Bioinfohazards 2019-09-17T02:41:30.175Z · score: 76 (29 votes)
Spiracular's Shortform Feed 2019-06-13T20:36:26.603Z · score: 36 (8 votes)


Comment by spiracular on Who lacks the qualia of consciousness? · 2019-10-06T22:43:32.566Z · score: 2 (2 votes) · LW · GW

In one of the sub-comments, I thought about the tests that identified mental-imagery, and started thinking of how you might test for several variants of "lack of sense of self" or some related attributes.

Related Tests for inner-sense or model of self

No-qualia seems challenging to test. But "no model of self" (one form of "lack-of-self-awareness") seems halfway-there, or at least in the correct spirit of the question? And I think that could be tested reliably; just get a group of people to predict their own behavior, and watch that subset of the group who reliably fail catastrophically at this.

For lack of consistency and other-awareness... There's a Nazi (ETA: Eichmann) who seemed likely to be a troublingly-vivid example of "no consistent worldview or other-awareness"; all his words and beliefs were inconsistent platitudes, and he seemed genuinely surprised when Jewish judges didn't feel sympathy for the difficulty in his attempts to get promoted by doing a "good job" optimizing trains for death. Unfortunately, I can't track down his name. If someone knows who I'm talking about, I'd love to be pointed at an article about his strange psychology again.

Lack-of-attachment-to-internal-identity seems to be another semi-related thing. I feel like there are some things where I care about "identity-alignment" a great deal, and other matters that others clearly care about where I just lack any feeling of identity euphoria/dysphoria around the matter regardless of what I do. I suspect there are some people who lack either sensation altogether. Probably some fraction of those people come across as identity chameleons; people who switch out identities according to external incentives, because they have no internal reason not to.

(Personally, meditation updated me considerably towards a reduced attachment to internal-identities, but there are still some I'm attached to and care about maintaining.)

Alexithymia is a phenomenon where you lack awareness of your own emotions, sometimes even as you are acting them out. This seems easy to test in a manner similar to green/red color-blindness; have the person try to appraise what sort of emotion they're feeling, and then read their circumstances or watch their behavior for a read of which emotion it actually is, and see who usually seems to misjudge it (or believe they're not feeling emotions entirely).

Another p-zombie variant

There's a different p-zombie subtype I've been thinking about a great deal myself..

If you set up a system where there's an observer, an actor, and an environment, then there are 2 kinds of consciousness:

  • The observer influences, controls, or is the actor who does things in the environment
    • A policy feedback loop
  • The observer is just modeling what the actor will do in the environment
    • A one-way modeling of the agent

I suspect the later can feel "conscious" even if the observer never influences the actor in any way.

Humans are usually a bit of both, but someone who only has "consciousness" in the later capacity feels a bit like a... "consciousness hitchhiking on a q-zombie" to me.

(Related: The Elephant and The Rider)

Comment by spiracular on Who lacks the qualia of consciousness? · 2019-10-06T22:36:06.132Z · score: 1 (1 votes) · LW · GW

Mental imagery: Drawing seems to reliably distinguish between coherent-mental-visualizers and those who aren't. Tasks like "count the stripes on the tiger" make sense to a vivid/detailed visualizer, but not to someone who is just holding on to the concept "striped big cat."

I suspect drawing-attempts would also reliably identify people like "The Man who Mistook His Wife for a Hat", who viewed people as a "disorganized bag of facial-features" and had to rely on a single distinctive trait to identify even people he knew well (ex: Albert Einstein & his eccentric hairstyle), and who described things that weren't there when trying to interpret a low-feature image like a picture of the dunes of the Sahara.

Comment by spiracular on Are there technical/object-level fields that make sense to recruit to LessWrong? · 2019-09-25T13:46:14.703Z · score: 18 (6 votes) · LW · GW

Biology-nerd LWer here (or ex-biology-nerd? I do programming as a job now, but still talk and think about bio as a fairly-high-investment hobby). BS in entomology. Disclaimer that I haven't done grad school or much research; I have just thought about doing it and talked with people who have.

I suspect one thing that might appeal to these sorts of people, which we have a chance of being able to provide, is an interesting applied-researcher-targeted semi-plain-language (or highly-visual, or flow-chart/checklist, or otherwise accessibly presented) explanation of certain aspects of statistics that are particularly likely to be relevant to these fields.

ETA: A few things I can think of as places to find these people are "research" and "conferences." There are google terms they're going to use a lot (due to research), and also a lot of them are going to be interested in publishing and conferences as a way to familiarize themselves with new research in their fields and further their careers.

Leaning towards the research funnel... here's some things I understand now that I did not understand when I graduated, many of which I got from talking/reading in this community, which I think a "counterfactual researcher me" would have benefited from a lucid explanation of:

  • Transferring intuitions around normalization
    • how to do it, why to do it
    • see how it eliminates spurious leads in data like goddamned magic
  • How to Handle Multiple-Hypothesis-Testing
  • A really good explanation of MCMC
  • Implied "priors," model assumptions, and how to guess which ones to reach for and screen out the ones that are wrong
  • When is trying to add ML to your research a good or bad idea

Things I think we've done that seem appealing from a researcher perspective include...

  • Some of the stats stuff (I may not remember the precise sources, but this is where I picked up an understanding of most of what I listed above)
  • Thoughtful summaries/critiques of certain papers
  • Scott Alexander's stuff on how to thoughtfully combine multiple semi-similar studies (so very much of it, but maybe one in particular that stood out for me was his bit on the funnel graph)
  • I vaguely remember seeing a good explanation of "how and why to use effect sizes" somewhere

(...damn, is Scott really carrying the team here, or is this a perception filter and I just really like his blog?)

Small sample sizes, but I think in the biology reference class, I've seen more people bounce off of Eliezer's writing style than the programming reference class does (fairly typical "reads-as-arrogant" stuff; I didn't personally bounce off it, so I'm transmitting this secondhand). I don't think there's anything to be done about this; just sharing the impression. Personally, I've felt moments of annoyance with random LWers who really don't have an intuitive feel for the nuances for evolution, but Eliezer is actually one of the people who seems to have a really solid grasp on this particular topic.

(I've tended to like Elizer's stuff on statistics, and I respected him pretty early on because he's one of the (minority of) people on here who have a really solid grasp of what evolution is/isn't, and what it does/doesn't do. Respect for his understanding of a field-of-study I did understand, rubbed off as respecting him in fields of study he understood better than I did (ex: ML) by default, at least until my knowledge caught up enough that I could reason about it on my own.)

((FWIW; I suspect people in finance might feel similarly about "Inadequate Equilibria," and I suspect they wouldn't be as turned off by the writing style. They are likely to be desirable recruits for other reasons: finance at its best is fast-turnaround and ruthlessly empirical, it's often programming or programming-adjacent, EA is essentially "charity for quantitatively-minded people who think about black swans," plus there's something of a cultural fit there.))

Networking and career-development-wise... quite frankly, I think we have some, but not a ton to offer biologists directly. Maybe some EA grants for academics and future academics that are good at self-advocacy and open to moving. I've met maybe a dozen rationalists I could talk heavy bio with, over half of which are primarily in some other field at this point. Whereas we have a ton to offer programmers, and at earlier stages of their careers.

(I say this partially from personal experience, although it's slightly out-of-date: I started my stay in the Berkeley rationalist community ~4 years ago with a biology-type degree. I had a strong interest in biorisk, and virology in particular. I still switched into programming. There weren't many resources pointed towards early-career people in bio at the time (this may have changed; a group of bio-minded people including myself got a grant to host a group giving presentations on this topic, and were recently able to get a grant to host a conference), and any that existed was pointed at getting people to go to grad school. Given that I had a distaste for academia and no intention of going to grad school, I eventually realized the level of resources or support that I could access around this at the time was effectively zero, so I did the rational thing and switched to something that pays well and plugged in with a massive network of community support. And yes, I'm a tad bitter about this. But that's partially because I just had miscalibrated expectations, which I'm trying to help someone else avoid.)

