TheJach.com

Jach's personal blog

(Largely containing a mind-dump to myselves: past, present, and future)
Current favorite quote: "Supposedly smart people are weirdly ignorant of Bayes' Rule." William B Vogt, 2010

Thankful for 2022

Non-exhaustive/unsorted list of things I'm thankful for this year: No new family deaths, no estrangements, still got my health, still got my long-term friend/housemate (despite wanting to move this summer to escape the rent mill), still got my other long-term friend in Utah to visit and chat with, still financially well enough despite the markets to keep doing my own thing and be more generous than last year at the same time, seeing a couple great vtuber revivals and another 3-sister collab, meeting Coyo / コヨちゃん and all the fun (and some worry) that has brought, regaining some ability/motivation to write things and subject myself to more (hopefully) strengthening stresses, not succumbing to certain things, learning more new things again despite slowly and despite many more to go / take up again, some good books, at least one really good game, can still program my way out of a paper bag just fine, have been able to travel (and just committed to one more big jaunt next month), several quality-of-life upgrades for my computer room I spend too much time in, and that I remain alive in relative comfort.

I hope those dealing with their own tragedies or just minor struggles this year can find something to be grateful for too. It's no good to dwell all the time on the negatives.

See Full Post and Comments

The allure of within our lifetime

More discussions lately on the webs around for lack of a better shorthand Singularity Pessimism. Thought I'd write down some ramblings to look back on later. A while ago the joking-but-no-we're-serious "die with dignity" piece came out, MIRI essentially raising a white flag to oblivion of humanity at the hands of unaligned AGI.

They've certainly thought about it a lot more than me. I remain in my not too sure state about whether we could in fact muddle through somehow. But while I have absorbed some of the pessimism, I think it's probably harmful overall.

"We'll have to work faster" was once a condolence in the face of death. The world death clock still stands at 1.8 per second after many years of global progress. This is unacceptable, and reducing this value to 0 should be a fiery motivator, especially for young people with a lot of talent.

See Full Post and Comments

Make someone feel

On a recent trip my Dad was telling a story which I guess everyone back home has heard numerous times by now, but was fresh to me, about a time he had to spend some time in jail. He got a weekends-jail deal, anyway grandma dropped him off once and to try cheering him up said something like (not sure of the exact phrasing) "Maybe you can make someone happier in there." Of course Dad then was like "No mom, it's jail! No one's going to be happy!" and slammed the door. But however she really said it, the sentiment is such a her-thing to have said. Maybe something like her life motto? She always tried to make people happy and was usually good at it. Super kind. It'd be nice if she were still around.

Despite not inheriting such great skill, I think it's a good idea to keep in mind especially for less than ideal situations for both yourself and those around you... If you can make someone happy, or at least try and think of some ideas to do so, that's often a lot better than many other alternatives like doing nothing or just wallowing. There may be other things you can do too though, so don't neglect them (and for some people kindness seems to hurt and they really would prefer you left them alone or expressed sympathy in less direct ways).

I think I recall reading somewhere of a parent making sure to ask their child after school over dinner what they did that day to make someone's day better, if anything. My immediate reaction to such an idea is that I wouldn't want to put this pressure on my own child as I sort of see it as a cousin to misguided philosophies about living for the sake of others, but I think that's more of an adult perspective and not really appropriate here, especially because such heavy-handed perspectives are cousins to other misguided philosophies about any form of sacrifice being evil. People are just trying to be more kind every day, that's typically a good thing. For a kid especially, the acts of kindness can be very small and still matter, like holding open a door, or a nice compliment, or giving up some table space at the cafeteria, or a hug to someone who needed one, or... it's probably good to engage a kid's creativity this way too.

See Full Post and Comments

Song Translations - Calling To The Night

Thought of a new little project to do: song translations! (Edit: and by 'project', maybe I'll post one of them before the year is out...) They're mostly going to be Japanese -> English, in an effort to help me learn more Japanese, but this first one is actually English -> French and something I did all the way back in high school for French class.

