@hooser's banner p

hooser


				

				

				
0 followers   follows 0 users  
joined 2022 October 02 12:32:20 UTC

				

User ID: 1399

hooser


				
				
				

				
0 followers   follows 0 users   joined 2022 October 02 12:32:20 UTC

					

No bio...


					

User ID: 1399

That's a responsible use of AI as a tool to refine your thinking and communication. I place that in the same bucket as using spellcheck or a calculator. Similarly, I would not expect a disclosure of the tool's use.

Sounds like she did the responsible thing and gave you the relevant info, now it's your-body-your-choice. There are ways you can minimize the chance of getting herpes infection. Even so, I would not advise it for anything less than your near-future wife. As your friend can attest, getting stuck with herpes for life tanks your love life.

If I wanted to read polished prose expressing RHLF-ed opinions on charged current-events topics, I would read the New York Times.

The Motte expects thoughtful engagement. Someone posting AI output in place of their own ideas is being neither thoughtful nor engaged.

I support having a rule disallowing blocks of AI-generated text, with exception of meta discussions (like, "Claude output X to prompt P; here's what I think", or "Given prompt P, O3 says X but R3 says Y; here's what I think.").

I have a suggestion: when you post parts of your work-in-progress and want feedback, start with a personal note that says so. I've been skipping these posts (sorry, by now something that looks like LLM doesn't get my attention), but I won't if I know that there's a human who is developing thoughts and welcomes feedback. Best of luck on the book!

Dogs: companions during good times, food during famine. That's what I tell my mutt. She'll have to learn to share the voles and frogs she hunts, if times turn bad.

I second the choice of OnShape! Yes, it's free for educational purposes. I volunteer at a local after-school STEM program, and we have elementary-school kids CAD designs for 3D printers.

OnShape looks a bit intimidating at first (it wasn't UX-ed to death), but there are lots of videos on YouTube that do a how-to. Get to the point where you can make a "Sketch" of something simple like a polygon, and then "Extrude" it, and you're on your way to make interesting designs. In my experience, so long as the kid is coordinated enough to use a mouse, he'll get comfy with the basics faster than an adult.

Best of luck!

Thanks for the thoughtful response. There is indeed a danger on "overtraining" Kindness on family (and by extension kin and friends) if one takes the Confucian idea of family being the root of Kindness. I think the metaphor still holds: a tree sapling that has healthy roots but fails to grow is a failed tree.

The advantage of training in Kindness on the people you actually know and interact with is that it gets quickly apparent why Kindness is a hard virtue to achieve. Especially in the original sense of virtue as a moral force, a form of personal excellence that is actually useful in accomplishing something. If your father is eating himself into an early grave, what's a Kind way to dissuade him? If your teenage daughter is driving herself insane with social media, what's the Kind way to wean her off? Is it even Kind to meddle into their affairs? Are you sure of the superiority of your judgement? These questions get much harder, the nearer the people are to you.

Whereas if I train in Kindness on strangers, the typical failure mode is that it devolves into simple politeness.

Kindness as virtue is similar to the Confucian highest virtue of [Ren](https://en.wikipedia.org/wiki/Ren_(philosophy)), which I have seen translated as "humaneness", "beneficence", or "kindness". Kong-Fu Tze came up with the term himself, and the kanji 仁 is literally two radicals: 'man' and 'two' (or 'also'). Like "kind-ness", in the sense of considering someone else as like yourself or your kin.

(I promise to have a question for you in the end, after I set up the premise.)

Confucius (or rather his school) falls within the general framework of the Chinese political schools of thought of the time, which rests on three main questions: What is the Way (to fix the society)? What virtue (in the sense of personal power) does one get for following this Way? What kind of society does this Way lead to? (I'm loosely paraphrasing Van Norton's intro to classical Chinese philosophy, which is excellent.)