Comment by spiracular on What are some of your "Crazy Ideas" that you're currently thinking about? · 2019-09-22T01:30:12.387Z · score: 6 (4 votes) · LW · GW

Many of the ideas that most alienated me from normal people are pretty mundane here.

(Years ago, some normal person asked me what I thought about what would happen in the future, in light of overpopulation and the climate crisis. When my response involved "AI-based catastrophe," "problems capitalism is or isn't adequate to solving," and geoengineering, they straight up checked out of that conversation and asked somebody else.)

So what am I thinking about that might seem a little strange even here...

I've apparently been putting a whole lot of thought in the last couple of months into the extent to which idealization (or the pairing of idealization/demonization, which are probably different sides of the same coin given how they turn on a dime) is utterly ubiquitous and seems to be extremely bad for good governance. Indirectly, it strongly incentivizes those in power to develop worse epistemics (cover things up, don't ask questions, be easy for others to model) no matter how good they originally were. Now that I've started looking for it, I keep seeing evidence everywhere.

I've gently-but-seriously considered trying out a process loosely based on the one described in this crazy notebooking write-up.

One belief I've had for a while, which might be slightly strange in this group, is that I'm not bothering with cryonics. It seems to break down into 2 factors... 1) I believe almost any post-Singularity "humans" will be radically different to be point of not identifying or being identifiable as the same thing, and even if "I" do make it to the end, I'll quickly modify myself into something neuroticism-free that I can't internally identify with (whether by means of myself, or AI-imposed values, the outcome is the same). Therefore, having my exact mental configuration probably doesn't matter too much. 2) Weird "bug" of prioritizing continuity only of "predictable external identity," while largely deprioritizing or treating as free-variables most of my internal continuity that's independent from that (in other words... I allow myself to make radical mental shifts, so long as I expect to be able to behave similarly, continue to serve my future self, keep my allies, and keep my word).

Probably the single highest "craziness index" idea I've mulled on in the past year is whether I wanted to track and see if there's correspondence between meditational vibrations (and where I "feel" them to be; I have a sense of "location" with them) and Brodmann Areas (or a similar location-numbering system, so that I can leave myself partially-blind on initially assigning them). It's the kind of thing where I expect the original framing to fail, but I also expect to learn something interesting in the process. Settled on "not worth the effort," though.

Most of what I'm thinking about is probably merely eccentric/special-interest... biology stuff, metaphorical correspondences between financial data and ideas from evolution or entropy, how I'm using intuitions/perceptions and getting better at communicating them clearly...

(Plus a fairly typical human baseline: social, emotional, productivity, self-improvement, identity, future planning)

Comment by spiracular on Spiracular's Shortform Feed · 2019-09-17T03:03:26.806Z · score: 7 (4 votes) · LW · GW

One of my favorite little tidbits from working on this post: realizing that idea innoculation and the Streisand effect are opposite sides of the same heuristic.

Comment by spiracular on Spiracular's Shortform Feed · 2019-09-17T02:49:20.467Z · score: 3 (2 votes) · LW · GW

Bubbles in Thingspace

It occurred to me recently that, by analogy with ML, definitions might occasionally be more like "boundaries and scoring-algorithms in thingspace" than clusters per-say (messier! no central example! no guaranteed contiguity!). Given the need to coordinate around definitions, most of them are going to have a simple and somewhat-meaningful center... but for some words, I suspect there are dislocated "bubbles" and oddly-shaped "smears" that use the same word for a completely different concept.

Homophones are one of the clearest examples; totally disconnected bubbles of substance.

Another example is when a word covers all cases except those where a different word applies better; in that case, you can expect to see a "bite" taken out of its space, or even a multidimensional empty bubble, or a doughnut-like gap in the definition. If the hole is centered ("the strongest cases go by a different term" actually seems like a very common phenomenon), it even makes the idea of a "central" definition rather meaningless, unless you're willing to fuse or switch terms.

Comment by spiracular on Looking for answers about quantum immortality. · 2019-09-15T21:09:36.798Z · score: 4 (3 votes) · LW · GW

Relatedly: I would bet someone money that Greg Egan does something insight-meditation-adjacent.

I started reading his work after someone noted my commentary on "the unsharableness of personal qualia" bore a considerable resemblance to Closer. And since then, whenever I read his stuff, I keep seeing him giving intelligent commentary and elaboration on things I had perceived and associated with deep meditation or LSD (the effects are sometimes similar for me). He's obviously a big physics fan, but I suspect insight meditation is another one of his big "creativity" generators. (Before someone inevitably asks: No, I don't say that about everything.)

To me, Egan's viewpoint reads as very atheist, but also very Buddhist. If you shear off all the woo and distill the remainder, Buddhism is very into seeing through "illusions" (even reassuring ones), and he seems to have a particular interest in this.

I can make up a plausible story that developing an obsession with how we coordinate-and-manifest the illusion of continuity from disparate brain-parts... could be a pretty natural side-effect of sometimes watching the mental sub-processes that generate the illusion of "a single, conscious, continuous self" fall apart from one another? (Meditation can do that, and it's very unsettling the first time you see it.).

Comment by spiracular on Looking for answers about quantum immortality. · 2019-09-15T21:07:20.004Z · score: 4 (-3 votes) · LW · GW

So, here's the specific thing I can think of that seems like it might be helpful...

I try to be cautious about using meditation-based wire-heading or emotional-dulling, but at minimum, there's a state one step down from enlightenment (equanimity) that perceives suffering as merely "dissonance" in vibrations. The judging/negative-connotation gets dropped, and internal-perception of emotional affect is pretty flat (Note of caution: the emotions probably aren't gone, it's more like you perceive them differently. I'm not 100% sure how it works, myself. While it might sound similar, it's not quite the same as dissociation; the movement is more like you lean into your experience rather than out of it. Also, I read in a paper that its painkiller properties are apparently not based on opiods? Weird, right? So neurologically, I don't really know how it works, although I might develop theories if I researched it a bit harder.).

Enlightenment/fruition proper doesn't even form memories, although I've never been able to sustain that state for longer than a few seconds. But when it drops, it usually drops back into equanimity... so I guess between the two, it'd be a serious improvement on "eternal conscious suffering"?

Unfortunately, to get into Enlightenment territory, there's a series of intermediate steps that tend to set off existential crises, of widely-varying severity. Any book or teacher that doesn't take this and the wireheading potential seriously, is probably less good than one who does. That said, I still recommend it, especially for people who seem to keep having existential crises anyway. But it's a perception-alteration workbench; its sub-skills can sometimes be used to detrimental ends, if people aren't careful about what they install.

Comment by spiracular on Looking for answers about quantum immortality. · 2019-09-15T21:06:43.824Z · score: 2 (2 votes) · LW · GW

Here's one plus-side that you don't need the additional context to understand: I kinda suspect that at least most people would eventually find the right combination of insights and existential-crises to bumble into enlightenment by themselves, if they had an eternity of consecutive experiences to work with. Especially given that there seem to be multiple simple practices that get around to it eventually (although it might take a couple of lifetimes for some people).

Comment by spiracular on Looking for answers about quantum immortality. · 2019-09-15T21:06:24.769Z · score: 2 (2 votes) · LW · GW

As someone who has had stream-entry, and the change-in-perception called Enlightenment... I endorse your read of it as being potentially useful in this case?

I'm going to give more details in a sub-comment, to give people who are already rolling their eyes a chance to skip over this.