The song is Calling To The Night, by Natasha Farrow, from the game Metal Gear Solid Portable Ops. (I think I heard it first in Smash Bros.)



See Full Post and Comments

Of course there's evidence

These last few weeks have been amusing for yet more failures to replicate (or just re-upping some other notable failures) in science. I expect some of these failures will themselves fail to replicate -- i.e. some flaw discovered -- and some future meta-analyses will try to pick up the pieces or attempt to synthesize or whatever. (Sometimes just pointing out inconsistencies is fine. There are two well-established methods to deal with overcoming traumas/phobias, but the original individual theories behind each would, if true, prohibit the other from working. There are more modern theories that can account for it, but it took people noticing.)

Some recent highlights (thanks to @hbdchick for hearing about many) include mere exposure effect not replicating, a bad history of hummingbird flap rates, vitamin D with/without calcium doesn't matter for bones, Alzheimer's research is bunk, no evidence for nudging, doubt on serotonin as a useful part of theory on depression (but some drugs still seem to work somehow sometimes anyway, so don't get too skeptical (but still skeptical and notice long term issues)).

An older highlight is around Koko the monkey. A lot of the flaws with Koko are repeated in small by so many published researchers, i.e. that there's just no data to even attempt an outside analysis or replication, here's another recent thread pointing this out.

See Full Post and Comments

CEO-monarchs would be best in a patchwork

I've liked the idea of neocameralism since I was first exposed to it. In short, the government runs as a classic monarchy with an absolute King, but it's really structured like a modern corporation and so the King is really a CEO-King who is formally accountable and replaceable. (By the board members, which themselves are replaceable by the shareholders.) This improves over classical succession issues where the best person can take the reigns when the CEO dies or retires, not necessarily the CEO's son. This also improves some accountability issues because if the country is dying and the CEO isn't seen as doing his best to stop it, he can be gotten rid of with a nice and clean process to let someone else have a go, much the same as within the company/country official org chart itself the CEO can replace anyone.

This has some obvious defects, though some maybe not worth too much concern about. The first is: will a CEO-King who is being replaced actually go quietly? Maybe in the distant future you can actually crypto-lock and secure all weapons of significance to make this a guaranteed peaceful transition, but even without such questionable tech, I think it'd usually go well anyway. For one, humans respect hierarchy quite a bit -- even when guns are in play, even when there are many reasons someone with a cynical disposition can question legitimacy. The historical case of Lincoln and his Generals is a good one to reflect on. Why should the President, who typically lacks any military training, be Commander-in-Chief after all? It's amazing this country has never had a successful military coup against the sitting President (I'm unaware of even an unsuccessful one). Instead, even top generals defer, and even go away, and are replaced by other generals. And Lincoln himself, had he not been shot, eventually would have deferred to let another President take office. (Though perhaps like FDR not immediately.)

Also, having a nice severance package is a common and good incentive to keep people from doing harm on their way out, or later on.

See Full Post and Comments

Grabby Aliens and Paperclips

In a recent bar-talk-style chat with a friend, we got on the topic of AI risk again. Somewhere in the conversation he expressed surprise at some of my arguments, given some of my older arguments. Namely, I put forth some arguments de-emphasizing the importance of recent AI advances when it comes to full-blown AGI, and generally I expressed skepticism that full-blown AGI was "near". At least, a lot more skepticism than I've ever done before with him, it seems.

Ultimately I think I've just been more and more influenced by Robin Hanson's viewpoints, and his recent work on grabby aliens may have been another push that I've only connected now.

But stepping back a bit, I want to describe the shape of my beliefs about AGI. Before I get to probabilities I have to first talk about possibilities. So first of all, it seems obviously possible that creatures of human intelligence could one day create machine intelligence that rivals and surpasses them. It does not at all seem that it's an impossible engineering challenge in the way that a perpetual motion machine is one**, nor does it seem like humans should just happen to be close to the limits of general intelligence. If nothing else, we could at least think faster, which alone would be a large advantage even without greater generality to thought.

See Full Post and Comments