pic#57300910 triangle

Recent books

Nine Perfect Strangers - Liane Moriarty

I recently said Liane Moriarty was getting a bit formulaic, but this book breaks out of the mould and is very good.
It's about the guests at a high-end health retreat run by a woman who's a bit intense and cultish.
I was glad they weren't actually all strangers to each other, because that would have been too hard to keep track of, and too hard to become emotionally invested in nine unrelated people all at once. In fact, they include a young couple and another couple with an adult daughter, and we don't get viewpoint narration from all nine, so it feels like a more manageable number.
The interconnecting plot arcs of the various characters, and the suspense of what's really going on with the retreat, are all quite interesting and compelling, and there's a good climactic final sequence.

The Nanny - Gilly Macmillan

This is about a woman who is back living in the house where she grew up, with her ten-year-old daughter, and her mother, who was always quite narcissistic and un-maternal. Her old childhood nanny, Hannah, is also back in the area. Hannah suddenly left one night without explanation when Jo was a child, and Jo was devastated because she felt closer to Hannah than to her mother. She is now keen to see Hannah again and find out what happened, and hire Hannah to look after her own daughter.
SpoilersCollapse )

Going Solo - Roald Dahl

Reading this was an exercise in completionism, as I read all Roald Dahl's children's books as a child except this one, and Bethany now has a box set of all of them.
I didn't read it as a child because it was about him when he was an adult, and as a child I had a strong preference for books about children and found books about adults boring. Now I'm an adult, I wanted to read about Roald Dahl's adult life, particularly his marriages and children (including the one who died of measles) and how his writing career took off. But it turns out Going Solo doesn't deal with any of that. It only covers the first few years of his adulthood, where he works in Africa for Shell Oil and then flies for the RAF in World War II. I can see why he wrote it as a children's book - it's quite an adventure story and I can definitely see it appealing to a certain type of child who likes things like Biggles (which my dad tried to get me into as a child and I found boring). I can't say I really enjoyed it, although his style made it more interesting than the same content written by a different author would have been.
I'm glad I didn't read it as a child, as some parts of it were a bit scary and/or gruesome, more so than his other children's books.

It was quite an interesting snapshot of writing about British colonial history. His time working for Shell in Africa before the war, the people he interacted with and the social structures he moved within were all very much a part of colonialism. And I think in a piece of writing written a bit earlier, this would have been described with pride and nostalgia, and in something written a bit later it would have been described in an embarrassed and apologetic tone, but the book was written around the transition point (sometime in the 80s), so it's presented as a kind of morally neutral curiosity.

When She Returned - Lucinda Berry

This is about a woman, Kate, who turns up after being missing for a decade and presumed dead. Her young daughter is now a teenager and her husband has remarried. She is in a bad way physically and the assumption is that she was kidnapped and held captive, but it gradually emerges that she joined a cult and went willingly.
The cult leader did lie to everyone ("helping" her husband and police with the search, while telling her that her husband didn't want her back), but she chose to join the cult, chose to take the next step of going to live with them full-time without telling her family, and chose to stay with them for a decade rather than leaving them to start a new life as a single person once she believed her husband wouldn't have her back.
I didn't find it very plausible – although the author is a psychologist who's studied cults, so probably knows much more than I do about the psychology and behaviour of cult members/victims. But there were several points throughout the story where the cult leader did something so outrageously abusive and/or illegal that I felt sure Kate and his other followers would see through him and turn away from him, but they didn't. Some of these were before Kate joined them full-time, so I'd have thought she wasn't so entrenched at that point that she would feel unable to leave.
I know that in the case of domestic abuse, naive observers think "why don't you just leave him?" and one answer is that the victim feels emotionally and often financially dependent on the abuser. Often they've been together a long time and/or the victim doesn't know any different (if they've had a series of abusive partners and/or an abusive parent). But none of that applies here - Kate had a stable, loving family before she joined the cult. The only thing wrong was that she was getting a bit bored of her husband and feeling she wanted broader horizons in life than he did. But that just makes me ask even more: why didn't she leave the cult and start a new life on her own? If the cult leader lied to her and told her her husband didn't want her back, she could have used that as an excuse to cut ties and go and travel the world or whatever it was she wanted.
In this book, some of the cult members were former drug addicts who the cult had helped to get clean. I can see how they would feel very emotionally committed and indebted, and would overlook the cult leader's actions, maybe rationalising and finding excuses - but that didn't apply to Kate.
I do know about things like Jonestown, so I guess it is true that cult members continue following their leaders even when they cross the moral event horizon, but I didn't feel this book did a good job of making it seem plausible from the inside. And I didn't enjoy it very much; it was a bit depressing.