Comment by spiracular on Spiracular's Shortform Feed · 2019-09-12T18:59:45.116Z · score: 6 (4 votes) · LW · GW

While I could rattle off the benefits of "delegating" or "ops people", I don't think I've seen a highly-discrete TAP + flowchart for realizing when you're at the point where you should ask yourself "Have my easy-to-delegate annoyances added up to enough that I should hire a full-time ops person? (or more)."

Many people whose time is valuable seem likely to put off making this call until they reach the point where it's glaringly obvious. Proposing an easy TAP-like decision-boundary seems like a potentially high-value post? Not my area of specialty, though.

Comment by spiracular on Bioinfohazards · 2019-09-11T02:57:32.073Z · score: 15 (8 votes) · LW · GW

Now that we've gone over some of the considerations, here's some of the concrete topics I see as generally high or low hazard for open discussion.

Good for Open Discussion

  • Broad-application antiviral developments and methods
    • Vaccines
    • Antivirals proper
    • T-cell therapy
    • Virus detection and monitoring
  • How to report lab hazards
    • ...and how to normalize and encourage this
  • Broadly-applicable protective measures
    • Sanitation
    • Bunkers?
  • The state of funding
  • The state of talent
    • What broad skills to develop
    • How to appeal to talent
    • Who talent should talk to

Bad for Open Discussion

These things may be worth specialists discussing among themselves, but are likely to do more harm than good in an open thread.

  • Disease delivery methods
  • Specific Threats
  • Specific Exploitable Flaws in Defense Systems
    • Ex: immune systems, hospital monitoring systems
    • It is especially bad to mention them if they are exploitable reliably
    • If you are simultaneously providing a comprehensive solution to the problem, this can become more of a gray-area. Partial-solutions, or challenging-to-implement solutions, are likely to fall on the bad side of this equation.
  • Much of the synthetic biology surrounding this topic
  • Arguments for and against various agents using disease as an M.O.
Comment by spiracular on What are the biggest "moonshots" currently in progress? · 2019-09-07T20:33:19.138Z · score: 6 (3 votes) · LW · GW

Warning: Big pile of text, Decomposing definitions

I notice that there's a lot of disputably-relevant axes for assessing if something is a moonshot, and that was making me hesitant to answer. So... I'm going to be "that guy" who deep-dives defining terms. Hopefully this will be constructive?

Here are some disputable axes for assessing "moonshots" that popped to mind:

  • How possible is it to profit off of incremental progress on this project's sub-goals?
    • At the limit: Does it have to be completely unprofitable until you win, after which there is a steep, step-like function after which there are massive returns?
      • But even space wasn't this extreme! There were rockets and missiles before there were rocket ships. And my loose understanding of the way Elon Musk is doing things is about as incremental as you can make space rockets and still have it be rocket science (admittedly: not very). Those many fancy landing-control tests don't make the final launch not a moonshot, though.
    • At lower levels: At what point is this just the kind of "profitable-incremental-progress via selection" algorithm that capitalism encourages and supports just fine?
  • How severe was the projects "start-up cost"? How much hard-to-consolidate infrastructure, intelligence, data, and resources were required before this project was even conceivable?
  • For one reason or another, is the most plausible counterfactual that if this one group wasn't doing this, no group would be doing this? Or would be doing this much more poorly?
  • Classically, megaprojects refer to large-scale works of architectural infrastructure, but we live in an information age. Do massive broadly-beneficial projects of informational infrastructure count, or not?
  • How much of humanity needs to be impacted by the outcome of this project? Does the outcome need to be beneficial? Or is this really about affecting the narrative humanity has about itself, and not the effect it has on the day-to-day way that people experience their lives? (ex: space projects)

(And... interesting! ryan_b selected a almost completely different set of axes. Maybe it's not a particularly consistently-defined concept?)

I feel conflicted about including any of these as requirements (several don't actually seem that desirable), but I think the normal way moonshots are thought of and defined tends to center around analogy to the Space Race. And therefore, tends to involve almost all of these features being present.

"No incremental progress measures" seems like a particularly key part of the definition, and yet a potentially negative thing to filter by. Whenever you can, good incremental progress assessment is usually a positive thing to add to a project.

Thought experiment: If you had to compare 2 identical cold-fusion projects, one of which came up with a bunch of intermediate steps and tested them, and one which didn't and just had one big "did you get everything right?" assessment right at the end... which one is the moonshot? But which one is probably the better project?

Under this lens, there's a pretty important distinction to be made between things that are being treated as moonshots, and problems that have to be approached as moonshots.

Maybe the right question is... "What are moonshot problems that someone is seriously tackling?" Or just dropping the moonshots framing, and getting a list of interesting megaprojects. Or just picking the 1-2 axes you most care about, and sorting on them explicitly.

(FWIW; any one of those is a valid thing to want, and it's a good question! That is part of why I put in the effort to try to break it down.)

P.S. Is there some way I should have used the Question-on-a-Question / Related Question function to do this? If so, could someone walk me through how that's supposed to work?

Comment by spiracular on What are the biggest "moonshots" currently in progress? · 2019-09-07T20:21:01.395Z · score: 6 (3 votes) · LW · GW

It's not quite moonshots, but here's wikipedia's list of Megaprojects.

When it comes to more classic architectural infrastructure projects, I'd be shocked if China (and maybe Singapore) don't have several. But I also expect them to lean towards the incremental-progress-is-visible model where they can. Because when you can, having assessable incremental progress really is almost always the better choice?

Serious New Physics seems to be full of nightmare-to-fund megaprojects. What's that upcoming telescope that's going to be able to test if weird extrasolar asteroids like 'Oumuamua passing by Earth are a common or rare event? Somebody's got to be building a new and more powerful collider, somewhere in the world. Who's working on fusion reactors? Who's working on crafting quantum computers?

Thinking of megaprojects & moonshots got me thinking of some information-infrastructure stuff that might or might not count, depending on definition. So I'm going to throw on a list of some tentative and disputable "Information Age Megaprojects" too...

  • Wikipedia
    • Someone back-of-envelope estimated that Wikipedia was the culmination of 100 mil hours of work in 2008, and assuredly more by today.
    • While it's success feels predestined nowadays, when it first outdid formal and privately-funded attempts at online encyclopedias, that was a very surprising outcome to many
  • If you loosen the restrictions, biology probably has tons of megaprojects. (BLAST, GenBank, PubChem... NIH is funding and maintaining vast databases, and all the infrastructure that makes them navigable. How much work that is probably depends on whether you include or exclude the work it took to extract the data they're peddling.)
    • It feels almost-incoherent for me to think of biological projects as "moonshots"? There are interesting and novel things being done constantly (Many of them clever, irreplacable things. Some of them fundamental things! Some of them potentially high-impact!), and yet the term just does not seem to fit. Just about every interesting synthetic biology project nowadays has a "merely" moderately-large to large start-up cost, a measure of incremental progress, and a huge element of chance. Some of those could have huge impacts if they work. Should I call those moonshots? They're not really big enough for "megaprojects" to fit, cost-wise.
  • Google (Alphabet? GoogleX?) seems to have a taste for this
    • Google's book-scanning project (before it was shut down)
    • Loon is their internet-access balloon project
    • The Android Operating System?
    • The Google search-engine itself?
  • The Internet Archive?
  • People building 3D models of entire cities, in games such as Second Life?
  • Does Bitcoin count?
    • I suspect not, but I couldn't really articulate why. If I wanted to argue one way or another with this one, I actually wouldn't know where to start.
Comment by spiracular on What are the biggest "moonshots" currently in progress? · 2019-09-07T20:14:38.097Z · score: 1 (1 votes) · LW · GW

I'm used to thinking of those more as "Megaprojects". "A List of Megaprojects" is a great goal, but I feel like it might be worth clarifying that in the question up-top a little bit.

Comment by spiracular on Paper Trauma · 2019-08-25T18:53:52.934Z · score: 2 (2 votes) · LW · GW

Single datapoint, but... I love markdown for notes. The formatting shorthands are great, and an automatically-generated Table of Contents is a nice plus.