So Confucian school regarded the virtue of Kindness as power, which makes sense: if you understand another person, does that not give you power to guide that other person in a way closer to your goals? The Confucian school also was adamant that this very useful power is hard to obtain. To truly be Kind, you need to spend years studying people, starting with those closest to you and whose foibles you are most familiar with. Thus the school emphasized family as the root of Kindness: if you can be Kind to your grouchy out-of-touch parents, your annoying siblings, your infuriating spouse, your disobedient children... well, then you're onto something. (In particular, maybe then you can transfer that power to being Kind to your grouchy out-of-touch boss, your annoying co-workers, your infuriating office mate, and your duty-shirking underlings.)

So my question for you is: do you regard the virtue of Kindness as something hard to obtain, something that requires years of diligent study, as opposed to a more common notion of "kindness" in a sense of good disposition or well-intention? And if you do: how do you go about obtaining this virtue? (I suspect that, as a modern progressive, your answer would be substantially different from Confucius.)

I just don’t think that a serious attempt at real data collection is going to happen for societal reasons

In "The Typical Man Disgusts the Typical Woman" post Update (the post that y'all are discussing), Caplan links to Emil Kirkegaard's analysis of four much more representative data collections:

  • General Social Survey (GSS), USA
  • NLSY Add Health, USA
  • Wisconsin longitudinal study (WLS), USA
  • German General Social Survey (ALLBUS), German

In all of these, the OK Cupid's stark disparity in ratings do not reproduce. Women's photos do tend to get slightly higher attractiveness ratings, but, you know, there's probably a reason why both men's and women's magazines are full of half-naked women.

Maybe you're right. I have drifted away from watching documentaries in the past decade, and even then my preference was for nature and science themes. It's possible that the standards of presenting evidence have significantly changed (deteriorated?) since then.

They experienced it, and they felt that the information from that experience could best be communicated to me by those fictional films. They felt it captured something about what it felt like.

I agree that the value of fiction is, among other things, in its success in conveying emotional truths. If "how it felt" is best conveyed with disturbing depictions of atrocious savagery told in flat matter-of-fact manner (like in Tim O'Brien's "The Things They Carried"), then that's what the author does. Nobody need question whether this specific instance of atrocious savagery happened, or even whether this type or this level of atrocious savagery happened somewhere in this time-and-place. Nobody need question such thing, because that's besides the point, so long as the depiction serves to convey "how it felt".

The problem arises when fiction gets presented as historical fact. I would have a problem if a documentary on Vietnam intermixed historical footage with scenes from "Apocalypse Now", while Tim O'Brien reads excerpts from "The Things They Carried", especially if the intended audience is not familiar with either work or the author and thus is unaware that they are works of fiction.

I did watch it on Kanopy, through my local library.

I am watching a film about a subject about which I have, at best, a cursory knowledge; how much can I rely on its factual claims? If it's a work of fiction: not at all. Just enjoy the story. Some historical background might be rooted in fact, but I am not in a position to tell. If it's a documentary: I expect that factual claims are, to a large extent, true. Sure, I expect a documentary to cherry-pick its facts to present a compelling narrative, but that's what distinguishes a documentary from a work of fiction: the narrative is constrained by at least a few asserted facts. It's common for a film-maker to outsource the actual statement of facts to an expert. The expert bears the cost of getting the facts wrong; the film-maker bears the cost of choosing experts poorly.

So here I am, watching this acclaimed documentary about a topic I know little about. It shows archival-looking clips, with meticulous citations -- I trust that those clips are what they appear to be. It shows quotes from diplomatic archives, with meticulous citations -- I trust that those quotes are from actual diplomatic archives. Twenty minutes in, it (for the first time) appears to have an expert contextualizing the main subject, again with citations. Do I continue to extend my trust to the presented facts, confident that a meticulously researched documentary would feature solid expertise in the subject matter?

The two-minute narration is a mixture of factual claims and narrative spin; yes, I understand that "Congo Inc." is a metaphor, but: Did Congo's rubber really "smooth[ed] the way to World War I", or is that a terrible pun? Was Congo's uranium key to US bombing Hiroshima, or was it just the most convenient source? How much did Congo copper contribute to the devastation of Vietnam, and how much of that devastation would have happened with other sources if Congo's copper was not available?

