And around 95% of the show is consistent with that interpretation.