[How dangerous is your world generally, if I try to do something like getting lunch without a psion protecting me?  Without the ability to coordinate, it doesn't seem like it should be able to protect itself against the kind of violence represented in Civ, since protection is a non-excludable good at that scale.  Is there a powerful mage that protects this whole city just to be nice, or...?]

[You are not especially likely to be mugged or attacked in this neighborhood but I suppose it seems likely that we have a higher crime rate than dath ilan. There's a psion who works in crime prevention but stuff is seldom escalated to him if it's anything short of arson or murder, since he's trying to cover a lot of people, so if you were lined up for a more minor crime then just being about to have a check-in with me would be your backup plan there, but it is not generally understood to be a particularly dangerous activity for people to go out and get lunch and go shopping in Manhattan.]

[I think I might be wiser to hide in your apartment of lower-environmental-complexity until I understand this world better.  There's just too much that isn't compressible and hits me in specifically low probability densities of my model, too many places where I'm stupid and not just ignorant.  Places where my model makes a routine strong confident prediction about how a human civilization at this tech level functions, and then yours does something wildly different, often not in a good way from my standpoint.  Your world's language has a shorter word for "genocide" than "cooperation", for example, and I would not have expected that to be the case for humans.  I might be having an easier time in some ways if I'd landed on a visibly alien world with crystal things or pulsating blobs.]

[Sorry for the lack of crystalline entities, I guess. Do you expect to need anything I haven't mentioned?]

[If the local environmental invariants like air oxygenation and thermal range stay within their previous range of variance, I should survive ten hours.  I hope I'm less stupid and bothersome when you come back.  Oh - you should either show me how to operate local microwaves and refrigerators, or trust to my ability to deduce that safely from whatever the Network claims when a weirdly stupid person enters search terms.]

[The refrigerator doesn't require operation, you just open it and then you close it when you're done and it will beep at you if you fail to close it, but fair enough on the microwave.] She demonstrates the microwave once she's put her bowl in the dishwasher, and then shows Thellim how to get to Wikipedia (the current featured article is about a historical painter) and Google (she Googles an arithmetic problem by way of example, and then at semirandom "Guam", to demonstrate the principle).

"Thank you again.  I'll try not to be even more of a bother to you."

"Don't worry too much about it, you're interesting." Isabella collects her cane and her coat and heads out.

Thellim gives her an Earth-wave-goodbye with one hand and a dath ilani handgesture with the other.

All right, step one in the investigation plan, see if Thellim can catch up on any more sleep.  This is going to be cognitively demanding and should be done on maximum available sleep!

All right, step one worked.  Now it's time for THINKING.  If she wants to actually do this correctly instead of turning into Science Maniac Verrez, she should actually list out all the anomalies and questions worth investigating, before she starts; then pick a first target based on its tractability X importance, instead of blindly maximizing one or the other; then try to come up with her own hypothesis before she starts looking at the data, so she knows what it is that she's learning and where her prior theoretical mistakes were.

Item one:  EVERYTHING IN THIS WORLD IS WRONG AND MAKES NO SENSE.

Item two:  The world doesn't appear to run on math!  Not that there's anything wrong with that.  Lots of things don't run on math!  Thellim personally knows several logical inconsistencies and they've always been very nice impossibilities.

Item two point one:  This world has many precogs, which seems to imply that the possibility that Thellim is in the top layer of reality right this second is correspondingly small.

Thellim is strongly tempted to go over on Wikipedia and look up this issue right now, except for the enthusiasm-draining strong prediction that Wikipedia isn't going to know anything about it either.

Item three:  This is apparently what happens to you when you die under circumstances where you could not possibly be cryonically preserved.  Thellim is surprised, on reflection, that there has not been more speculation about this topic inside her own Civilization.  Perhaps it is considered an infohazard, since even thinking about this for ten seconds is enough for Thellim to start seeing ways that thinking about this topic could drive people insane.  Maybe 'Wikipedia' has articles about that topic, since nobody here seems to have the concept of information with negative value to imperfectly rational agents, possibly because they've never invented the idea of rational agency in the first place or realized that information is normally supposed to have positive value.