It hasn't been easy for me to find anything that offered both flow-chart diagramming and cloud-sync, though. And both markdown and HTML are bafflingly awkward to make tables with (Why can't somebody just set up a painless csv wrapper? Or if that exists, please tell me?).

Most decent markdown editors are also LaTeX capable, which I probably use even more than diagrams. There's a bit of a learning-curve for that, but nowadays I can type out most equations without so much as looking at the keyboard (let alone looking up symbols).

Comment by spiracular on Why is the nitrogen cycle so under-emphasized compared to climate change · 2019-08-14T07:02:28.109Z · score: 2 (2 votes) · LW · GW

But when the argument for the alternative boils down to "eat shit, not chemicals"...

I'm kidding, but only slightly. Organic fertilization is a bit gross, and I think most food companies would prefer to not be associated with any of dead leaf matter, rotting leftovers, or manure.

Counterargument: Composting totally became a thing, and that potentially puts the grossness right in your backyard.

(Huh. Composting is certainly something people do on an individual level to micro-combat the usage of nitrogen fertilizers, with probably a very negligible effect. And some people do seem dedicated to it. But I suspect that if you asked most people who do it, they would claim it's about landfills or something, not soil nitrogen content.)

Comment by spiracular on Why do humans not have built-in neural i/o channels? · 2019-08-14T05:39:20.493Z · score: 24 (6 votes) · LW · GW

Even if it had evolved, any detailed form of communication that had the potential to transmit hard-to-break imperatives is something you want to be very, very careful with.

Defection, manipulation, and novel avenues for disease-transmission or parasitism heavily disincentivize this. It's intuitively "gross" for a reason.

TL;DR: Infections and defections would probably utterly wreck this. The blood-brain barrier exists for a reason. While we did get language, in other ways we've evolved specifically to hide information from each other; it's not straightforwardly evolutionarily favored. Large clusters of highly-related organisms have more incentive to do this (bacterial mats, ants, our own cells, etc.), and the information-bandwidth they share with each other through pheromones and chemical signals is actually pretty staggering. But at a glance, I do think they pay a cost in increased (and more elaborate) avenues for manipulations and infections to reap the benefits of this privilege.

Edit to add: Linking some additional strongly-related articles! SSC's Maybe Your Zoloft Stopped Working Because a Liver Fluke Tried To Turn Your Nth-Great-Grandmother into a Zombie and the paper it centers on, Invisible Designers: Brain Evolution Through the Lens of Parasite Manipulation

Isolating Brains from Infection

It bears pointing out that evolution usually seems to want the opposite of this for our centralized-decision-maker organs. Notice that instead of making the brain more open-access over time, most species went the direction of making it isolated from even our own bodies, using things like the heavy-filtering brain/blood barrier. The risk of that sensitive organ getting poisoned, diseased, or biologically manipulated was just too high to risk it.

(Nematodes make what humans call "embodied thinking" look like a joke; the serotonin from their digestive tract is felt by their brains directly.)

People used to die in droves (and at young ages!) of measles, tuberculosis, and a million other things. Even before cities, herd-living put us under immense pressure to develop a pretty intensely specialized immune system. If we had to worry about giving a disease a highway to our central nervous system, that would be... very, very bad. It might even guarantee that such a species would never get to centralize such things at all, such a force is the risk of infection.

Not to even get into the possibility of physical brain-hijackings by concepts (like memes, but oh-so-much-worse!), or even just catching communication-transmitted Kuru... but here's a pretty vivid speculative description of just how bad being an evolved "open book" that granted others write-access would probably get.

And with regards to the "benefits" of open communication -of information conveyed in a language that's very hard to fake- we do still have some information transmitted in body-language and words. That certainly captures some of the benefit. But it bears pointing out that we're a species that un-evolved any obvious presentation of whether a female is in estrus, and has very strong inhibitions around trying not to gain information from each other's body odor. "Complete, total honesty" is not something evolution typically selects for, and it didn't veer entirely that way for us. Even in the less-cutthroat modern era (at least, compared to our distant savannah past), Greg Egan's Closer feels like a pretty realistic depiction of how we might feel about it if we ever did fully share our mental experience with even one another person. I'll avoid spoiling it too badly, but we'd probably quickly uncover a lot of things about one other that we wished we didn't know.

Some Living Approximates

Bacterial mats, giant networks of fungi, and eusocial insects with strong genetic kinship might have strong enough evolutionary incentives for this to line up, although higher relatedness actually exacerbates the infection concern. And between the cells of multicellular creatues, certainly quite a lot does get communicated. Many of these examples do seem to "share their mind" in at least some meaningful sense. They transmit a lot of information and orders to one another, and have a communal decision-making process at varying degrees of centralization/decentralization. Pheromones for insects, various signalling secretions from bacteria, the oodles of transactions and deliveries between our cells at every moment... combined these can be very high-bandwidth. Almost incomprehensibly so, if you've ever seen attempts to measure and chart such things.

Ants could practically be said to have a pheromone "language," complete with clan-identification tags. And as a way to selectively trigger a highly-specific neural pathway, or activate a known set of behaviors in a conspecific with the same brain-configuration, pheromones are not a bad way to go? The behavior patterns pheromones set off can get oddly specific at times.

And... ants also get tricked by pheromones into feeding the brood parasites that eat their own young. And what we call "bacterial sex" (high-bandwidth communication of DNA?) is actually virus-esque plasmids trying to transmit themselves to new bacteria, like an infection. Some plasmids might even come with addiction molecules, which is an extra-douchey way for an plasmid to convey "replicate me, or die." And in coordinated bacteria, you do sometimes see defectors. So... it's still pretty manipulable, and it sure gets manipulated.

The more stereotyped behaviors you can set off through external signals, the more you have a "broader attack surface," in the cybersecurity lingo. And biological parasitism is ubiquitous, and fractaline, and adaptive, and uses any damn attack surface it can get.

Humans? A fluke. Parasites are evolution's true darlings.

Comment by spiracular on Why is the nitrogen cycle so under-emphasized compared to climate change · 2019-08-07T22:20:27.702Z · score: 9 (5 votes) · LW · GW

I thought this was an interesting question... although I definitely get the feeling like I'm missing some of the context behind this "planetary boundaries" write-up.

(What is the Stockholm Resilience Center? What are its motives and methods? Why was it doing this analysis, and how did you end up running into it?)

I agree that fertilizer runoff gets talked about a lot less than climate change, and I'm not entirely sure of why that is. I just looked it up, and "Organic"-labeled things apparently do already mandate organic fertilizers (which should be N-neutral on net?). So there's at least that.

Regarding their assessment... one of the factors that seemed to be weighted a lot in their assessment was "level of antropogenic change vs natural variation in Process X." I'd expect that to have heavily-weighted the Nitrogen Cycle, since our interference in it dwarves the variability due to natural processes (something they themselves spell out. The Haber process is a strange, powerful, magical, inorganic thing the humans cooked up).

This metric is... not precisely the same as "level of damage this change can cause." They seem to have set up some sort of threshold for what they consider "dangerously high", and I don't really understand the thought-process or reasoning they used in picking those thresholds. But I think they factored "change vs natural variation" into their thinking a good deal.

My understanding from my own Ag background is that fertilizer runoff really can be a big problem; nitrogen and phosphorus are a major limiting nutrient for plant-life, both on land and in freshwater ecosystems (the ocean surface is iron-limited, interestingly). In a sense, this is exactly why we use it; we want our food crops to grow at a wild rate, one rarely seen in nature.