In Five Years - Rebecca Serle

This starts with a woman who's just got engaged and has a vivid dream about herself in five years' time, where she's living with a different man. She later meets him in real life, and he's her best friend's new boyfriend.
I thought this would be another interesting "what if?" story, like After I've Gone,the one where the woman sees posts on her Facebook wall from a future where she's dead. This one has similarities, in that Dannie worries about the predicted future and tries to avert it. But the glimpse into the future is a much smaller part of this book than it was in that one. In this book, the dream serves as an inciting incident, but the book is more just about a relationship gradually ending for boring and prosaic reasons, and it just wasn't very interesting to read. I didn't really like any of the characters very much.
The book is actually more about Dannie's friendship with her best friend Bella than her relationship with her fiancé.
SpoilersCollapse )
Also, minor quibble, but the "present" bit is set in 2020 and the "future" bit in 2025. Why not just have them in 2014 and 2019, to avoid having to set most of the book in an unknown future? That always bugs me a bit - unless you're writing sci-fi that has to be set in the future, writing ordinary fiction set in the future just seems unnecessarily set up to fail. A book covering the current period but not mentioning the coronavirus pandemic breaks suspension of disbelief, and who knows what other world events will happen in the next 5 years that will have the same problem.

The Other You - J S Monroe

This was extremely good - very sinister. The author is really good at ratcheting up the tension.
It's about a "superrecogniser" (the other end of the face recognition spectrum from prosopagnosia/face-blindness, which I have to some extent) who used to work for the police, identifying criminals, and is now worried her boyfriend has been replaced by a double. I had heard of the Capgras delusion (where you think a loved one has been replaced by an impostor) and I was worried the book would be a mediocre one-dimensional dance between "she has Capgras" and "he really is an impostor", but actually there was a lot more to it than that: it was twisty and intelligently written, and "is he an impostor?" was only one of several plot arcs, which were all resolved in a satifyingly interconnected way.
I guessed the two main twists before the end, but I don't feel this detracted from my enjoyment, because it was as much a "how will the good guys get out of this scary situation" thriller as a "what's really going on here" one. If anything, the way the clues were stacking up without the protagonist realising it added to the sense of foreboding.
It's closer to being sci-fi than a lot of the thrillers I read these days. Face recognition technology is a major theme, and so is the Internet of Things. The latter makes it feel a bit like a poltergeist story, without the supernatural but with the sinister sense of malevolence from the everyday objects around you: houses where the lights go out and the hob burners come on by themselves, and cars where the doors lock and the heaters come on full-blast.

I have a slight quibble in that a character uses the principle "First do no harm" to persuade a doctor to risk his life by giving up information to help save another character from danger. The principle means that it's better to err on the side of inaction rather than take actions that may harm people, but both characters seem to assume it means intervening to prevent a harm that someone else would commit if you didn't act. That's a perfectly valid principle, but it's not what "First do no harm" means.

I was glad this was so good after the previous two books I read were a bit meh. I want to read more by this author. My LJ helpfully tells me I've already read one of his (Find Me). That was definitely good, but not as outstanding as this one.
pic#57300910 triangle

Sugar-free Lent II

Five years ago I gave up sugar for Lent, partly as a spiritual discipline and partly as a biological experiment, to see if it would improve my energy levels (after a brief initial withdrawal period).
It really didn't, and caused me to be more run-down than usual for the whole six weeks.
I think most or all of the intervening years I've done some form of mobile-phone fast for Lent instead.
This year it occurred to me, fairly last-minute, to try giving up sugar again this Lent - again, for both reasons, and re-running the biological experiment now I don't have gluten-intolerance-related tiredness affecting my energy levels as a confounder (and also have two kids at school rather than a toddler at home).

It's going at least as badly as last year (this is day 9). I'm feeling really run down again (and, thanks to giving up gluten, this is more of an unusual state of affairs - which is wonderful generally, but means I'm less used to it and find it harder to accept).
In 2016 I briefly tried a strict diet with no sugar (even fruit), no dairy and practically no complex carbs, as one of many failed attempts to understand or cure the mystery tiredness I had at the time, before I linked it to gluten. I wrote then "Not eating any sweet things at all, even fruit, I feel like some of the joy has been sucked out of life; which does sound a bit sad and addicted." I'm feeling that way again now, even on a less restricted regimen. I keep being sad because several times a day I think "Ooh, I could have... oh, no I couldn't." I'm also feeling a bit listless and lethargic and generally like everything is a bit grey and pointless. It's like a small, temporary insight into what it must be like for people with depression (and is also further evidence that I DON'T HAVE DEPRESSION normally, thank you, all you medical professionals who thought I did). I've hardly got any work done this week, or housework.
I feel a bit embarrassed that I'm so unspiritual and fleshly that I derive so much joy from eating sweet things and lose so much joy (and functionality) from not eating them.