...well, it's sort of obvious how those three issues fit together, isn't it?  This reality apparently possesses a capability of internal division that lets it obtain 'precogged' answers and outcomes from sub-layers of reality; metaphorically and perhaps literally, it's a complexity-theoretic-superior computer to the physics of Thellim's home.

Some layer of that reality contains either the original of Thellim's universe or an isomorphic copy of its physics, the same way that Isabella's universe contains all the sublayers that get precogged.

And even the top layer of Isabella's world is a sublayer of some other region of this hierarchically(?) divisible reality; and the layers above Earth somehow put the lunar eclipses into place, as an equilibrium of unknown other factors and forces and optimization pressures there.

Then, some rule of some layer above both Earth and dath ilan in the hierarchy, fired to copy Thellim from dath ilan into Earth...

Is any of this concentrating probability, though?  Or did she just come up with a hypothesis isomorphic to "Anything can happen?" and start coding out the anythings by hand instead of compressing them?  It sounds like mostly that, so far.

Still, when your previous hypothesis was Stupid in the sense of definitely prohibiting materializing from dath ilan into Earth, there is something to be said for jumping to a hypothesis of greater ignorance which allows more things to happen.  And there is also something to be said for allowing that hypothesis of ignorance to be specific and have plausible-sounding specific structure inside it, just in case it does start concentrating probability or suggesting an agenda of investigation.  It could be, for example, that 'Wikipedia' will have something to say about the precog-supporting sub-layers of reality, and that some experimentally-confirmed theory about them will look like definite affirmation that there could be a copy or original of dath ilan inside one.

Also, Thellim is starting to notice that, if she switches her attention from nonfiction to fiction, there is a strangely large amount of fiction that would prepare somebody to analyze her situation, if she died under circumstances not permitting cryonic preservation, and woke up in another world with higher information-theoretic complexity (implying more destinations than source events, which is why her transportation here would be rare as seen by the destination).  Were those novels Keeper-influenced, encoding hidden advice from Keepers secretly meant to help people like her?  "Be very careful about which nonhuman superpowers you bargain with or give your allegiance to" and "nonhuman superpowers may not respond to your requests the way you expect, not because they lack information about you, but because they have inhuman utility functions over responding to requests" are both very common themes in Portal Fantasies, now that she thinks about it.  Those are both pieces of advice that she might have guessed for dead people, if she was guessing blind.  None of that advice seems obviously relevant to her own actual situation; but it wouldn't be surprising if the Keepers guessed wrong, guessing that blindly.

...This mostly feels like the wrong thing to consider right now, even if it is in some sense the most important.

Go back to breadth-first.  What, specifically, in this world is wrong and makes no sense?  What is every anomaly Thellim can remember seeing since coming here?

1.  Her literal materialization in somebody's home.  Actually, that it was inside somebody's home and not over the middle of the ocean is sufficiently improbable-given-randomness to be worth noting itself.

2.  The home contained two males in a couple, something far more frequent here than in dath ilan; it would later be claimed to her that sexual attraction on Earth doesn't go by reproductive potential of couplings at all, in defiance of the evolutionary equilibria that are supposed to be where biology and biological cognition originally come from.

3.  Thellim tried to speak a different language to show that she was from outside their world; which didn't work, in retrospect, according to later info, because Earth had more than one widely used language; the first of many, many cases of Earth being way off the multiagent-optimal boundary despite all the theorems showing that shouldn't happen for intelligent agents in general.

4.  Thellim was contacted by a voice in her head, because Brian and Jackson would have contacted Isabella later, and Isabella is a precog...

Thellim is starting to notice an isomorphism between the set "anomalous events" and the set "all events since landing on Earth".

The thought of trying to list out everything that happened to her in order is making Thellim's brain feel tired, even if it would be the methodologically optimal thing to do.  Maybe she'll just forgive herself for being less perfect, and do this more sloppily.  Yes, even on this irreplaceable first-pass-at-the-problem where her thoughts are fresh and hindsight traps are minimal.  There's a reason why Thellim's original employment wasn't as a scientist; she didn't want that much rigor in her daily mental life.

Okay, rougher view of the anomalousness of this world.  There's all sorts of places where pieces of Earth ought to be generated by an optimized equilibrium, observable facets of reality where the derivatives of some fitness gradient should have come to an approximate rest along most dimensions as pushed by an optimization process.  Like natural selection, and sexual desire; or offered economic goods, and people selecting where to spend their money.  Then instead there's sex-indifferent sexual attraction, or airplanes with terrible food and nonsensical advertisements.