When there's a sudden influx of nitrogen into a freshwater system, one of the possible consequences is an algal bloom. These go through a boom-and-bust cycle (with the seasons, or resource-availability patterns), leading to a massive algal die-off. During this die-off, decay bacteria wipe out the underwater oxygen-supply, leading to knock-on effects on freshwater ecosystems like massive fish die-offs. Enclosed spaces like lakes are perhaps especially vulnerable, since there is no way for the N and P to ever exit the system, potentially perpetuating the cycle indefinitely.

That said, suddenly cutting fertilizer usage would practically ensure that food production would suddenly drop way down, and that would ill-serve a lot of other human values. Reducing runoff looks like a very hard problem, to me. Finding ways to remove excess P and N from environment seemed to be an area with at least a little bit of interest, but it doesn't seem very actionable on an individual rather than city or state level, which might explain the low publicity? Unsure.

Comment by spiracular on Permissions in Governance · 2019-08-05T23:36:49.293Z · score: 7 (4 votes) · LW · GW

Some rules seem to have an element of "cost of compliance" that relies on being able to predict the future, in a way that even specialists have little hope of doing. Sometimes, this leads to a risk-taker-enriched (aka asshole-filtered) gray zone surrounding a well of value which may-or-may-not have been poisoned (something of a Shroedinger's Well?).

If the gray zone is valuable, then for a while, the market might heavily favor gray-area violators of this type. But occasionally, one of these zones suddenly transforms into a trap for gray-area violators at all levels of competence. At least in theory, I could see some law-makers deploying this variety of illegible rules on purpose, to use that as a wasp-trap for sneaky, competent boundary-pushers. I have very little idea of whether this actually happens on-purpose much, and a lot of things that might initially look like this probably turn out to be "you gotta know a guy" in retrospect.

Some examples I can think of that might fit this pattern...

The question of exactly which financial regulations would be deemed "applicable to Bitcoin" was going to have to be done largely in retrospect, and I think even specialists largely couldn't make reliable predictions about this. On a related note, I know a story where a pre-blockchain gold-backed online currency tried to get a permit and was prohibited, due to regulators deciding that they didn't fall into the relevant reference class. This company was later penalized into bankruptcy for operating without that permit, when a later court ruling decided that they did fall into that reference class.

More broadly, the question of "will rules around patents will be leveraged against you" seems to sometimes fall near this gray-zone. That one's dampening profile is a bit weirder and complicated, though. The dampening-effect there seems to be disproportionately borne by the medium- to well-funded. It seems to reward obscurity, because small corporations are usually not worth going after about patent violations. Medium-sized ones might be, and often settle out of court to avoid a lawsuit, making it profitable to go after them; I would guess that they're the ones penalized the worst by this, but I'm not certain. Top corporations probably fall somewhere in-between; on the one hand, they tend to have good lawyers (repelling frivolous lawsuits), but on the other hand, they might stand to lose a very large sum in court.

Possibly anything where court rulings being made on the same case seem to see-saw back-and-forth as it goes up levels.

Comment by spiracular on Spiracular's Shortform Feed · 2019-07-06T00:39:39.385Z · score: 12 (3 votes) · LW · GW

I stumbled into a satisfying visual description:

3D "Renderers" of Random Sampling/Monte Carlo estimation

This came out of thinking about explaining sparseness in higher dimensions*, and I started getting into visuals (as I am known to do). I've see most of the things below described in 2D (sometimes incredibly well!), but 3D is a tad trickier, while not being nearly as painfully-unintuitive as 4D. Here's what I came up with!

Also, someone's probably already done this, but... shrug?

The Setup

In a simple case of Monte Carlo/random-sampling from a three-dimensional space with a one-dimensional binary or float output...

You're trying to work out what is in a sealed box using only a thin needle (analogy for the function that produces your results). The needle has a limited number of draws, but you can aim the needle really well. You can use it to find out what color the object in the box is at each of the exact points you decided to measure. You also have an open box in which to mark off the exact spot in space at which you did the draw, using Unobtainium to keep it in place.

Each of your samples is a fixed point in space, with a color set by the needle(function/sampler's) output/result. These are going to be the cores that you base your render around. From there, you can...

Nearest Neighbor Render

For Nearest Neighbor (NN), you can see what happens when you situate colored-or-clear balloons at each of the sample points, and start simultaneously blowing up all of them up with their sample point as a loose center. Gradually, this will give you a bad "rendering" of whatever the color-results were describing.

(Technically it's Voronoi regions; balloons are more evocative)

Decision Trees

Decision trees are when you fill in the space using colored rectangular blocks of varying sizes, based on the sample points.


Most other methods involve creating a gradient between 2 different points that yielded different colors. It could be a linear gradient, an exponential gradient, a step function, a sine wave, some other complicated gradient... as you get more complicated, you better hope that your priors were really good!

Neural Net

A human with lots of prior experience looking at similar objects tries to infer what the rest looks like. (This is kinda a troll answer, I know. Thinking about it.)

Sample All The Things!

3D Pointillism even just sounds like a really bad idea. Pointillism on a 2D surface in a 3D space, not as bad of an idea! Sparsity at work?


The Blind Men and an Elephant fable is an example of why you don't want to do a render from a single datapoint, or else you might decide that the entire world is "pure rope." If they'd known from sound where each other were located when they touched the elephant, and had listened to what the others were observing, they might have actually Nearest Neighbor'd themselves a pretty decent approximate elephant.

*Specifically, how it's dramatically more effort to get a high-resolution pixel-by-pixel picture of a high-dimensional space. But if you're willing to sacrifice accuracy for faster coverage, in high dimensions a single point's Nearest Neighbor zone can sloppily cover a whole lot of ground (or multidimensional "volume").

Comment by spiracular on Spiracular's Shortform Feed · 2019-07-02T21:53:15.787Z · score: 5 (3 votes) · LW · GW

I have this vague intuition that some water-living organisms have fuzzier self-environment boundaries than most land organisms. It sorta makes sense. Living in water, you might at least get close to matching your nutritional, hydrational, thermal, and dispersion needs via low-effort diffusion. In contrast, on land you almost certainly need to create strongly-defined boundaries between yourself and "air" or "rock", since environments can reach dangerous extremes.

Fungi feel like a bit of an exception, though, with their extracellular digestion habits.

Comment by spiracular on Spiracular's Shortform Feed · 2019-07-02T21:53:00.136Z · score: 8 (4 votes) · LW · GW

If The Ocean was a DM:

Player: I want to play a brainless incompletely-multiceullular glob with no separation between its anus and its mouth and who might have an ability like "glow in the dark" or "eject organs and then grow them back." It eats tiny rocks in the water column.

The Land DM: No! At least have some sensible distribution-of-nutrients gimmick! Like, you're not even trying! Like, I'm not anti-slimes, I'll allow some of this for the Slime Mold, but they had a clever mechanic for succinctly generating wider pathways for nutrient transfer between high-nutrition areas that I'd be happy to...

The Ocean DM: OMG THAT'S MY FAVORITE THING. Do you have any other character concepts? Like, maybe it can also eject its brain and grow it back? Or maybe its gonads are also in the single hole where the mouth and anus are?

The Land DM: ...

Comment by spiracular on Unrolling social metacognition: Three levels of meta are not enough. · 2019-06-19T17:14:46.660Z · score: 9 (2 votes) · LW · GW

...huh. I guess I know of one particular variety, and that variety is very self-contained and circling-adjacent (I almost could have called it "Narrative Circling", if that didn't seem like such a contradiction-in-terms). But from the wiki article, T-Group appears to refer to a more nebulous and broad category of things, some of which seem not nearly so self-contained.

The thing I had run into functioned basically as described here (scroll down for the written description). This read to me as clearly cicling-adjacent, and I didn't think all that hard about where it had come from.

The wikipedia description struck me as surprisingly uninformative about the details of the practice itself. But from poking around a bit on the internet just now... I get the impression that T-Group can refer to something similar to what I described, but can also be used to refer to something close to an experimental leadership/decision-making structure that uses the "T-Group" as part of their intragroup conflict-resolution method?

