Friday, December 15, 2017

The Worst Argument Against Being Cautious About AI

Note: throughout this article when I use the term "AI" I'm talking about superhuman artificial intelligence, and I'm assuming the counterarguments are referring to same.

There is a species of article or blog post that basically boils down to "People who worry about AI as an existential risk are being irrational for reasons the follies of human psychology, and/or overarching principle linking intelligence and moral behavior." (I won't bother linking to one, you can do a quick web search and fine a lot of them.) In other words, they're attacking a straw-man argument that states: the reason to worry about AI is that it will necessarily be malevolent toward humans.

A side comment often appended to these arguments make is that to believe AI will be malevolent toward humans is to be comically anthrocentric. While this observation is quite correct, misunderstanding this point is exactly why these arguments are so wrong. You have to be comically anthrocentric to think that we are immune from the disruptions caused by superintelligence. That is to say: for AI to hurt us, it doesn't have to come after us specifically. It just has to not specifically care about us.

Case in point: settlers from Europe in the Americas were not malevolent toward the native humans, for the most part. They just found arable land that they wanted, and the didn't especially care who was already there - resulting in plague and death and the loss of many cultures and languages. (That was just between two flavors of humans where one side had about a four thousand year technological head start.) In the same way, the developers at the edge of my city don't have anything special against the deer and coyotes and worms in the soil - but the spreading pavement and light and traffic and noise has affected them just the same.

To repeat: for AI to hurt us, all that has to be true is that it does not care about us. It doesn't even have to have any special malevolent intentions.

There are projects that are essentially trying to solve morality before the first AI goes online, for exactly this reason. I wish them luck, but we've had a few dozen centuries on this project and not gotten very far.

Wednesday, December 13, 2017

San Diego from Space, By Balloon

High altitude balloon launches are cool, especially when you can clearly see the San Diego coastline. It always amazes me that in the distance you can drive in a half hour, you're in space. (Now where's my check from the Round Earth Conspiracy! Come on guys!)

Friday, November 24, 2017

The Hidden, 1987

I always thought The Hidden was an unjustifiably forgotten scifi action movie from the 80s, and this is reinforced by the positive reviews it got and still gets. Basically, imagine the Terminator meets Fallen - except instead of a time-traveling unstoppable cold-blooded robot pursuing/being pursued by a good human from the future and shooting up Los Angeles in the process, and a demon possessing suspects and detectives, they're space-traveling unstoppable aliens (one cold-blooded and one nice) chasing each other and shooting up Los Angeles, and taking over suspects' and detectives' bodies. It's more smartly written and filmed than you might expect from that description (I still think the intro is quite clever.) And, the good alien is played by Kyle MacLachlan, and one of the evil-alien-possessed-people by Claudia Christian, before she was in Babylon 5. You may also recognize Ed O'Ross, who showed up in a number of action/scifi movies in the 80s and 90s. A good number of one-liners are had by all, with a possible explanation toward the end of the movie of what's happened to our political system recently.

As a side note, that narration on trailers from 80s can sound so dated makes me realize how very old I truly am.

The Hidden Car Chase (1987) from Sergei Nazarov on Vimeo.

Sunday, October 29, 2017

First Interstellar Asteroid? It's Interstellar, But Not the First We've Seen

Information here and here. Based on the velocity and path, this asteroid originated from outside the solar system. This is a great additional finding, but not actually news! Comet Wild-2 was the subject of the Stardust sample return mission, and analysis showed more than a few interesting things: that it contained the amino acid glycine, and that the nitrogen isotope ratio showed that the object likedly originated from a different solar system.

A point of interest here is that since the solar system's origin, there must have been multiple close passes by other stars - close enough that our respect Oort clouds would mix at the margins, and material would be exchanged between star systems. We have now verified this logical inference visually, and through direct chemical evidence.

Previous post about alien evolution, Life's Origins at Four Billion Years Ago; Implications For Our Future

Friday, October 27, 2017

If You Take Parfit Seriously, You Should Commit Yourself To Creating Superintelligence

Cross posted at Cognition and Evolution and The Late Enlightenment.

Derek Parfit makes the argument that if utilitarianism as it is commonly understood is to be taken to its conclusion - the greatest good for the greatest number - that mathematically we should care not just about making individuals happy, but making more individuals, to be happy. If you can have a world of a billion people all just as happy as a world of a million people, then that that's a no brainer.

The problem is when you get to the math of it. The "repugnant conclusion" that if the total amount of happiness is what matters, then you should favor numbers over quality of life. That is, a world of a hundred billion people with lives just barely worth living is better than a world of a hundred people with great lives - because the great lives are probably not a billion times greater than those of the hundred billion in almost total misery.

The obvious objection is that you're talking about theoretical people when you talk about those hundred billion. The counterargument is that we do care about theoretical people - our descendants - and you might already make environmental decisions to preserve the environment for the happiness of your grandchildren; right now you avoid (hopefully) littering the street to avoid upsetting people you've never met and will probably never meet.

There are other objections of course; for instance, that experienced happiness in an individual is what matters; otherwise slave plantations could be (in fact, probably are) morally acceptable.

But following Parfit's repugnant conclusion to its end, if the total amount of utility is what matters, then increasing the amount of utility possible to be experienced also matters. That is to say, there is no reason to stop at considering theoretical people, but rather we should consider theoretical kinds of experience, and theoretical kinds of experiencers. And there is nothing in Parfit's thesis provincial to or chauvinistic about humans. (If there were, that might solve the problem, because you could say "the closer something is related to me, the more I should be concerned with its happiness" - me and my brother against my cousin, et cetera - which, at very close genetic distances, is in fact what most humans already do.)

Therefore, we should try to make a world of a hundred million bipolar (manic) people who can experience hedonic value far in excess of what most of us ever do (assuming we can keep them manic and not depressed.) Or, even better, created an artificial superintelligence capable of experiencing these states, and not devoting all our resources to creating as many copies of it as possible. But cast aside those constraints - if you believe it is possible for a self-modifying general artificial intelligence with consciousness (and pleasure) to exist, then by Parfit, the only moral act is to give up all your recreation and resources to live in misery and dedicate your life to the single-minded pursuit of getting us one second closer to the creation of this superintelligence. The total suffering and happiness of life on Earth up until the moment of the singularity would quickly shrink to a rounding error, compared to the higher states these replicating conscious superintelligences might experience. Therefore, if you are not already singlemindedly dedicating yourself to bringing such a superintelligence to life, you are forestalling seconds of these agents' pleasurable experiences (which far offset your own suffering and maybe those of all living things) and you are committing the most immoral act possible.

This problem is superficially similar to Roko's basilisk (in the sense of your actions being changed by knowledge of a possible superintelligence) but I think it should still be called Caton's basilisk.

As a result of these objections, I do not think we need to take the repugnant conclusion seriously, and I do not think not dedicating yourself to creating a super-hedonic superintelligence is immoral.

Tuesday, October 24, 2017

Organics on Ceres Are From Ceres (not from other impacting bodies)

The organic material on Ceres, while intriguing, appears to be native, rather than delivered from other impactors. So says data from the Southwest Research Institute at the 2017 Astronomical Society meeting. The possibility of simple organic replicators on low-gravity bodies in the solar system ("space viruses", to be dramatic) an interesting one, and is one form (or one part) of the pan-spermia hypothesis that's been considered for over a century, going back at least to Arrhenius. (Space viruses might also be the only evidence we would ever see of alien life or even an alien singularity.) What this tells us is that the large majority of material on Ceres, and presumably on most large old asteroids, is native to those bodies since the dawn of the solar system.