It's possible that I'm feeling so run down because of the drop in calorie intake (I concluded that was what was wrong with the strict 2016 anti-candida diet, and that was why I abandoned it; after about a week I was finding it difficult to stay standing up). It sounds a bit silly because I'm still eating a normal diet of three savoury meals a day and not restricting them in quantity or ingredients, so that ought to be enough for anybody; but OTOH I normally pig out so much on sweets, chocolate, cake and fizzy drinks that by cutting them out I've probably reduced my total calorie intake by at least a quarter, if not a third, which is going to be noticeable.

I am really questioning whether I can get through the remaining 5-and-a-bit weeks, and wondering whether to pivot to another mobile-phone fast instead for the rest of Lent; but I would feel quite ashamed and pathetic doing that and admitting I couldn't even live without sugar for six and a half weeks (especially as I managed it in 2015, so am presumably more addicted to it now than I was then, which is bad).
pic#57300910 triangle

Recent books

You Look Like A Thing And I Love You - Janelle Shane
Janelle Shane writes the excellent AI Weirdness blog, where she trains neural networks to do things like write fanfiction or recipes, or come up with names for rock bands or ice cream flavours, and posts the hilarious results. I'm often reduced to helpless giggles reading it.
You Look Like A Thing And I Love You is the spin-off book from the blog (named after an example of an AI-generated chat-up line).
It's an easy and entertainin read, but not as funny as I hoped: it's at least 80% a beginner's guide to how AI works, occasionally illustrated with funny examples. My understanding of AI was already at about the level that the book teaches: I've picked up bits and pieces from Slate Star Codex and things linked from it, and from work, where I've edited two full-length lay-audience books on AI, and also read several papers on the topic in order to write summaries of them (which involves understanding them in broad brushstrokes, even if not understanding all the technical detail). I would quite like a book similar to this but at the next level up, to advance my understanding.

Machines Like Me - Ian McEwan
This is about a man and his neighbour-slash-girlfriend (the relationship evolves over the course of a book) who jointly adopt and train an android.
There's a lot of philosophising about what it means to be human, which feels a bit in-your-face and sophomoric, and I'm sure that if I as an amateur writer wrote a book with the same philosophising, the book would be thought weaker for it.
But the book is still very readable and enjoyable, and later it explores an interesting and compelling moral dilemma, where, unsurprisingly, the robot's view is more black-and-white and absolute than that of the humans, and emotionally I sympathise with the humans, but I'm not sure I can rationally justify it. I think that part of the book is very well crafted.

Extraction - R R Haywood
Sci-fi thriller about three people who get "extracted": snatched away moments before their deaths by a time-traveller who's assembling a team of heroes to help stop the world ending.
The plot is a bit slow (it's a trilogy, and the first book just covers the team getting to know each other and training, before they go on any actual missions), but the characters are fun and the interactions between them are fun (lots of banter and snark). It's a bit violent and the time travel doesn't play as big a part as I'd ideally like, and I nearly stopped reading during a long, dull chapter of scenery-porn. But I enjoyed it enough to go on to read the next one in the series.
There's a Prime Minister who is never named, but the timing and the description of his looks and personality mean he basically has to be Boris, and he is portrayed very unfavourably indeed, as an openly-racist sexual predator who has an undignified and cowardly meltdown when the terrorists attack. It seems to be stretching the "any resemblance is coincidental" principle to breaking point, and I wonder what the outcome would be if the PM decided it was libellous.

Editorial aside: There was a bizarre kind of anti-typo, where one character said something like "He's going into hypothermic shock" and the other said "It's hypothermic, you idiot - hypodermic is to do with needles." I'm guessing the first character originally said "hypodermic shock", and either a program like Grammarly or a human editor who wasn't paying attention corrected it to "hypothermic shock" without bothering to read the following line.

Consider the Fork - Bee Wilson
A history of food and eating and cooking. Each chapter focuses on an aspect of food technology (cooking, cutting, storage, etc) and focuses on how it evolved from prehistoric times through to today. I find this sort of thing really interesting: looking into how the everyday innovations we rely on came about, and how they shaped culture. Thanks to toothycat for lending it to me.