I knew that the variety I had run into was a bit homebrew, and probably had aspects of circling bred into it. I don't think I appreciated just how different it could be from other people's usage of/context for the term. That said, I do see some signs of shared lineage.

The techniques feel related, and the facilitating ethos of awareness, learning, honesty, and goallessness feels similar. But the variety I ran into felt more tightly-defined and compartmentalized, and I was mostly doing it with strangers.

I admit that with a high bar of trust and decently committed participants, I could actually see it working well as a social-information-gathering method? But the idea of being dragged into doing it with coworkers, or of treating it like a primary conflict-resolution technique, seems quite troubling to me.

Comment by spiracular on Unrolling social metacognition: Three levels of meta are not enough. · 2019-06-17T19:42:19.097Z · score: 3 (2 votes) · LW · GW

Circling is working on a similar problem, and training capacities that are used as workarounds: This feels true to me.

I think this even more visible in the circling-variant called "T-Group," where people tell a short narrative on why they think they're having a described emotional reaction. Very frequently, the explanations encapsulate 1-2 layers of meta, or explicitly gesture at certain/uncertain pieces of common knowledge.

(ex: Joe is responding to Jane's response and feels U, Jane is responding to Jane's interpretation of Joe's response to Jane and feels V. Albert notes X, models that Betty would be troubled by X, and feels concern that there might not be common knowledge of this. Betty believes that there is common knowledge of X, and that everyone feels wary about it. Betty queries if anyone disagrees with that interpretation. Carrie notes that she hadn't initially been aware of X, but is now aware of X, and feels sad and a little scared.)

When I look at it this way, it becomes even clearer why T-Group comes with an exhortation to always include the experiencer as an object in your sentence ("I feel", "I make it mean", "I infer"). If the next person is going to do meta on your meta, it helps if they don't need to recalculate out the layer representing "you," and it's useful to explicitly differentiate between yourself and common knowledge.

(Actually, the more I think about it, the more T-Group looks like a hybrid between circling and explicit modeling. And the fact that they work well together suggests to me that not everything in the circling skillset gets eclipsed when you switch to explicit-modeling.)

Comment by spiracular on Naked mole-rats: A case study in biological weirdness · 2019-06-17T17:15:19.994Z · score: 4 (3 votes) · LW · GW

Ticks just repeatedly break my intuitions. How does something that small and R-selected have a 2-year-long lifecycle?

Oh goodness yes, fungi. Two-nuclei sexual stages startled me when I first learned of it. It's a highly-diverse and successful clade, filling in an array of niches all across the specialist/generalist spectrum, and ranging anywhere from unicellular to syncytial to organisms the size of a city. Plus, many of them seem to manifest that evolutionary pseudo-"inventiveness" that I usually associate with bacteria.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-15T02:10:41.393Z · score: 10 (5 votes) · LW · GW

Yeah... this is reading as more "moralizing" and "combative" than as "trying to understand and model my view," to me. I do not feel like putting more time into hashing this out with you, so I most likely won't reply.

It has a very... "gotcha" feel to it. Even the curiosity seems to be phrased to be slightly accusatory, which really doesn't help matters. Maybe we have incompatible conversation styles.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-15T00:58:14.264Z · score: 14 (4 votes) · LW · GW

(Meta-note: I spent more time on this than I wanted to. I think if I run into this sort of question again, I'm going to ask clarifying questions about what it was that you in particular wanted, because answering broad questions without having a clear internal idea of a target audience contributed to both spending too much time on this, and contributed to feeling bad about it. Oops.)

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-15T00:57:24.857Z · score: 1 (3 votes) · LW · GW

Personally, one of the most damning marks against the physical-distance intuition in particular is its rampant exploitability in the modern world, where distance is so incredibly easily abridged. If someone set up a deal where they extract some money and kidnap 2 far-away people in exchange for letting 1 nearby person go, someone with physical-distance-discounting might keep making this deal, and the only thing the kidnappers would need to use to exploit it is a truck. If view through a camera is enough to abridge the physical distance, it's even easier to exploit. I think this premise is played around with in media like The Box, but I've also heard of some really awful real-world cases, especially if phone calls or video count as abridging distance (for a lot of people, it seems to). The ease and severity of exploitation of it definitely contributes to why, in the modern world, I don't just call it unintuitive, I call it straight-up broken.

When the going exchange rate between time and physical distance was higher, this intuition might not have been so broken. With the speed of transport where it is now...

Maybe at bottom, it's also just not very intuitive to me. I find a certain humor in parodies of it, which I'm going to badly approximate here.

"As you move away from me at a constant speed, how rapidly should I morally discount you? Should the discounting be exponential, or logarithmic?"

"A runaway trolley will kill 5 people tied to the train track, unless you pull a switch and redirect it to a nearby track with only 1 person on it. Do you pull the switch?" "Well, that depends. Are the 5 people further away from me?"

I do wonder where that lack-of-intuition comes from... Maybe my lack of this intuition was originally because when I imagine things happening to someone nearby and someone far away, the real object I'm interacting with in judging /comparing is the imagination-construct in my head, and if they're both equally vivid that collapses all felt physical distance? Who can say.

In my heart of hearts, though... if all else is truly equal, it also does just feel obvious that a person's physical distance from you really should not affect your sense of their moral worth.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-15T00:50:11.815Z · score: 3 (2 votes) · LW · GW

I think you're inferring some things that aren't there. I'm not claiming an agent-neutral morality. I'm claiming that "physical proximity," in particular, being a major factor of moral worth in-and-of-itself never really made sense to me, and always seemed a bit cringey.

Using physical proximity as a relevant metric in judging the value of alliances? Factoring other metrics of proximity into my personal assessments of moral worth? I do both.

(Although I think using agent-neutral methods to generate Schelling Points for coordination reasons is quite valuable, and at times where that coordination is really important, I tend to weight it extremely heavily.)

When I limit myself to looking at charity and not alliance-formation, all types of proximity-encouraging motives get drowned out by the sheer size of the difference in magnitude-of-need and the drastically-increased buying power of first-world money in parts of the third world. I think that's a pretty common feeling among EAs. That said, I do conduct a stronger level of time- and uncertainty-discounting, but I still ended up being pretty concerned about existential risk.

Comment by spiracular on Naked mole-rats: A case study in biological weirdness · 2019-06-14T19:36:05.278Z · score: 9 (4 votes) · LW · GW

The strange pressures of their subterranean lifestyle, which eukaryote described somewhat, probably covers most of it. Inbreeding/isolation is probably the other half of the puzzle.

I'll try to show how some of these traits tie in with low-oxygen and subterranean living, in those places where it wasn't already covered.

  1. A lot of these do bottom out to the pressures of creating a large nest, and dealing with an underground low-oxygen environment.

Eusociality and large protective communal nest-building mesh together really well, which I think fed into a lot of the items in the Richard Alexander list of predictions mentioned above (the accuracy is really impressive!).

Lack-of-hair and weird teeth seem pretty obviously developed for digging/crawling lifestyle. Acid-pain immunity has been proposed to be a consequence of having to handle an otherwise-intolerable level of lactic acid buildup ('sore muscles') while digging in low-oxygen zones. The strange metabolic properties (which probably feed into cancer resistance considerably) also seem to be a way to handle their lower-oxygen-availability lifestyle; endothermy can be surprisingly energy-intensive.

Really, living a low-temperature low-sugar careful-energy-usage low-ambient-cell-replication lifestyle tends to defend against cancer and improve lifespan quite a bit in general (ex: calorie restriction for mammals often extends lifespan, raising fruit flies at low temperatures can full-on double it). Molerats seem to have been under heavy pressure to biologically enforce a strict energy-usage regimen, and they take this to an incredible extreme. So you'd expect to see some cancer resistance, although it's still crazy in terms of degree.