What is this guy's expertise in, anyway? Fiction. He is a writer of fiction. He may be a very good writer, and he may even meticulously research the background setting for his novels. But he claims no historic expertise. And the work he's reading in claims no historical accuracy.

There was a saying from this past election, something like "Trump lies like a used car salesman, Harris misleads like a lawyer." (No offense to lawyers.) The film didn't lie, it misled, and it misled subtly; it misled about the apparent level of expertise.

If I were already an expert in DRC history, it wouldn't matter. As an expert myself, I would evaluate any claims by their content not provenance. But I am the opposite of an expert; I can point to DRC on the map and I have some vague knowledge of the 20th-century history of Sub-Saharan Africa in terms of de-colonization and a tug-of-political-influence between USA and USSR (and later China). So I cannot possibly evaluate these claims by their content, and I must cautiously rely on expertise. That's why presenting someone as an expert when he's not is a big deal for me.

So, maybe, this film just isn't for me? Maybe it's aimed at people who are far more knowledgeable about the subject matter, who would not possibly mistake the expertise of Bofane? The film-maker is, after all, Belgian, and maybe the local audience is far more steeped in the history of the country's former colony. But I don't buy it. The film premiered at the Sundance Film Festival, it's clearly shooting for a broad audience.

I would therefore like to make a prediction, even if I am too lazy to actually carry it out. Let's say that a poll is conducted among the film's audience. The poll takers watch the two-minute clip of Bofane talking (22:56 to 24:19). Then they respond to a question like this one:

Which best describes "Congo Inc.": (a) an academic publication by a professional historian, (b) a non-fiction account by a professional journalist, or (c) a work of fiction by a professional fiction writer.

My prediction is that less than 10% would choose (c).

What an impressive propaganda technique. That's my one-line review to the "Soundtrack to a Coup d'Etat", and I mean it most sincerely. I really am impressed.

This quote from a New York Times film critic serves both as a quick plot summary and as the main impression the film conveys:

... a sprawling film that's a well-researched essay about the 1960 regime change in the Democratic Republic of Congo and the part the United States, particularly the C.I.A., played.

Let's focus on the "well-researched" part, the part that lends the film a documentary gravitas, the propaganda technique I so admire.

The documentary is a collage of footage, archival audio and video clips, and quotes with careful citations that briefly appear on screen. It doesn't have a narrator--except occasionally it does, like from 22:56 to 24:19, where English text quoting In Koli Jean Bofane's Congo.Inc overlays archival footage while the said author reads his work in original French:

The algorithm Congo Inc. was invented Africa was carved up. Capitalized by Leopold II, it was quickly developed to supply the whole world with rubber and smooth the way to World War I. The contribution of Congo Inc. to the 2nd World War was key. It provided the U.S. with uranium from Shinkolobwe that wiped Hiroshima and Nagasaki from the face of the earth while it planted the concept of 'mutual assured destruction'. During the so-called Cold War the algorithm remained red-hot. It contributed vastly to the devastation of Vietnam allowing Bell UH1-Huey helicopters, sides gaping wide, to spit millions of copper bullets from Kolwezi over the countryside from Hanoi to Hue via Danang all the way to the port of Haiphong.

Here's the beauty: "Congo Inc." is a work of fiction. It is a novel. It is not, and never claimed to be, an accurate and contextualized account of history, nor is it subject to the kind of critique for accuracy that a work of non-fiction would receive.

The technique allows the film to convey the impression of historical gravitas while absolving it of any responsibility for truth, accuracy, or context. What is there to criticize? All the film does is feature a Belgian writer connected to Congo by birth and some years of residence, reading from his work. It's a work of fiction--so what, when the main theme of the film is to suggest the interweaving of art and politics. The film's omission of the category of the work is completely in line with their omission of such information about their other sources. Surely the film has done its due diligence by accurately citing the sources, thus providing any interested viewer with the requisite information to establish the necessary level of epistemology for the content of any citation it happens to feature. If anything, it's a mark of respect for the sophistication of the viewer that the film doesn't bother contextualizing these works, since surely the viewer is quite familiar with both the history of Sub-Saharan Africa in general, and prominent literary works of authors with Sub-Saharan African ties in particular.