An obvious thought is that the layer of reality she's now inside didn't generate those features by tracing a long causal history to find equilibria, and instead just... filled things in, sort of.  Like an author making up features of a book, only not quite like an author, because a human author paying attention and trying for realism wouldn't have non-upgradable bad food on an expensive airplane trip.  The people inside the world are allowed to notice these facts, but not allowed to realize that they're anomalous, or they've come up with overfitted hypotheses that manage to explain what should not be explainable.

Alternatively, another obvious thought is that there's perfectly sensible reasons for everything 'anomalous', and Thellim just doesn't know what they are.  Like with 'subways' seeming at first to be moving much too many people from one place and time to another, but hiddenly, subways just make additional stops... actually that still doesn't make very much sense?  There were still way too many people, even if they were going to twenty destinations instead of one destination.  But it makes twenty much times as much sense as before Isabella explained that part to Thellim, and maybe that reflects a predictable direction of updating where Thellim could learn more and more facts and things would stop looking out-of-equilibrium.

This seems like a good general point to try to narrow down first, if Thellim isn't going to do the proper, rigorous, mental-effort-intensive, slow process of collecting all her data and considering all her questions before investigating any.

In particular - though maybe this is just recency bias with respect to what she's thought about last - the couple-sex-mix frequencies seem most likely to have an explanation that is in Wikipedia, if events in this world have explanations at all.  Economics is complicated and changes rapidly as lots of agents compete to participate in more gainful-trades.  Sexual attraction moves more slowly over evolutionary time and is easier to study.

Also, the pattern of romantic couplings generates a huge fraction of all the goodness/fun/utility in Civilization, so even here somebody should be studying that?  But no, you could say the same thing or even more so about everybody's genetic material, and here that is considered a bad thing to try to optimize.  Well, either Wikipedia will have something to say about it, or there will be no articles in Wikipedia on sexuality; and later Isabella will explain that one time a scientist wondered out loud what makes two people fall in love, and then everybody on that continent murdered each other, after which nobody ever asked the question again.

Actually, no, if that scenario is true, the whole story will be on Wikipedia!  Because Earth doesn't infohazards!

Thellim takes a private moment to complain to herself about all the dubious looks Isabella gives her, every time Thellim mentions Civilization being grownup enough to do something that Isabella's version of 'civilization' can't manage to do without people randomly killing each other, for what doesn't even sound to Thellim like reasons, and it is obvious to her which of these two ways of organizing a planet is doing better, and it's the one that can manage to subsidize childcare for unusually smart or healthy people without that turning into immediate mass murder for reasons that Thellim still cannot begin to understand even after being told the end result in this world, but fine.

And Civilization isn't a monoculture either!  Just because their world can manage to coordinate on anything does not make it a monoculture!  Thellim bets that her Civilization has managed to produce one thousand times as much informed analysis of the cost-benefit tradeoffs of how much exploration and exploitation to do in particular dimensions of society!  And then meta-optimize their world for the correct amount of diversity versus optimization!  But Isabella would just point to that as even more proof that dath ilan is a monoculture!  When what they actually have is a planet whose qualities vital to life and happiness are distinguishable from noise!

She's getting sidetracked, here.  Probably because Isabella is hosting her and being incredibly generous to a stranger without - apparently, this part itself is bizarre - any expectation that Thellim will be able to parlay her knowledge into trillionaire status and repay her.  But sometimes Isabella acts disapproving, and doesn't seem to want to clarify their relationship with more explicit terms; which makes Thellim feel unsure whether maybe the support gets shut off, if she says the wrong thing.  Thellim is at least temporarily dependent on Isabella, right now, more than she's been dependent on anyone since she was eight years old and passed the tests for initial-level economic self-determination.  Which means that Thellim has things she wants to say back to Isabella on certain subjects, but has been suppressing herself from saying, because it may or may not be safe; which means those rejoinders keep running through her mind, a loop of cognition that can't complete itself.

Thellim acknowledges this feeling; tells the feeling that it will have a chance to speak again later, when she understands her relationship with Isabella more solidly; and turns her attention back to analyzing Earth.