What the findings mean for the "space virus" hypothesis is that we can be more confident that Ceres is not crawling with foreign space viruses - although if there is a replicator that can use the typical organics on large asteroids as building materials, that's not what you would usually see. That is to say, when an organism gets infected by a virus, the organism isn't infiltrated with foreign matter, but rather with a tiny bit of foreign matter that then rearranges the atoms in the organism into copies of itself.

Saturday, October 7, 2017

Review, Blade Runner 2049 (Containing Many Spoilers)

tl;dr It's good, worthy of the original, with a much cleaner script, and what shortcomings it does have it shares with the original. Go see it now, on the big screen.


The cinematography (given other reviews to this effect I need not belabor this point) and the sound, which was fantastic in the first one and is a hallmark of Villeneuve's other work. It's not surprising that a director who clearly focuses on being a good story-teller - to the narrative flow - also pays so much attention to the sound in his movies. In at several interviews the actors described him as tending toward verbal sound effects in his direction. Roger Deakins should get the Oscar, [added later: he did] and you should watch this on the big screen.

In his movies, Villeneuve always goes for the emotional jugular, usually with the death of loved ones and harm being done to children, both happening here. From a distance it seems like manipulation but it feels wonderful while you're watching.

I really liked the shot looking down onto the city during the initial approach to Los Angeles. You see the detailed metal surface of the city, reminiscent of Latin American shanty-towns on the mountains surrounding the city centers - and then, best of all, you realize that's just the roof of packed high-rises, and you catch glimpses of neon down to the streets. And then there are the police and corporate towers above it all, implying the class stratification as in the first one - i.e., if you're not cops (or corporate) you're little people. This movies continues the trend of moving away from CGI and building real sets and doing real stunts (used to excellent effect in Mad Max: Fury Road.)

I liked the transition from slave to working skinjob, and the attendant shift social status represented as out of the frying pan and into the fire in a sense. In the first one after the term skinjob is introduced, in the narrated version we hear "O'Brien is the kind of man in history books used to call black men n-----s." In this one, skinjobs are walking around free but clearly despised by police and civilians alike. This closely parallels the status change of African-Americans from slave to free but living under Jim Crow and segregation.

Also a clear parallel - the police chief ("Madame") has more than a hint of plantation owner or county sheriff to her, and treats K as her boy. In the scene where she gets a little drunk there is clearly sexual tension and she talks to K teasingly in a way I doubt she would talk to her human employees. (She also at one point asks "How long have you been under me?")

The introductory scene continues the moral ambiguity of Blade Runner vs replicant. In the first one, it's "Did you ever take that test yourself Mr. Deckard?" that starts many viewers wondering. In the very beginning, a hard-working farmer (producing food for an overpopulated world and otherwise minding his own business) enters his kitchen, there's a thug sitting in the shadows who coldly dispatches him.

I like that we get out of LA in this one. If the cities are so miserable, now we understand why people aren't just moving elsewhere. If you've made the LA-Vegas drive on the 15, you've seen the single solar tower surrounded by mirrors, and in this movie we see a whole plain covered with them (I imagine Villeneuve doing his research by driving or flying around the desert and saying, "What is that? That's real, right now in 2017?") I like that LAX is now separated from the rest of LA by a fair stretch of Pacific and that there's a massive tide wall protecting LA. In general I appreciate movies that really use the places as they are instead of a made-up generic American city. In this one we see maps and landmarks - except during the visit to San Diego, which is seldom featured in movies. (Incidentally, that's exactly how I pictured the Dog Solitudes from Gibson's Mona Lisa Overdrive.)

They eliminate the question of whether K is a replicant very early on, which I like. They then play with the "are the memories real or not?" quite a bit. Favorite scene in this regard: K gets angry when he sees memory woman, and she assumes it's because he is finding out his memories are false, but the truth is much more complicated. And I'm still not clear that K really isn't the male twin who survived.

We still don't know if Deckard is a replicant or not. Villeneuve says in so many words that they were quite careful to avoid answering this.

We see a big Atari logo, but there's no explanation at all as to the resurrection of this company from 2017 to a corporate giant. In general the movie stays away from explaining how we get from 2017 of today to 2019 of the movies, and that's not what's important, so I'm glad they don't get bogged down in it. There continues to be an 80s flair to the styling, especially the cars (K's car looks like an old Toyota with flimsy plastic over it) although it's clearly shown to be a Peugot. (I wonder if companies actively fought to keep their logos out of this movie, given the fortunes of those prominently featured in the first.)

The Blackout is a way to keep the movie from being over in five minutes ("ah, I'll just go on Google 2049, and voila, there's Deckard!") but to the movie's credit, they don't belabor it. I had previously worried that this movie would suffer from the same problem that all paradigm-shift-reliant science fiction does, but I think I was correct that they aren't bogged down by it. the last one?)

Sylvia Hoeks said in one interview that the out-of-place crying was her idea and that Villeneuve trusted her to go with it. (Evan Rachel Wood has to do something like this in Westworld for her character Dolores and likens it to the "acting Olympics.") This is her first American movie, so I'm sure part of it is her showing off her acting skills to future directors. Even if this is confusing or even showboating, I don't mind, because we in the audience get to benefit from her performance.

The parallelism between Deckard's rejection of neo-Rachel and K's rejection of the not-true-Joi hologram is nice.

The fight scene and choking of Luv is genuinely disturbing.

The use of monumental architecture gives the movie a bit of an interesting 70s scifi feel (think of the concrete bunker where memory woman lives.) I wonder if they built this or just used some building on UC Irvine's campus, like everyone does.

There is rare use of humor, just enough to keep the movie from being too heavy or pretentious ("You must have been adorable" to the Wallace clerk.)

NEGATIVE or NOT FULLY UNDERSTOOD - the more I think about them, the less important they are.

There's maybe just a little too much correspondence between characters in the first and second movies, although many are hybrids. Memory girl is J.F. Sebastian (a replicant subcontractor who keeps to herself and has a genetic defect that keeps her from going off-world.) Wallace is Tyrell. Robin Wright is O'Brien. Blonde replicant girl is Pris. The man running the salvage-orphange is half Chu (the eye guy) and half Taffy Lewis (incidentally, I didn't like his performance - too over the top crazy.) K is a combination of Roy Baty and Deckard, Luv is a combination of Roy Baty and Rachel - but much more Roy. (Sylvia Hoeks didn't make up lines for Luv like Hauer did for Roy, but she did take liberties with emotional expression, i.e. the random flat crying.) She also auditioned by recording herself delivering Roy Baty lines. And she certainly perfected the facial-expression-of-naughty-child-while-doing-horrible-murderous-things look that Hauer used for Roy. You could say Joi is this movie's Rachel, but not really - she's the one that most does not fit.

One of the things that made Blade Runner unique was its depiction of 2019 Los Angeles as a multiethnic mess. When I first watched it I was an easily scared East Coaster, and I signed on to the sentiment that if there are that many Asians in an American city and white people eating with chopsticks, it must be a dystopia! Now I watch the first one and think, I wonder if that noodle stand is any good? This tapped into the growing fear at the time that Japan would take over the world. Today, implying a dystopian tone is much better done with climate change and much less with ethnic mixing. Therefore, oddly, in this one, there are remarkably few Asians (or non-white people for that matter - did they all go off-world?) and furthermore, the use of language doesn't make sense from a world-building perspective (Cyrillic on the farm modules; Devanagari in the police station; Korean on the casino.)

David Bowie was originally supposed to play Niander Wallace. That would've been perfect. I kind of wonder at Jared Leto as a choice. His dedication to the role can't be questioned, but I found his slow, staggered delivery a bit contrived and self-conscious.

I really liked some Robin Wright scenes, like the drunk one I mentioned. But some of her dialogue was clunkily written. "You've bought yourself a war." Sticks out to me; no good way to deliver it.