Yes, its Sundance Festival Special Jury Award for Cinematic Innovation is well-deserved. I look forward to future adaptations of this technique, where documentaries about the CIA quote John Grisham's novels, and documentaries about the Catholic Church quote "The Da Vinci Code".

Clear commitment to a shared future helps. Seven months apart is not that long in light of years of marriage. My husband and spend several months apart while engaged, and I know married couples who spend a few years working in different states (and even countries).

Does anything prevent the two of you from getting engaged?

Sounds like you are most excited by the possibility of switching to HR. You can:

  • (1) Get the SHRM-CP (Society for Human Resource Management Certified Professional) certification. You can do it on your own time and at your own pace.

  • (2) Ask your employer to either transfer into their HR full-time, or to take on more HR tasks. If the latter, ask that your title reflect this new responsibility.

  • (3) Document all the HR-adjacent tasks you have already done, and keep documenting it. Use the documentation to negotiate (2).

  • (4) If for some reason you look into (1) and decide against it, "Professional in Human Resources" certificate from Human Resource Standards Institute is also good.

  • (5) If you are working through (1) or (4): learn how to use an AI to help you learn. I recommend Claude. Don't let it do your thinking for you, but do use it as a broadly knowledgeable tutor who sometimes goes off the rocker (so validate any concrete piece of info you absolutely need to rely on).

Good luck!

and currently am doing my PhD in Baltimore

Let's have some straight talk about the unspoken expectations of PhD and beyond.

Successfully finishing and defending your dissertation means very little if you haven't used your time while in the PhD program to establish a strong professional network. Without the latter, all you have is an extra line on your CV (or resume), and there are plenty of others out there with a similar or more impressive-sounding line in their CV. This is true even if you turn your dissertation into several publications, and even if those publications actually find readers beyond Reviewers #1 and #2. None of that is a substitute for a strong professional network.

Fortunately, building a strong professional network in graduate school coincides precisely with your desire for a community. Right now, you have fellow PhD students in relatively close physical proximity and in sufficiently close sub-fields / fields, pursuing similar-enough goals. All want to successfully complete their dissertations. All are working on something that (at least at the beginning) they found interesting. Quite a few of them will be your future professional colleagues. Building a strong professional network starts with organizing your fellow PhD students into a mutually supportive network.

Does your department have a weekly graduate student seminar, where grad students can present an interesting article or some partial progress on their dissertation? If yes, attend it and present in it, and hang out afterwards to casually discuss stuff with the presenter. If not, organize it. Ask your department head for pizza funds, chances are pretty good they'd be thrilled that someone is willing to take on the organizational task.

Are you in a program with too few grad students? Well, are there grad students in adjacent programs? It's very useful to be able to talk about your research to people outside of your field, and a bit of cross-discipline pollination goes a long way. Again, ask for pizza funds.

Have the seminar repeat at a regular time, so people get used to it being a thing. If weekly is too frequent, have it bi-weekly. Or first and third Thursday of the month. Invite undergrads that are heading into similar fields. Invite professors; quite a few appreciate the opportunity for low-stress chats about something in or close to their field. If there are local people in the industry that are relevant to your field, invite them too; industry people can bring boots-on-the-ground perspective that academics miss.

Do you or your fellow PhD students take classes? If yes, do they have informal study sessions? If yes, make a point to attend those. If no, organize one. It could start small: just you and one other student, and then make it generally known that others are welcome. Have it at a regular time and place, and be consistent about showing up.

Have you stopped by the office of every professor in your department to chat about their research? Do that. Ask also about the social aspects of their field: Where are the people who work in that field? Is it a more-or-less cohesive group, or are there rival factions? What conferences / forums do people in this field use to informally exchange ideas? Which journals do they value, and which are junk?