(I'd totally buy that they probably have some additional nutty things going for them. I think I've heard a theory that they have unusually-stringent cell checkpoints pre-division?)

  1. Inbreeding and genetic isolation

No, really. It's both a strong push towards kin-selection (a basis of eusociality if there ever was one), and an exacerbator of genetic drift. The changes might not always be anywhere near this favorable, but isolation still tilts things towards the weird, and the faster rate to saturation increases your ability to build adaptations on top of adaptations. (see also: high weirdness for species living in islands, caves).

Comment by spiracular on Naked mole-rats: A case study in biological weirdness · 2019-06-14T17:00:19.383Z · score: 10 (5 votes) · LW · GW

Adding a couple of mine to this list, will probably add more as I think of them...

Invertebrates: Sacculina (parasitic barnacle that hormonally manipulates crabs), Sawflies (primitive hymenopterans with caterpillar-like larvae, some love forest fires), Ticks, Phengaris arion

Vertebrates: Chameleon, Hoatzin, Bats (just... bats)

Plants: Amorphophallus genus (The entire stalk is a single giant leaf. It sheds that leaf every few years, puts up a giant flower, then sheds that and goes back to growing another single giant leaf?), Orchids, Magnolia (ancient monocots that convergently evolved to look like dicots, flowers that predate bees)

I'm giving partial-points to: scale insects, fig wasps, termites, Neotrogla curvata, Pandoravirus, Neuroptera

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-14T01:50:30.493Z · score: 10 (4 votes) · LW · GW

Out-of-context quote: I'd like for things I do to be either ignored or deeply-understood and I don't know what to do with the in-betweens.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-14T01:17:48.863Z · score: 7 (3 votes) · LW · GW

After seeing a particularly overt form of morality-of-proximity argued (poorly) in a book...

Something interesting is going on in my head in forming a better understanding of what the best steelman of morality-of-proximity looks like? The standard forms are obviously super-broken (there's a lot of good reasons why EA partially builds itself as a strong reaction against that; a lot of us cringe at "local is better" charity speak unless it gets tied into "capacity building"). But I'm noticing that if I squint, the intuition bears a trace of similarity to my attitude on "What if we're part of an "all possible worlds" multiverse, in which the best and worst and every in-between happens, and where on a multiversal scale... anything you do doesn't matter, because deterministically all spaces of possibility will be inevitably filled."

Obviously, the response is: "Then try to have a positive influence on the proximal. Try to have a positive influence on the things you can control." And sure, part of that comes from giving some weight to the possibility that the universe is not wired that way, but it's also just a very broadly-useful heuristic to adopt if you want to do useful things in the world. "Focus on acting on the things you can control" is generally pretty sound advice, and I think it forms a core part of the (Stoic?) concept of rating your morals on the basis of your personal actions and not the noisy environmental feedback you're getting in reply. Which is... at least half-right, but runs the risk of disconnecting you from the rest of the world too much.

If we actually can have the largest positive impact on the proximal, then absolutely help the proximal. This intuition just breaks down in a world where money is incredibly liquid and transportable and is worth dramatically more in value/utility in different places.

(Of course, another element of that POV might center around a stronger sense that sys1 empathetic feelings are a terminal good in-and-of-itself, which is something I can't possibly agree with. I've seen more than enough of emotional contagion and epidemics-of-stress for a lifetime. Some empathy is useful and good, but too much sounds awful. Moving on.)

I guess there needs to be a distinction made between "quantity of power" and "finesse of power," and while quantity clearly can be put to a lot more use if you're willing to cross state (country, continent...) lines, finesse at a distance is still hard. And you need at least some of both (at this point) to make a lasting positive impact (convo for later: what those important components of finesse are, which I think might trigger some interesting disagreements). Finding specialists you trust to handle the finesse (read: steering) of large amounts of power is often hard, and it's even harder to judge for people who don't have that finesse themselves.

And "Finesse of power" re: technical impact, re: social impact, re: group empowerment/disempowerment (whether by choice or de-facto), re: PR or symbology... they can be really different skills.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-13T22:07:47.994Z · score: 3 (2 votes) · LW · GW

Good to know! I'm assuming that's mostly an automatic consequence of everything being a comment, but that also fits with what I'd want from this sort of thing.

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-13T21:56:59.665Z · score: 6 (4 votes) · LW · GW

Damn it, Oli. Keep your popularity away from my very-fake pseudoprivacy :P

(Tone almost exactly half-serious half-jokingly-sarcastic?)

((It's fine, but I'm a little stressed by it.))

Comment by spiracular on Spiracular's Shortform Feed · 2019-06-13T21:49:02.622Z · score: 3 (2 votes) · LW · GW

This is pasted in from a very different context, so it doesn't match how I would format it as a post at all. But it paints a sketch of a system I might want to flesh out into a post sometime (although possibly not on LessWrong): Predictiveness vs Alignment as sometimes-diverging axes of Identity.

There's probably some very interesting things to be said about the points where external predictiveness and internal alignment differ, and I might be a bit odd in that I think I have strong reason to believe that a divergence between them is usually not a bad thing. In fact, at times, I'd call it a legitimately good thing, especially in cases where it feeds into motivation or counterbalances other forces.

Ex: For me in particular, anti-identifying as "sensitive" is load-bearing, and props up my ability to sometimes function as a less-than-excessively-sensitive human being. However, the sensitive aspect is always going to show through in the contexts where there's no action I can take to cover for it (ex: very easily disturbed by loud noise).

There's still some question in my head about what internal alignment even is. It feels to me like "Internal Predictiveness" is a subcomponent of it, and aspiration is another part of it, but neither describes it in full. But it felt like there was a meaningful cluster there, and it was easier to feel-out and measure than it was to describe, so for now I've stuck with that...

Q: What are some identity labels that apply to you? How aligned with or alienated from them do you feel?

I was only able to do this exercise if I separated out "Externally Predictive" (P-, NP, and P+) and "Internal Alignment," (AN-, N/A, AG+, and sometimes 404). These sometimes feel like totally divergent axes, and trying to catch both in a single axis felt deeply dishonest to me. Besides, finding the divergence between them is fun.

(I apologize for the sparse formatting on this; I'll try to find a way to reformat it better later.)

Predictive-First Identities

  • Nerd & Geek
    • Predictive (P+), Aligned (AG)
  • Rationalist
    • P+, AG+
  • Biologist
    • P+, full range from AG+ to bitterly AN- (Alienated)
  • Programmer
    • Moderately P+, N/A
  • Extroverted Geek
    • P+, Highly AN
  • Introvert
    • Slightly P- (used to be P+), AG
  • Anxious
    • Slightly P+, Slightly AN

Alignment/Alienation-First Identities

  • Healthy Boundaries
    • P+, AG+
  • Visual Thinker
    • NP to P+, AG+
  • Queer
    • NP (by design!), AG
  • Smart
    • P+, 404 (I can't properly internalize it in a healthy way unless I add more nuance to it)
  • Not Smart/Good Enough but Getting Better by Doing it
    • NP to Moderately P+, AG+
  • Sensitive
    • P+, AN
  • Immature
    • NP (used to be P-), AN
  • Self-centered
    • P-, AN-
Comment by spiracular on Spiracular's Shortform Feed · 2019-06-13T20:44:07.308Z · score: 14 (5 votes) · LW · GW

Thinking about value-drift:

It seems correct to me to term the earlier set of values a more "naive set.*" To keep a semi-consistent emotional tone across classes, I'll call the values you get later the "jaded set" (although I think jadedness is only one of several possible later-life attractor states, which I'll go over later).

I believe it's unreasonable to assume that any person's set of naive values are, or were, perfect values. But it's worth noting that there are several useful properties that you can generally associate with them, which I'll list below.