In the first one I wondered what the point was of using emotion to distinguish human and replicant. Because they then become dangerous? Pris was supposed to be part of a kick murder squad! In this movie, the unclear point is why being "born" makes such a distinction. But you can only explore the dimensions of the human condition so much in one movie.

Everyone who has seen Her will think of that movie during the sex scene, and I think the scene in Her was more poignant and better done - although technically I did really like the video effect here. That said, it still felt very crowbarred-in. The function this scene serves in the plot is to increase the emotional resonance of K's relationship with Joi, as well as to get a tracer into K's pocket, but this could have been accomplished in other ways.

I could've used less of the screen investigations (looking at bone fragments, looking at DNA.)

Where did Gaff's accent and use of cityspeak go?

There has already been discussion of another movie. Alien: Covenant has extras that strongly suggest continuity between Blade Runner and Alien (i.e. Weyland's mentor was Tyrell) but this has several problems: a) by 2049 there are at least 9 off-world colonies. In Prometheus, fifty years after BR2049, there are none. b) The late 21st century citizens of Prometheus seemed much happier than the people in Blade Runner. c) This also implies continuing between Blade Runner and Predator. d) There has even been implied continuity between two PKD-inspired movies, Blade Runner and Total Recall. What a mess! I hope that Villeneuve or his descendants don't think about this at all, but if these movies are commercially successful, I can easily see pressure being applied to make this into a parallel to the Marvel Universe. Which will not likely benefit the quality of these movies.

Friday, September 29, 2017

Life's Origins at Four Billion Years Ago; Implications for Our Future

A group from the University of Tokyo (Tashiro et al, 2017) argues in a Nature paper that carbon isotope ratios in rocks in northern Labrador, Canada means that those rocks harbored life almost four billion years ago. This pushes back the early bound on origin of life almost two hundred million years, almost to the Hadean eon. To be sure this finding has not been universally accepted, but it's worth thinking about what it would mean. In particular, and perhaps not coincidentally, this is also right about when the Earth's surface transitioned from molten to solid.

A recent paper reconstructing the last universal common ancestor (LUCA's) genome from a massive tree of millions of genes showed that it was pretty clearly a sulfur-vent organism. This is good news if you're looking for life on Europa or Enceladus, because that means that life on Earth didn't need the sun (and neither would any life that could evolve along vents under Europa's icy crust.) If you assume that the chance of life evolving by 3.5 billion years ago on Earth was 50%, and that the chance of life evolving is based on surface area, and all other things are equal (admittedly speculative when we don't even have all the information for our N=1) then there is a one in three chance of life on Europa. (If that probability correlates instead with the volume of water, then it was overwhelmingly more likely for life to evolve on Europa!)

[Added several days later: someone has finally run the numbers. A model of RNA polymer formation by Pearce et al suggests that the first RNA world molecules were most likely to have formed in small surface pools rather than sulfur vents - but even earlier, 4.17 billion years ago. If a wet-dry cycle is needed, this suggests ocean worlds like Europa are less likely than once-wet places with exposed land like Mars. The lesson of this paper is that you need puddles, not bone-dry deserts or world-spanning oceans. In this model, a world with puddles and organics seems all but certain to develop into an RNA world. A paper by Cardenas et al from the Geological Society of America Bulletin strongly suggests that 3.5 billion years ago, Mars was exactly the kind of place to have puddles. The logical argument is that life, or at least an RNA world, also developed very quickly there, and we should look for similar deposits to the ones found by Tashiro et al. If Pearce's argument does not produce findings like Tashiro's on Mars, we at least can start looking for differences in the early environments of the two.]

Two things to keep in mind about the LUCA paper: 1) LUCA is the last universal common ancestor. There could be a long lineage before it; and 2) the smaller and simpler a system, the more profound the changes possible in that system. If at one point Earth was an RNA World, molecular clock techniques developed based on modern DNA metabolism would probably be pretty bad at retrodicting LUCA. That two hundred million year gap map be exactly that. All that carbon might be free-floating ribosomes, or peri-biotic viroids.

Even more importantly, this has implications for the likelihood of the evolution of life. This discovery should worry you if you consider the Great Filter. The idea is that it seems very likely that life would evolve anywhere there's liquid water. Yet the universe is not obviously filled with intelligent life. Something is therefore stopping the progression from the evolution of life, to that life spreading from its home planet. (This is typically assumed to be some natural event and need not be some science fiction plot of an alien menace stamping out intelligence wherever it appears.) And every time that the origin of life is pushed back a bit further - that gives greater cause to worry, because where probabilistic events are concerned, the faster something happened, the more likely it was. If this paper is correct, then life on Earth appeared essentially as soon as the surface cooled from magma to solid. [Added several days later:

The real question is whether the Great Filter is behind us (we're freaks that got more complicated than algae) or in front of us (every intelligence is powerful but short-sighted and wrecks its own ecology before it can escape its home planet.) Therefore, a very reassuring discovery would be simple life - the local flavor of blue-green algae - under the ice Europa of Enceladus,* and in the ancient mud of dried Martian riverbeds, and baked into Venusian bedrock. That would mean that somehow, we got past the gate - still no guarantees, but we already passed the filter. This would mean that if we do manage to get out of the solar system, we'll find a lot of alien bacterial mats, but no alien minds. Boring? That idea is actually quite reassuring.

On the other hand, a bad discovery would be mass fossil beds of complex multicellular things (like the radioactive squid in Europa Report), especially ones with extrasomatic adaptations (tools.) We have had a number of landers on Mars and Venus, and none of them captured any obvious macroscale life. But a positive finding by SETI would be even more harrowing, especially because it's unlikely that there would be only one other intelligence that happens to be even within a million years of our technology - even if they're within 1% as old as we are, that's a gap of 40 million years in either direction! In such a situation we would have to include they must be legion. In such a situation, we would have to reason: we can hear them, but for some reason they never get away from their home planet - and we are unlikely to be any different.

*If indeed we believe that Enceladus only formed in the Cretaceous, then there is much less likely to be life there than Europa, and we should focus on Europa.

Previous post about alien evolution, Vast Cool and Unsympathetic: Other Worlds Detecting Earth

Benjamin T. Cardenas, David Mohrig, Timothy A. Goudge. Fluvial stratigraphy of valley fills at Aeolis Dorsa, Mars: Evidence for base-level fluctuations controlled by a downstream water body. GSA Bulletin, 2017; DOI: 10.1130/B31567.1

Pearce BKD, Pudritz RE, Semenov DA, Henning TK. Origin of the RNA world: The fate of nucleobases in warm little ponds. 10.1073/pnas.1710339114 PNAS October 2, 2017

Tashiro T, Ishida A, Masako Hori M, Motoko Igisu M, Mizuho Koike M, Pauline Méjean P, Naoto Takahata N, Yuji Sano Y, Komiya T. Early trace of life from 3.95 Ga sedimentary rocks in Labrador, Canada. Nature 549, 516–518 (28 September 2017) doi:10.1038/nature24019

Tuesday, September 26, 2017

Vast, Cool and Unsympathetic: Other Worlds Detecting Earth

If aliens visited Earth in any numbers, the result would likely be a disaster for our species. This has been the consensus of the not-inconsiderable number of scientists who have given this topic serious thought, Stephen Hawking not least among them.

And yet, there are people who deliberately try to signal our presence to aliens. If you take the possibility of intelligent aliens seriously, you should support a ban on this activity. These people either don't seriously believe they'll be heard, or they're willing to risk the end of life on Earth at some future time, all for their little project. It's as if the Mayans had built signal fires on the beach to show the locations of their cities to any helpful people navigating the coast in giant canoes.