Are there local or regional conferences in your field? Do go to those. Preferably, organize some of your fellow PhD students to come with you. If there aren't... maybe there are, but you don't know it. Chat with your professors. Baltimore is plenty big, and it's close to so many other centers of academia.

And yes, by all means go running and attend church. Touch grass. Do what you need to keep healthy and grounded. But understand that, at this juncture, those are unlikely to be the communities that you'll keep.

I'm fine with accepting that Saul of Tarsus is not only a historical figure, but that the legends about him are sufficiently close to what happened to that figure in reality (+/- miracles). I am fine with having a high likelihood of a historical Jesus, and that this man was an object of a cult following, though I find it unlikely that the historical Jesus would match the Jesus of Christian mythology to any reasonable degree. I doubt the existence of a historical Judas, he's too convenient as a one-stop-scapegoat literary character.

For the purposes of the game hydroacetylene proposed, I am primarily interested in the literary characters of Jesus, Paul, and Judas, and I would consider their historicity only because it makes the read-the-Bible-as-if-it-has-unreliable-narrator more plausible. They can then write some "tell it like it really was" books.

Let's say that for two out of the three of these figures, there is a lack of evidence outside of biblical literary traditions, which could well be apocryphal.

Can I bring back semi-historical figures? Because I would love to bring back Saul of Tarsus (aka Saint Paul), Judas Iscariot, and Jesus of Nazareth, and have them simultaneously do the talk-show/podcast circuit to promote their various new books:

  • "The Art of Sacred Sass"
  • "Kiss & Tell"
  • "Loaves and Fishes: a Cook Book"

I’m currently just thinking about how weird all of this is.

This here, this is a great place to be. When the world feels weird, that's a palpable sign that something is the matter with my internal model of the world. And that's as it should be! If my internal model of the world is so snug and secure that I feel not a twinge of discomfort, of puzzle, of wonder, that's when I am most in danger in getting blindsided when that internal model falls short of reality. And the plain fact is that my internal model of the world will always fall short of reality. So the best I can do is to hold my model lightly, play with it, and always be willing to adjust it as new information arises.

As to a story that places my consciousness into the grand scheme of things, I am quite partial to the one from "The Elephant in the Brain". The "I" -- that feeling of consciousness -- is not the captain of this mind/body, is not even the team leader, but is rather the spokesman--the spokesman of a generally disorganized cabal pretending to be a well-organized administration. The cabal does something, and "I" stand in front of the members of the press and spin it as best as "I" can.

I like this story because, paradoxically, it gives me agency.

"I" can't compel the cabal against the cabal's wishes. "I" am not even directly privy to the inner politics of the cabal: "I" don't really know why the cabal did what it did. But "I" do have influence on the cabal, because the cabal cares about self-image and public-image, and "I" am the one who goes in front of the members of the press and spins those stories. So if, as part of a story about "myself" (especially to actual other people), "I" commit myself to some action in the future (an action not immediately salient so as to not step on any of the cabal's current sore points, and one that's not too difficult), then the cabal is incentivized in following through that commitment to avoid negative publicity.

I got into the habit of daily jogging this way.

It's not easy to start jogging. The inertia of habit is against you, and the activity isn't rewarding. If you aren't already used to it, sustained cardiovascular exercise feels bad. You are out of breath, you feel nauseous, and (at least for me jogging) you feel distracted by the jagged vision produced by the bouncing eyeballs. That's quite a barrier to overcome.

So "I" spun it, and "I" spun it hard. "I" told stories of heroic effort, of commitment, of taking on the unpleasant hard tasks for the greater good. "I" advertised my intentions to my spouse and my friends, and "I" updated them on my progress and setbacks. Two months it took me to stop hating the actual act of jogging (though even then I felt great afterwards). After another two months, jogging got kind of enjoyable. Now, it's a habit, and I get the jogging itch if I skip a day.

A quick cross-cultural comparison: Wife-beating is common in Eastern-European cultures (say, 1-in-10 couples). A stereotypical Russian phrase, uttered by a Russian wife about to get beaten, is "Just don't punch the stomach!". It's reasonable to suppose that, in the past, the rate at which pregnant women got beaten was higher than now.