My internal model of how value drift has worked within myself looks a lot more like "updating payoff matrices" and "re-binning things." Its direction feels determined by not purely by drift, but by some weird mix of deliberate steering, getting dragged by currents, random walk, and more accurately mapping the surrounding environment.

That said, here's some generalizations...


  • Often oversimplified
    • This is useful in generating ideals, but bad for murphyjitsu
  • Some zones are sparsely-populated
    • A single strong value (positive or negative) anywhere near those zones will tend to color the assumed value of a large nearby area. Sometimes it does this incorrectly.
  • Generally stronger emotional affects (utility assignments), and larger step sizes
    • Fast, large-step, more likely to see erratic changes
  • Closer to black-and-white thinking; some zones have intense positive or negative associations, and if you perform nearest neighbor on sparse data, the cutoff can be steep


  • Usually, your direct experience of reality has been fed more data and is more useful.
    • Things might shift so the experientially-determined system plays a larger role determining what actions you take relative to theory
      • Experiential ~ Sys1, Theoretical ~ Sys2, but the match isn't perfect. The former handles the bulk of the data right as it comes in, the later sees heavily-filtered useful stuff and makes cleaner models.**
  • Speaking generally... while a select set of things may be represented in a more complete and complex way (ex: aspects of jobs, some matters of practical and realistic politics, most of the things we call "wisdom"), other axes of value/development/learning may have gotten stuck.
    • Here are names for some of those sticky states: jaded, bitter, stuck, and conformist. See below for more detail.
    • Related: Freedomspotting

** What are some problems with each? Theoretical can be bad at speed and time-discounting, while Experimental is bad at noticing the possibility of dangerous unknown-unknowns?

Jaded modes in more detail

Here are several standard archetypical jaded/stuck error-modes. I tend to think about this in terms of neural nets and markov chains, so I also put down the difficult-learning-terrain structures that I tend to associate with them.

Jaded: Flat low-affect zones offer no discernible learning slope. If you have lost the ability to assign different values to different states, that spells Game Over for learning along that axis.

Bitter: Overlearned/overgeneralized negatives are assigned to what would otherwise have been inferred to be a positive action by naive theory. Thenceforth, you are unable to collect data from that zone due to the strong expectations of intense negative-utility. Keep in mind that this is not always wrong. But when it is wrong, this can be massively frustrating.

(Related: The problems with reasoning about counterfactual actions as an embedded agent with a self-model)

Stuck: Occupying an eccentric zones on the wrong side of an uncanny valley. One common example is people who specialized into a dead field, and now refuse to switch.

Conformist: Situated in some strongly self-reinforcing zone, often at the peak of some locally-maximal hill. If the hill has steep moat-like drop-off, this is a common subtype of Stuck. In some sense, hill-climbing is just correct neural net behavior; this is only a problem because you might be missing a better global maximum, or might have different endorsed values than your felt ones.

An added element that can make this particularly problematic is that if you stay on a steep, tiny hill for long enough, the gains from hill-climbing can set up a positive incentive for learning small step size. If you believe that different hills have radically different maximums, and you aspire for global maximization, then this is something you very strongly want to avoid.

*From context, I'm inferring that we're discussing not completely naive values, but something closer to the values people have in their late-teens early-twenties.

Naive modes in more detail

Naive values are often simplified values; younger people have many zones of sparse or merely theorized data, and are more prone to black-and-white thinking over large areas . They are also perhaps notable sometimes for their lack of stability, and large step-size (hence the phrase "it's just a phase"). Someone in their early 20s is less likely than someone in their 40s to be in a stable holding pattern. There's been less time to get stuck in a self-perpetuating rut, and your environment and mind are often changing fast enough that any holding patterns will often get shaken up when the outside environment alters if you give it enough time.

The pattern often does seem to be towards jaded (flat low-affect zones that offer no discernible learning slope), bitter (overlearned/overgeneralized negatives assigned to what would otherwise have been inferred to be a positive action), stuck (occupying eccentric zones on the wrong side of an uncanny valley), or conformist (strongly self-reinforcing zones, often at the peak of some locally-maximal hill).

Many people I know have at east one internal-measurement instrument that is so far skewed that you may as well consider it broken. Take, for instance, people who never believe that other people like them despite all evidence to the contrary. This is not always a given, but some lucky jaded people will have identified some of the zones where they cannot trust their own internal measuring devices, and may even be capable of having a bit of perspective about that in ways that are far less common among the naive. They might also have the epistemic humility to have a designated friend filling in that particular gap in their perception, something that requires trust built over time.

Comment by spiracular on Global insect declines: Why aren't we all dead yet? · 2018-04-08T03:25:02.680Z · score: 8 (3 votes) · LW · GW

Part of why this is so confusing to me is that if you take some of these invasive-like species, and introduce them to new habitats, they DO cause huge problems (see: Kudzu, certain species of mussel... I could go on and on). And due to combinations of things like "toxins" or "low nutritional content" or various really good counter-herbivore adaptations, I actually wouldn't be that surprised if only 1 or 2 things in their native environment actually subsist predominantly off of the invader, or at least eat it at such a level that it would kill large numbers of them.

Comment by spiracular on Global insect declines: Why aren't we all dead yet? · 2018-04-08T03:22:21.097Z · score: 11 (3 votes) · LW · GW

Entomology bacchelor here. My naiive model would have been that termites and earthworms, followed maybe by "the entirety of all pollinators", would have major effects on ecology at the macro-scale. But my default model also has comments along the lines of "there are species that will trigger black swan effects, and you can't always predict which ones they are." And that later part of me is very confused.

The model is less that every single species matters, and more that there are "keystone species": the often-highly-specialized regulator (a predator, parasitoid, or lethal disease) of a toxic/high-reproductive-rate/invasive-like species, where if that invasive species were to be left unchecked, it would dominate the environment in a manner detrimental to just about everything else living there. See: Otters that kill sea urchins which would otherwise detatch kelp from the sea floor, or... in a case closer to what I'd expect here, things like small wasps that specifically kill a beetle that would otherwise kill large numbers of trees.

Comment by spiracular on Global insect declines: Why aren't we all dead yet? · 2018-04-08T02:41:47.371Z · score: 1 (1 votes) · LW · GW

On a related vein of thought, one can almost model any currently-visible adults of a species with a high potential number of offspring-per-adult as a "surplus" that doesn't necessarily have a large effect on trophic "throughput". Phytoplankton come to mind as the extreme case; they sequester a huge quantity of carbon dioxide, but the oceans aren't green (usually...) because huge quantities of them are constantly sinking to the ocean floor or being eaten. That small surviving fraction still reproduces at a high enough level to maintain themselves. (Land plants seem to have a very different equilibrium, which probably has something to do with... better herbivore control by predators, and maybe also counter-herbivore adaptations and the necessity of infrastructure-deployment to handle water scarcity? Not especially confident on this.)

Insects don't have the reproductive rate of phytoplankton, though. And from the other comments, it sounds like this really is starving out some members of higher tropic levels.

Comment by spiracular on Global insect declines: Why aren't we all dead yet? · 2018-04-08T02:08:41.870Z · score: 1 (1 votes) · LW · GW

For the former (German study with the biomass numbers), perhaps? For the review study (which reported on worldwide species numbers and Britain range distributions), I flipped through the supplement and it looks like they actually did try very hard to adjust for the fact that they were working off of volunteer data (Filtered for surveys that had identified certain other similar species, and which had lasted for over 1hr. Also, excluded some species that they expected to have "gone into hiding" more). Something interesting might be going on with the Beetles, but it seemed suggest that Butterfly/Moth and Wasp/Bee/Ant species are almost all either holding steady, or dropping. I think significant reductions in flying insect catches might actually be plausible.