If you think life on Earth is the result of evolution, and for some reason you're hesitant to extend Darwinian principles to the rest of the universe, think about it this way:
  • a) the Mayans encountered individuals from the same species, a mere four thousand years more technologically advanced than themselves, and the results were catastrophic to Mayan civilization and the New World's ecosystem, and

  • b) any idea that humans are somehow nastier than any advanced organisms that might visit from the stars is based on nothing, except wishful thinking and a desire for moral signaling.
If you like Earth's ecosystem, and you despair of the way that invasive species from the Old World (for the most part) have rolled over those in the New World and on island ecosystems like New Zealand, imagine the damage to Earth's biomes from invasive alien microorganisms. (Again, if we take the possibility of aliens seriously, then this should be considered as low probability, very high consequence threat, i.e. an existential threat, along the lines of an asteroid impact or gamma ray burst.)

Therefore, it's worth worrying about how easy we are to detect. This paper proposes a way to cloak the Earth with lasers. Another way to think about it is to establish a detectability index, and a useful one might be: how far away could a parallel Earth (with the same EM emissions) be, for us to detect it? Or, for them to detect us? I call this the C-index, and XKCD's What If addressed the same question. Now, astronomers have asked what other solar systems are ideally positioned to witness a transit of Earth against the sun, even without hearing EM emissions.

Astronomers have debated what types of planets are most likely to develop life, and a good summary might be that we should like for habitable-zone super-Earths that are closer to the center of the galaxy and have had frequent proper motion close passes by other stars. We should see if any of these stars meet those conditions, and study them exhaustively.

Previous post on alien evolution, Influence of Interstellar Proximity on Interstellar Exploration and Evidence of Extraterrestrial Visitation

Friday, September 15, 2017

The Battle of Mohi

Today, this place is in Hungary. In 1241, it was also in Hungary - and it's where the Mongols eviscerated the only army that had a serious chance of stopping their advance to the Atlantic...and then soon after, vanished back into Central Asia. What if...

The Paradigm Shift Problem in Science Fiction Movies: How Will Blade Runner 2049 Solve It?

A paradigm shift is a plot twist that changes the setting and therefore the whole meaning of what the characters are doing. Science fiction suffers uniquely from this because it's unique in having paradigm shifts (the rare action movie notwithstanding.) In Alien, we find that the crew is not dealing just with a face-crab, but with a predator that bursts from people's stomachs. In V, it's that the friendly human-like aliens are literal cold-blooded reptiles. In the Matrix, it's that we're all in a simulation. In Star Wars it's that Darth Vader is Luke's father - a bit more like a traditional plot twist, but still profound.

When a movie depends on the paradigm shifts for its money shot, what do you do to keep the fires burning? You could either make an entirely different reveal in the next movie - and that's hard. You can just keep having things bursting out of people's stomachs (boring!) or just have, I don't know, the personification of the Matrix reveal you know, a little bit more (so lame!) Or, you could add a whole NEW paradigm shift. That's difficult and dangerous. Not only are good paradigm shifts are hard to come by, but introduce a new one, and suddenly you've changed the whole tone, and it no longer seems like a movie from that franchise. This is hardly ever done, although Highlander 2 does come to mind, and it was so bad that Highlander 3 ignored it as essentially un-canonical.

Of course audiences are getting smarter, and harder to please, as we see the man behind the curtain, because we're saturated in media and we've become a bit too savvy and aware of how plot structures work and we roll our eyes at call-backs (Sarah Connor telling Kyle Reese "come with me if you want to live", WOW! I can't believe that it's reversed now!) The innovation here is the J.J. Abrams trick of making the characters aware of their past and having them consciously do the call-backs - Kylo Ren looking at Vader's helmet - and possibly even having the reversed Vader-Luke interaction, and having the bad son succeed in killing the good father.

I'm nervous about Blade Runner 2049 running into the same problem, but there's a difference, and here I'm afraid the title of the post is a little misleading. Blade Runner kind of doesn't have to solve the paradigm shift problem. Blade Runner doesn't rely on "Is Deckard a replicant?!?" for its punch to the same degree that the franchises above do. Its success is in the overall tone set by the first movie, so Villeneuve isn't as trapped into focusing on the paradigm shift as the core of the story. The Fifth Element was probably the movie that was at its debut most compared to Blade Runner, and the film-making is such that, if it has a sequel, it also won't be forced to make such difficult no-win writing decisions. The new short that takes place in 2048 is obviousy a Blade Runner film, by the sound just as much as the famous vision. My main complaint is that it seems that nothing much in that world seems to have changed in 30 years. But we will see!

Sunday, September 10, 2017

Anticipation: Blade Runner 2049 - and Kubrick?

At least one person has the theory - and enough confidence in it to post a Youtube video about it - that Blade Runner is deliberately full of references to Kubrick's 2001. Of course, the ending credits are actual Kubrick footage from the Shining, but there are a couple other things. And of course, these are my own two favorite movies in the world, so I would love for this to be true. Consequently I'm immediately skeptical.

In Blade Runner there are clear references to other movies - for instance, the lighting in the Tyrell Corp. office where Deckard interviews Rachel is unambiguously straight out of an early scene from Orwell's Citizen Kane (another great one.) But I have in fact noticed two odd connections recently, not between these two movies, but between Kubrick and the (now we must use the word) franchise. To wit: Tyrell is played by Joe Turkel - who also played the bartender in The Shining. Perhaps more significantly, Denis Villeneuve, the director of Blade Runner 2049, is a certified Kubrick freak.

Friday, September 8, 2017

Review: Alien Covenant

tl;dr Not terrible, and definitely not as bad as Prometheus, with good acting and better-developed and organized themes but still with some ridiculous plot holes.

I had said I was not going to worsen my experience of the Blade Runner sequel by over-anticipating it, and I am failing spectacularly. I'm beginning an effort to see the recent Ridley Scott movies (he executive produced Blade Runner 2049) and will also be seeing all Villeneuve's movies before I see the new Blade Runner. This is my excuse for having seen Alien: Covenant.

I reviewed Prometheus quite negatively several years back and while Covenant is not as bad, nothing will ever live up to the original, through no fault of Scott's. Science fiction relies on paradigm shifts more than other genres. Paradigm shifts are a subspecies of plot twists. Plot twists are an event unexpected to plot and characters (the Red Wedding in Game of Thrones), or the revelation of information about characters or events in the past that changes the motivaitons of characters and the meaning to audience of events. A paradigm shift is a revelation about the very nature of the world that the characters are living in; e.g., we actually live inside a simulation! (the Matrix); the V aliens are not really humanoid but actually reptiles!

And therein is the problem. Soap operas and action movies can and do have multiple plot twists. But since the paradigm shift is about the whole world on the screen, it necessarily becomes the central aspect of the story. And you're trapped: either you just rely on the same shift (and wear it out) or you shift to a new one. Rely on it, and you bore your audience (the second and third Matrix movies; the utter lack of surprise that the V aliens are reptiles; the eye-rolling boredom of "Wow, aliens are bursting out of people's bodies? Who knew!") Shift to a new one and you're making a different movie. Studios and directors really don't want to change the underlying formula of a successful franchise, so more xenomorphs bursting out of chests it is! J.J. Abrams developed a neat trick for coming into mature franchises and simultaneously appealing to fans and casual viewers, but even this trick has worn out a bit. Without further ado then, the good bad and ugly of Covenant.


- Great location and cinematography. For that matter I even liked that in Prometheus. Then again, you could film anything in South Island, New Zealand and I would probably like it.