I recon that the force experienced from a car deceleration is smaller than an occasional drunken punch on the womb.

Meta ends its DEI program (internal memo, Ars Technica verification). The company is disbanding its DEI team. It will no longer use "diverse slate hiring" (intentional seeking-out of candidates of particular underrepresented minorities). It is "sunsetting our supplier diversity efforts", which probably means that they will no longer privilege minority/women-owned suppliers.

It is ending the perception that it has representation goals. Yes that's convoluted, but how else does one interpret this statement:

"We previously ended representation goals for women and ethnic minorities. Having goals can create the impression that decisions are being made based on race or gender. While this has never been our practice, we want to eliminate any impression of it."

The stated reason for the shift in policy:

The legal and policy landscape surrounding diversity, equity and inclusion efforts in the United States is changing. The Supreme Court of the United States has recently made decisions signaling a shift in how courts will approach DEI. It reaffirms longstanding principles that discrimination should not be tolerated or promoted on the basis of inherent characteristics.

That is, they expect to no longer be sued based on "disparate impacts", but possibly sued based on preferential treatments. This... makes sense for a company to do. McDonalds is doing it; Walmart did it more than a month ago.

I expect more companies to follow suit (quietly or loudly). My question is: are there any corporate for-profit true-believers who will stick with the DEI initiatives? Ben and Jerry's, maybe?

Day 14 part 2 asked to find the first output configuration containing a christmas tree. This is essentially impossible to solve independently for an LLM since the problem didn't even specify what the christmas tree would look like and there are many plausible ways to draw a christmas tree with pixel art.

I thought about how to solve it in an automated way, and came out with the following approach. I figured that any recognizable pixel art of a christmas tree will have a lot of straight lines or filled-in spaces. So:

  1. For a configuration of robots, count the number of robots that have two other robots right next to them in a straight line. (E.g., for a robot in position (x,y), are there also robots on (x-1,y-1) and (x+1,y+1)?)

  2. Collect this measure for n-th move, for n from 0 to 10000.

  3. Submit n with the highest measure.

(If that's not the answer: repeat for another 10000 moves, etc. Turned out this wasn't necessary.)

More than a decade ago--before this war--I happened to go back to Kiev/Kyiv for the first and possibly the last time. My older brother, more familiar with the place and with better recollection, guided us to the particular nine-story apartment building we used to live in. Its yard used to have an old chestnut tree, The Climbing Tree, the one that only the older kids could climb, the one with a straight trunk too wide to hug and too smooth for footholds, whose lowest branch was just low enough that I could touch it when I jumped.

That tree was still there. That lowest branch was now chest-level. No child climbed it.

Yes, and those are the "successful" graduates.

My niece goes to Cupertino HS. She claims to suffer "trauma", like 50%+ of her classmates. Her psychologist agrees (just like the psychologists of those 50%+ of her classmates do).

The amount of pressure those students heap on themselves, on top of the high expectations of the parents, seriously distorts their perception of reality. What does it mean to be "successful"? Is it enough to graduate HS and get a job / start a family? No, of course not. A "successful" person successfully founds a start-up, or at the very least goes to one of the universities on The List, where they will successfully found a start-up (or, as a distant second, get a highly remunerative PMC career).

And the alternative? If you don't have it in you to write that killer app by the age of 15, if you gods-forbid don't get into any of the universities on The List... well, that's it, you failed. And indeed you did, if your definition of "success" is to have the means to continue to live in the Bay Area. And no, you didn't make those nice networking connections with the "successful" classmates--they are not interested in burdening their networks with failures.

Fortunately, the Bay Area culture also offers a ready alternative to owning your failure: you are a Person with Disability, it's not your fault. My niece is on all the trendy spectra.

My niece is a bright girl, and I have urged that she come live with me for a while in the Flyover Country and go to a regular school for a change. Recalibrating would do her a world of good. Alas, no luck so far.