- Scott did a better job of developing the theme that we live in a pitiless and mechanistic Darwinian universe, and that maybe the machines really are seeing accurately (in the original Alien, "Survival - unclouded by conscience, remorse, or delusions of morality.") The cruelest can, and probably usually do, win, despite our strivings and the trappings of our comforting values. I also like the animalistic nature of the aliens throughout the franchise, again on display. Nature doesn't produce culture, it produces survival, and culture is a by-product that helped survival (for a while.) David's creations have a predatory cleverness but there won't be any exchanging art or technology with them any time soon. I really liked the very biological touch of the dead facehugger curling up like a dying spider. There's also the theme of Weyland creates android; android creates xenomorph. Weyland wasn't very nice to begin with. They also explained why the later version of the android didn't just run over the earlier version (Walter explains that David disturbed people and so there were limitations placed on later models.)

- Fassbender is an outstanding actor and I might even have enjoyed the movie more if it was just David and Walter talking and fighting. The flute-teaching scene was very suspenseful, erotic, and in general made me uncomfortable, which I very much appreciated; by far the scene in the movie that bothered me most (yes, even including creatures bursting out of organs.) There was a criticism of Prometheus that the only character that didn't act like a robot (Fassbender's David) was, in fact, playing a robot, especially against Charlize Theron's woodenness. (Did you see her in Mad Max? She stole the show. If somehow you made Charlize Theron boring, you really screwed up.) I wonder if Scott was creating this contrast intentionally, but it's hard to defend this as intentional direction when the rest of that movie seemed like a disorganized pastiche of half-developed ideas. Here, the contrast is clearly a central feature of the themes of a much better developed story.

- In some of the lines and themes, there was more a sense of the film-maker's call back to previous movies (and his own outside the franchise)
  • While Daniels is resisting David's attack, David says, "That's the spirit." I'd argue that's an intentional Roy Baty reference.
  • David refers to the xenomorph as the "perfect organism", which Ash calls it in Alien. This one is particularly interesting because you wonder if somehow David influenced Ash, or if androids have a tendency to be psychopaths that arrive at the same admiration for the Darwinian killing machine.
  • At first I thought Scott was plagiarizing himself in the flush-the-alien-out-the-airlock scene, but he's actually plagiarizing Cameron after taking over the franchise.
  • It has nothing to do with Scott, but when David and Walter fight, you can't help but think of scenes from a certain other Cameron series with two terminators going mano a mano. Their relationship very much invites Data and Lore comparisons as well.
- David is a character study of a psychopath. His narcissism, manipulativeness and lack of empathy are clearly on display (an android admiring himself in a mirror?) Walter points out his flaw in misattributing a poem. David's crying is clearly sometimes for show, but he does seem to have been genuinely moved, at least abstractly, by the slaughter of the Engineers. On the other hand, Walter's innocent Kantian adherence to "duty" in contrast makes him seem safe, boring and even child-like.

- More interesting to think about in terms of where Scott could take the franchise - if David's point in creating the xenomorphs was to wipe out organic life, or at least both humans and engineers, then it makes sense that in Aliens 4, Ripley and her new android friend finally make it back to an obviously ruined Earth. If Scott continues to make movies, he's either going to have to declare other directors' work on the franchise not cannon, or do a lot of maneuvering. (Until this decade Scott had made only one of the four Alien movies. Each one had a different director; and of course this is even more complicated since xenomorphs have also been kidnapped into another franchise entirely (Predator.)

Interesting, so I guess positive: James Franco's role in the movie was near zero, other than two lines of dialogue on a video the protagonist was watching, as well as being a Pittsburgh-rare corpse. There was really a lot that was released online that was not in the movie.

Neutral, but I expected it to be bad, so I guess that means good: I was expecting to hate Danny McBride in this (boy, what a casting decision) and I had no feelings one way or the other.

Puzzling: Mr. Foreign-Accent McBurny-Face had a facehugger on him for all of 30 seconds and still managed to get a xenomorph implanted. Why do the things hang on to faces (in later-in-the-timeline-movies) for hours to days? The chest bursters popped out in a matter of minutes too. It could be that the later ones had deteriorated somewhat from David's original design. Or, it could be that Ridley Scott cares more about rapid action than continuity, because audiences tend to ignore plot problems when there's higher action density.

The Negative

- So stupid I can't even believe it: trained interstellar professionals go gallivanting off the ship into the brush on an alien planet, one they weren't even planning on landing on, and no concern for pathogens? (You could get rid of some of this with a single line in the plot without changing the visual story much at all, e.g. "remember when we got our universal vaccine shots", etc.) And very little comment about "Wow, we're setting foot in an alien ecosystem!" And very little comment about "Wait, there's wheat here, humans must have been here before us!" Fine, maybe Scott thought audiences are a little tired of every movie showing the wonder that the first settlers felt...but worst of all, when the crew goes out with THE BIOLOGIST to collect samples, a crewmember smokes and just casually tosses the cigarette butt. Seriously!?!? And then when the crew is returning with THE BIOLOGIST, two of them are obviously, suddenly, severely sick with a local pathogen, and THE BIOLOGIST is in full contact with them and doesn't think to put them in quarantine outside the ship. I mean come on. This is worse even than a bad horror movie where they say, "Hey, inhuman howling in the basement of this abandoned house, we should go down and see what it is!"

- Also really stupid: Yes Scott went out of the way to show us that Billy Crudup was a bad captain. But still, he doesn't appear to be a complete moron or an emotionally vacant nitwit. "Hey, a creature just killed one of my crewmembers and not only were you communing with it, but you were mad at me when I killed it!...okay, I'll follow you into your horror chamber and lean my face into this weird pod when you tell me to."

If Scott had made the characters behave a little more circumspectly - maybe even have viruses that could get through air filters on space suits, have David do his explication to Walter and have a facehugger drop on bad-captain from above as he was about to shoot David - this would have been a much, much better movie.

- Confession, I knew how it ended, but come on - how was it possible not to realize that it was David and not Walter when he was proudly watching the xenomorph run up and down the deck? Waiting for the reveal at the very end was quite annoying for that reason.

- Also, once they're in the dead colonial city - I guess there's barely a word uttered about the massive carved monoliths around them because they've already heard of the Engineers from the earlier mission? (Although again, a little more awe might still be on order when you walk into an alien city.)

- A neutrino burst? Come on, you can do better than that. There are only enough neutrinos to do serious damage if you're right next to a supernova, and then you have other things to worry about. If David can trigger neutrinos and supernovas, why doesn't he just go back to Earth and do that? The plot would've lost nothing if a stray shower of comet fragments damaged the ship and woke them up.

- A colony ship with a specific destination goes off-route to land on a planet emitting a strange (but ultimately human) signal. Alright; that's fair enough. But how did they miss a planet that was habitable? Yes it's a trap that David set, but the planet was habitable prior to any interference by David. Note that the ship has hyperdrive and it's the early twenty-second century, and it's interesting that we feel we can say that won't happen. (This is a reality criticism, not film criticism.)

- The bad android has an English accent and the good one has an American accent. I want to see an evil robot from Alabama or Queensland. (Gary Oldman as the villain in Fifth Element did have a Southern American accent, to that movie's credit.)

Monday, September 4, 2017

General AI: Computation versus Survival, Superintelligent is Not Omniscient

It is usually assumed that a superintelligent AI would maniacally focus on improving computation. Just to highlight the centrality of computation, a recent paper in the British Interplanetary Society Journal argued that the reason we don't see aliens is they're sleeping, waiting for a time when the universe is cool enough that their computations are more efficient. The alien singularities are waiting until they don't need to be cooled.

The most common concern associated with this line of thinking is that the technological singularity would be bad because the AIs would use all available resources - starting with all matter on Earth, including us - as computational resources. While I think a technological singularity would be catastrophic, I think the reason is eve more mundane.

Of course, this assumes that the AIs in all their power are maximizing computation. I don't think this is questioned nearly enough, and a good bit of the inertia around it stems from the cultural assumptions of the programmers and engineers making the argument. The singularity is thought of as a logical outcome of Moore's law, which concerns exponential growth in computation. It's not clear that this is what an AI would necessarily be maximizing. For our part, humans and other animals maximize a host of confused and often contradictory goals. Of course we remain in this mess because we are not recursively self-modifying. Assuming that AIs with such an ability aren't automatically condemned to wirehead, it's not unreasonable to ask whether there are things to maximize that increasing computations just wouldn't fix.

Replicators whose descendants are present into the future are the result of selection for one thing - making copies - and to the extent that extra computation can improve that, then the AIs present in the future will be selecting for computation that helps them reproduce and sustain themselves. But even a superintelligent AI is not an omniscient AI, and cannot see infinitely into the future and understand ahead of time the impact of all its actions in maximizing its survival OR computation. My strong suspicion is that a hard takeoff will likely be an apocalyptic gray goo explosion, much more thorough and faster than the mass extinctions so far in the much more comparatively mildly ecocidal anthropocene, and that furthermore this is a strong candidate for the Great Filter and the Fermi paradox. That is to say, we're more likely to find the simple but fecund survivors of such an event as something that looks like post-singularity AI-algae (or free-roaming AI "cancer") than alien AIs that are interested in philosophy.

Sunday, September 3, 2017

Influence of Interstellar Proximity on Interstellar Exploration and Evidence of Extraterrestrial Visitation

It's easy to despair at the gulf between stars and the millennia of time it would take to get a ship there. The fastest spacecraft humans have yet produced was Helios 2, which after a slingshot maneuver in 1989 was moving at about 103 kilometers per second. There are two things to note about that statement. The first is that it was a slingshot maneuver, not an acceleration achieved under its own power (which is always the case in space exploration.) The second is that breakneck speed would deliver Helios 2 to our nearest neighbor Alpha Centauri right around 13,000 years from now. (This also means there are stars in our neighborhood that even if we aimed our fastest-yet probe at them, we could never reach, because they're moving away from us faster than our fastest spacecraft.)

Those frustratingly quarantine-like time spans suddenly seem much shorter when we consider the geologically brief time spans between close passes of the sun and nearby stars, resulting from proper motion. These changes in interstellar geography actually occur much faster than plate tectonics. In the space of a few tens of thousands of years we might go from having our nearest neighbor five light years away, to close enough to impinge on the Oort cloud and send comets falling toward the inner system. It's amazing to think but since our ancestors were first using fire, multiple close-passes between other stars have occurred. Merely 70,000 years ago we had a star 0.82 light years away (Scholz's Star), and in another 1.3 million, we'll have another (Gliese 710).

There are a number of clear inferences to be drawn from the frequency of such close passes.

1) It bears repeating, in geologic time, 70,000 years is really not long. Humans may have already started leaving Africa when this occurred. We should assume the sun is not unique in this, and our close-pass rate is about once per million years. That means, since the solar system formed, this has occurred 4,500 times. The Wild-2 comet, from which we retrieved material that we've analayzed on Earth, has a nitrogen isotope ratio that strongly suggests it's a comet formed around another star. It also has the amino acid glycine.

2) Impact ejecta from large bodies like Earth can make it into orbit. We have Martian rocks here on Earth from such events having happened on Mars. Space is not hospitable, but even metazoans have survived fairly harsh exposures; for example, C. elegans worms from space shuttle Columbia experiments survived uncontrolled re-entry and were found alive on the ground weeks after the crash. They weren't even protected inside large space rocks. Some exobiologists expect that for this reason, if we do find life elsewhere in the solar system, it will be related to life on Earth (ejected and diffused during the Archaean?) essentially a long-lost branch of archaebacteria. While such a process would be less likely, i.e. take longer in the much greater volume of the outer solar system, if it is not less than 1 in 4,500, it has probably already occurred.

3) It is very likely that the amazingly short 66 years from first manned powered flight to first human on the Moon occurred in part because of the Moon's relative proximity. More than twice that duration has now elapsed and we are still only in the talking stages about a landing on Mars. A species that has the good fortune to "come of age" in terms of space faring technology, when the next closest star is a mere 0.82 light years from their own, has an easier task of proving the possibility of interstellar space flight than we do, coming of age when we're about equidistant from everything. We might therefore narrow our search for intelligent life to super-Earths around sun-like stars with close neighbors. (Of course, there is also a not-unreasonable argument to be made that close passes, or any distant large bodies disturbing the local Oort cloud, increase the chance of major impact events and decrease the chance of the kind of complexity developing that would allow long-distance space travel.) It's worth noting that in view of the high frequently of close passes, a problem for the Oort shower hypothesis of mass extinctions is that it does not happen more frequently, i.e., every million years.)

4) Recalling that our fastest spacecraft have all used gravitational slingshot maneuvers - while we might speculate wildly about the amazing propulsion technology visiting aliens would have, we can be 100% certain that they will have gravity maneuvers at their disposal, because we use it. It's easy and cheap (free, really.) Therefore, there may be interstellar "backwaters" that will not necessarily be empty spots in the galaxy, but places that are difficult to approach from nearby stars and then slingshot away from to another nearby star. If you're a species in such a backwater, you're not going to get visited very much, and you'll ask "Where is everybody?" As the stars shift, your interstellar geography status may change quickly, within a few thousand years. One of the problems with detecting aliens, particularly well-advanced ones, is we don't know what we're looking for. We may be looking right at evidence of their existence and miss it because they don't use our provincial communication methods, or because we're used to it and we explain it in terms of the background operation of "dumb matter". Or, the periodic mass extinctions that are sometimes claimed to be associated with close passes could in fact be associated with close passes - but because of an ecosystem-collapsing alien visitation as Stephen Hawking envisions, rather than because of Oort cloud impactors.

Previous post on alien evolution, There's (at least) a 1-in-3 Chance of Life on Europa

Friday, February 3, 2017

Review of Journey to Fusang, and Some Problems with Alternate History

This is cross-posted at the Late Enlightenment.

I'd had William Sanders's Journey to Fusang on my to-read list for years, and I finally got around to it. As much as I enjoyed the book, it exposes some of the problems common to the alternate history genre and even those specific to what-if-the-Mongols-invaded-Europe stories. It also made me wonder which historical branchpoints are most valued by "the academy". Each of these is covered herein.

Part I: Review of Journey to Fusang

After five centuries of European cultural domination of the world, it's difficult not to be fascinated by the now almost fantastic idea of non-Christian non-Europeans conquering parts of Europe. Of course this did happen in three corners of Europe - the Moors in Spain, the Ottoman Turks in the Balkans, and of course the Mongols. And if there were an award for depth of impression per years occupying European territory, the Mongols would win hands-down. Their advance beyond Russia lasted just a few years. And in the territory they actually held onto, modern Russia, they were there for a scant two centuries. Even in the East the dynasty they established (the Yuan) fell apart in about the same length of time, despite some useful innovations - among them, the world's first adoption of paper money, not to mention an honest attempt to change the Chinese writing system to a sensible phonetic one.

The Mongols' western high water mark was Poland and the Balkans, where they achieved several characteristically cunning and blitzkrieg-like victories before making winter camp. In the spring of 1242 they decamped just as suddenly and faded back into Central Asia. The reason typically cited for their withdrawal is the death of Ogedei Khan; in the Mongol institution of selecting a new khan, the kuriltai, you get a vote only if you're physically present at the succession talks, and the sub-Khans leading the European campaign did not want to be left out of the decision. As with most things in history, the reality is probably more multicausal and complicated. For example, Europe was littered with well fortified castles and their sieges were making for frustratingly slow going.  What's more, Mongols were not all-weather super-men, and they didn't do well in cold, wet, marshy areas, e.g. Europe.

But simplifying history for the sake of a good story (this is after all a novel, not a textbook), in Journey to Fusang, Ogedei's sickness does not lead to his death, and the spring of 1242 sees the Mongols conquer Europe from the Vistula to the North Sea and the Mediterranean. England and Ireland are spared the ravages of the grim horsemen, as there is apparently no western Khan ("khan of all the Germanies") as enterprising as Kublai (or there is, but England is protected by the same divine winds that defend Japan.) In the late 1600s when the action takes place, England and Ireland both call themselves Catholic though each has its own Pope. Milton is a gambler in London, and Shakespeare escaped to Ireland and wrote light comedy. Though, as a result of the conquest, nothing like the Renaissance happens in Europe, nor any Reformation, and there is certainly no European discovery of America. The European continent remains a cultural backwater stuck in a Malthusian trap of subsistence agriculture and ignorance, leading the protagonist at one point to make a comment about French women rearing litters of children named Pierre and Temujin. The East Coast of North America is being colonized by Moors and various other Islamic people, and the West Coast by the Chinese - the crown jewel of whose possessions is sunny Fusang and its colonial capital, a hilly metropolis by a bay. Somehow amidst all this the Aztecs not only survived but prospered, accumulating guns from Muslim traders, and they're a constant concern along the colonists' borders.

The protagonist is Finn, an Irishman, and a trickster and con-man extraordinaire, who finds himself captured by a Moorish slaving ship, which he escapes with a similarly-minded Hebrew indentured crew member. They find themselves in the great city of Dar Al Islam (New Orleans). The book does rely on incredible turns of fortune (in one moment the protagonist is in deep, the next he's on top of the world) and the plot is rarely predictable. Eventually our hero makes his way to the interior where we meet Comanches with names like Muhammad Ten-Bears. The large, dull English slave whose dumb luck brings him along for the adventure is a clear parallel to Lewis and Clarke's black slave York (the natives are amazed at his size and complexiion and, as York did, the fellow takes advantage of their fascination by fornicating with scores of their women.) Eventually the group of Islamic frontiersmen end up in Taos, where they encounter Chinese colonists for the first time. I particularly liked the description of the strange energy of Taos - I'm about as un-spiritual as they come but I find the place oddly unsettling. They also hole up in Acoma for a siege, which would be as good a place as any. In the real world, Acoma Pueblo, which is a cliff dwelling that's been continuously occupied since well before Columbus. I should point out that the breathless description of the ninja they encounter is one of the few things in the book which marks it as a work of the eighties.

Tension builds as both the Chinese and Muslims discuss nameless troubles in the north, which turns out to be a Russian Khan and his horde that crossed the North Pacific and invaded from - not the soft underbelly, but the undefended top half of North America. What results is very much a recapitulation of Mongol sieges and brutality, but in the American heartland. (Among other towns, Taos is completely destroyed.) The Russian khan's ultimate goal is the conquest of the Aztec empire from which he can invade the settled parts of the continent, which would precipitate a kind of world war. Of course through a combination of cunning and luck, our hero is able to stop the invasion, and after recovering from the injuries he sustains in doing so, he migrates to Fusang.

For me, what really sets this apart from other alternate histories was the scoundrel protagonist's cynicism, insulting descriptions and occasional misleading narration. For instance, as he and his Hebrew companion are traveling up the Mississippi to their frontier outpost (and not helping with the work at all - remember, these are escaped slaves that people are tolerating!) the manager of the expedition has finally had it with their shiftlessness:
Ibrahim had discovered Yusuf and myself standing in the shade of a cypress tree, overseeing the work and occasionally calling out useful suggestions, and had delivered a lengthy and surprisingly emotional speech in which he listed various discrepancies and points of dissatisfaction with our general performance to date. It worried me to see a man his age and weight get so worked up in that hot climate.
The Irishman abounds with politically incorrect observations of the many people he meets, although usually not without self deprecation. Even the people in whose colony he finds himself are not safe:
Arabs, on the whole, are an amazingly mendacious people; they will lie even when the truth would serve their purposes better, and show no shame even when caught at it. Among themselves this does no harm, since they never believe each other anyway; but it can be annoying for a man brought up in a scrupulously truthful country such as Ireland.
And near the end of the novel in Fusang, the protagonist and his sidekick, having made the acquaintance of a member of a shady Fusang organization, are recommended to one of the city's underground bosses, who deals with them thus:
He said, "Lu Hsu says you wish to work for us. He says," he said, rattling the letter, "that you are a pair of liars, cheats, seducers, and thieves, and cold-blooded killers if need be. He says that one of you would steal a hot stove and the other would sell it to a man dying of sunstroke. He says only a fool would trust either of you within a thousand li of his cash box, his stable, or his wife, and that even for Europeans you have set new marks in treachery, fraud, and deceit. Gentlemen," he cried warmly, "I have never seen a finer letter of reference. I am prepared to offer you immediate employment, early promotion, and full benefits."
In short, the book works as a historical thought experiment about the impact of Europe on North American cultures, the protagonist is hilarious, and it's a solid adventure novel to boot. Highly recommended.

Part II: Problems of Alternate History, and of Mongols-Conquered-Europe Stories Specifically

There are really three main problems (or assumptions) that alternate history novels have to address to succeed, and this novel is an excellent vehicle for exploring them. But it's useful to compare to three other alternate history novels, because you'll see some repeated motifs.

Book, AuthorBranchpoint and Changes
Lion's Blood, Stephen BarnesHannibal defeats Rome; North America colonized by Egypt in north and Ethiopia in south using European slaves, with Aztecs getting guns and flourishing
Conquistador, S.M. StirlingAlexander the Great does not die young; Europe and Middle East merge into millennia-long superstate, scientific progress retarded
Years of Rice and Salt, Kim Stanley RobinsonBlack Plague kills all Europeans; Americas settled by China and Islamic world, history becomes a struggle between the two
Journey to FusangMongols conquer Europe , leaving it a backwater; Americas settled by China and Islamic world, with Aztecs getting guns and flourishing

Alternate History Problem #1: What is the point? Why would you read a book about a history that never happened? Do you just want to be shocked by strange people in familiar places (mosques in frontier-era Arkansas, and Native Americans named Muhammad Ten-Bears), or are you trying to isolate historical cause and effect (did Chinese and African colonization of North America only fail to occur because of European domination?) I couldn't shake the feeling that Sanders wanted a Mongol-like horde in the American plains just for the sake of it. Granted, a Mongol horde on the American prairie is kind of cool, but that could not have carried a whole novel. At the same time, even the shocking reversal of oppressors and slaves here or in Lion's Blood was clearly written in the service of a thought experiment. Was there something about various human cultures that predisposed one people to enslave, or be enslaved by, another? Or is this something we would've all done to each other given the chance, and it's all just random, and history really is just one damn thing after another? (These days we get very uncomfortable saying that one culture might somehow be "better" than another, but here's a fact: either culture has some impact on human flourishing, or it's meaningless background noise. There's no middle ground; we have to pick one of those, and being uncomfortable with the choice doesn't make it go away.)

Alternate History Problem #2: Is warfare really the only thing that makes a difference in history? Most alternate history is about what would have happened if some episode of mass violence had transpired differently. Is everything else we do really so meaningless? In the four novels cited here, the branchpoints for two out of four is a war, one out of four is a change in the life of someone who's famous because of war, and one is from a plague. What if the steam engine were developed earlier? What about antibiotics, or intensive agriculture, or monotheism, or electricity? (For example: why no Byzantine industrial revolution? People have asked the same question of China but it should count for something that China developed intensive agriculture, centralized state bureaucracy, literacy, paper money, and gunpowder, some of which happened after getting sacked by the Mongols.)

These questions are far from useless. Economists are constantly asking questions of how developing nations can improve themselves, which invariably become questions about why certain advances occurred certain places, and not elsewhere. You might argue that it's not the wars themselves where the future is set; that Sun Tzu was right, and the outcomes of battles are determined long before they begin, by the culture and technology of the countries that produce the armies. Case in point: a normally clear-thinking friend once asked me seriously about the possibility of Mexico having won the Mexican-American War. Once the American military of the day was fully engaged with its Mexican counterpart, there was no question of the outcome. One question at least implied by many alternate histories is the impact of culture and political situation on the advance of science. Conquistador assumes that a unified Greece-Egypt-Persia would have halted progress, where Robinson assumes it would have advanced essentially the same in the absence of Europeans, merely with name substitutions (qi for electricity for example.) (I would argue the contrary in both cases.)

And here we come to some of the problems specific to Mongols-conquering-Europe stories. It's assumed that a Mongol sack of Europe would have flattened the West and removed it from the world stage. No doubt it would have been an unpleasant time to live in Paris or Rome. But in China, a Mongol conquest is not alternate history. The town of Yamen, a coastal town two hours' driving time from the metropolis of Guangzhou, was the site of the last surrender (or suicide) of Song officials to Kublai Khan. In case you haven't noticed, China did not collapse into a permanent dark age. You could argue that without the Mongol conquest, China might today be even further ahead (maybe it's time for an alternate history where the Great Wall was a little higher and kept the Mongols out, and by 2017 China has colonized the solar system.) A further problem in Mongol conquest-related alternate histories is that in two novels with Europe eliminated as a serious contender for founder of global civilization (whether due to Mongols or plague), why is it that automatically China and the Arab world (both also sacked by the Mongols!) who colonize the Americas? (In two books out of four on this list!) Even in real history, there was nothing stopping the Chinese or Arabs from such voyages, and in fact the famous Chinese treasure fleet was sailing before Columbus, and less than two centuries after Kublai's conquest. It was not because of European oppression or Mongol depradations that they were brought home, but rather due to domestic Chinese politics. (Arguments for an age of discovery driven by the Islamic world are even more obscure; despite this, three of the four novels cited feature Arabs colonizing North America's Atlantic coast, with only two novels showing Chinese colonization from the Pacific.)

Finally, there seems a strange urge to describe an armed modern Aztec nation-state, flourishing unmolested by non-European colonial powers who give them guns. In both novels cited above that contain an armed modern Aztec nation-state, their colonizing neighbors came from the Islamic world. The Spanish were not famous for their tolerance of Aztecs or their culture, but it's absurd to argue that Islamic conquistadors would have been more progressive. On the other hand, I freely admit that Barnes's Aztec knights were extremely cool, almost like an alternate history answer to Niven's kzin. In Journey to Fusang, Yusuf describes Tenochtitlan thusly:
He sat down on a coil of rope. "How can I describe it to you? What's the biggest, finest city you've ever seen?"

"Tangier," I confessed. "By a great margin."

"Tanger?" He made a scornful snorting sound. "Tangier wouldn't make a minor suburb of Tenochtitlan. It's bigger than Rome or Constantinople or even Baghdad - why, there are independent kingdoms in Europe and Africa that cover less area than Tenochtitlan alone. And all of it laid out carefully with long straight streets, even canals in some parts like those of Venice, and market squares bigger than most European towns. Flowers growing everywhere, and the people looking so clean and well-fed, even the poor...and in the center, dominating the whole city from wherever you stand, that great pyramid, with its two temples on top - the blue one for Tlaloc the rain god, the red one for Huizilopochtli, he's the really nasty one - with a whole city-within-a-city of lesser temples and palaces clustered at its base. Finn, I've seen the ancient buildings of Greece and Rome and Egypt, and I'd put these people's work up against any of it.

"And then," he said, still in that quiet, almost toneless voice, "while you stand there trying to take it all in, trying to grasp the wonder and the beauty of it all, like St. John seeing the New Jerusalem - just then you glance across the street and see a priest striding along in his black cloak, face painted black, hair hanging to his knees and matted solid with years of accumulated dried blood, and the stink of rotting blood coming off him like a walking slaughterhouse - and then you remember, and you can't believe it, and yet there it is. The same people. How can it be?"

Note again: in the real world Baghdad was sacked quite badly by Hulagu Khan; while not today the center of global learning that it once was, it was hardly reduced to an irrelevant backwater.

A summary of the motifs across these novels:

Book, AuthorBlacks Enslave WhitesMuslims Colonize Eastern N.Am.China Colonizes Western N.Am.Modern Aztecs w/ GunsBack-ward Europe
Lion's Blood
Years of Rice and Salt
Journey to Fusang

Alternate History Problem #3: Balancing realism with entertainment. This is fiction. It can usefully explore ideas, but ultimately it has to be carried by a narrative, and if that is lost, the medium fails. If Hannibal had won the Second Punic War, the map of Europe (or whatever it would be called) would look completely different and be populated by people who looked and spoke and worshiped completely differently from how they do today. If a branchpoint is set long ago and still faithfully followed through, it would result in a world so bewildering and unfamiliar to us that its people would be uninteresting. Conquistador's branchpoint was Alexander's surviving his fever in Babylon, but Stirling mostly used that as a way to produce a mostly-empty North America to be settled from our timeline. Of the four novels here, Lion's Blood is the worst offender in terms of a long-ago branchpoint magically still producing a familiar world, but neither is Fusang innocent. In the case of Lion's Blood, there's a fourth century B.C. branchpoint, but Christ - the same Christ - is still born 250 years later, and founds the same religion? And Mohammed more than 800 years later? Of course, a modern world without Christianity or Islam would essentially be invented out of a whole cloth - so such a degree of honesty is inadvisable, unless it's directly relevant to your thought experiment; e.g., whether evangelical monotheism would still inevitably have appeared. Sanders peripherally mentions Shakespeare and Milton but has the good form at least to make their output different.

There is also the problem that people with a different viewpoint would not use the same names for the same places (obviously) or imbue them with the same significance. By putting cities in recognizable locales (the mouth of the Great River, or the hilly city by the bay) you can cheat to some degree. Rice and Salt opens with a Central Asian horseman riding alone through an oppressively silent city of white bone-like monuments under the moonlight, and ascends a hill to inspect the most prominent. Ignorant of European history as you would expect such a person to be, he can only describe it in terms of pure sense experience, and it takes a few pages before the reader realizes he is plodding through the Acropolis in a decimated Athens emptied by the plague. I respect Robinson for this opening, because it risks the unobservant reader's missing the symbolism, but there's a substantial reward to the reader who recognizes the landmarks. (Many writers would have succumbed to the temptation of exposition, with the rider mumbling to himself "Well you know, these structures symbolize the foundational philosophy of this continent...")

Part III: Branchpoints Favored By the Academy

I compiled some stats for the Sidewise Awards. Uchronia's overall list would be a more accurate measure of what people are writing, but this is a better approximation to what people value. Order is by number of entries in category in the novel awards.

The distribution between short- and long-form is clearly different. Possibly, people are more likely to explore new territory in shorter form. Interestingly, from the novels, the post-WWII awards all went to stories set in the Kennedy or early Johnson administration - two of which are about the Cuban Missile Crisis, which terrifyingly could easily have gone differently, and not nearly as well. Also note that the short form pieces were more likely not about war or tragedy, with 5 of 21 titles being about some form of science or technology that developed differently (as opposed to 0 out of 21 of the novels.) Notably, of the ~42 titles I went through, there was a one triple winner author (Ian MacLeod), then two each for Stephen Baxter, Harry Turtledove, Chris Roberson, and William Sanders.

Also notable - the closest any of the branchpoints was to the release of a novel was about 50 years; for short stories, 38 years. Is that because if we don't wait a generation to write our alternate history - until all the stakeholders are dead or out of power - it's mere polemic?