The Other Daughter - Shalini Boland
Twisty psych thriller about Catriona, whose toddler daughter died unexpectedly. Confused by grief, she goes to a shopping mall the same day and abducts a similar-aged girl and brings her home, and ineffectually tries to convince her husband when he gets home that evening that it's their daughter. There's a scene which is like the inversion of the one in Little Face, with the mother insisting the child is their baby and the father insisting it isn't, and each calling the other crazy. The story is interspersed with chapters from the POV of Rachel, the mother of the abducted girl, living in a new town ten years later, and meeting a teenage girl who she thinks is her long-lost daughter.
The eventual resolution is very convoluted and maybe not especially convincing.
pic#57300910 triangle

More Zoe catchphrases

This is an update of this post from September 2017.

Good morning, lovely Mummy and lovely Daddy, my clock says seven... one... three. (with variations in the last two numbers). It's remarkable how this one has remained almost unchanged for over two years, just becoming more polite and... lovelier (previously "Hello, Mummy and Daddy, my clock says seven... one... three.") She still speaks the digits individually, even though she can read 2-digit numbers correctly most of the time.

Best! Multiple times per day, sometimes even per hour, Zoe will hide somewhere (sometimes in a very token way, like ducking her head behind the back of the chair) and jump out and exclaim "Best!" It seems to be a kind of nickname or title for me, Alex and Bethany; sometimes she'll say to us "Thank you, best" or "See you later, best." She likes to do this so much that sometimes she'll explicitly ask us to let her go into a room first and wait a few seconds before coming in ourselves so that she has time to hide and jump out.

If I could cuddle kiss you much I give you thousand cuddle kisses." I don't know where the weird grammar in this comes from (her syntax is normal the rest of the time), but she says it quite a lot with exactly the same wording every time. Bethany sometimes says it too. Zoe often says it when saying goodnight or when saying goodbye before school in the morning.

Yay, same! This is more of a shared Bethany-and-Zoe catchphrase. They use it the way people use "Jinx" or similar, to comment on when they say the same thing at the same time. They say it in quite a drawn-out way ("Yaaaay, saaaaame") and in a bit of a baby voice, so it sounds more like "Yay, shame."
pic#57300910 triangle

Twitter shoots itself in the foot

BBC article today: "Twitter apologises for letting ads target neo-Nazis and bigots".

"Twitter has apologised for allowing adverts to be micro-targeted at certain users such as neo-Nazis, homophobes and other hate groups.
The BBC discovered the issue and that prompted the tech firm to act.
Our investigation found it possible to target users who had shown an interest in keywords including "transphobic", "white supremacists" and "anti-gay".
Twitter allows ads to be directed at users who have posted about or searched for specific topics.
But the firm has now said it is sorry for failing to exclude discriminatory terms."

Putting aside moral questions of whether keyword-targeted ads should exist at all, and whether, if they do, they should be allowed to exclude certain keywords, and which ones... I think there's a really flawed linguistic assumption here.

No one says "I am transphobic" (except maybe very occasionally in a self-aware, self-deprecating way). They say "I just don't think it's right" or "Men can't become women" or "Bruce Jenner".

The word "transphobic" is used far more often by trans people and trans allies. "So-and-so used a transphobic slur." "Women-only domestic violence shelters are transphobic." "I am taking a break from Twitter because of transphobic bullying."

"White supremacists" isn't as much of an outgroup-term as "transphobic" - it's possible that some people do describe themselves as "white supremacist" in their bio - but I still think the term "white supremacist" is going to be used much less in tweets by white supremacists than by their opponents. White supremacists tweet about black crime rates or interracial personality differences; anti-white-supremacists tweet about white supremacists.

The image in the BBC article shows a data table with a "Keyword" column where one of the keywords is "homophobes". This is an even stronger example of the issue than "transphobic" or "white supremacists". At least with "transphobic", it's possible someone with anti-trans views could use the word, to write "I'm not transphobic, but..." or "Some people think this policy is transphobic, but actually..." Whereas "homophobes" is a noun - and a plural noun, so it's not even going to come up when people say "I'm not a homophobe, but..." or "Someone called me a homophobe". "Homophobes" (plural) refers to a group, and it's a group that's only going to be talked about by people opposed to it. "Homophobes" are the outgroup for gay people and their allies, just like "gays" or even "faggots" are the outgroup for homophobes. If you actually want to exclude keywords homophobes are using, you should exclude "gays" or "faggots", not "homophobes"!

Later in the article, it says "A campaign using the keywords "islamophobes", "islamaphobia"[sic], "islamophobic" and '#‌islamophobic' had a potential to reach 92,900 to 114,000 Twitter users, according to Twitter's tool." Same issue there, especially with the plural noun "islamophobes". And which do you think is a more likely use of the "#‌islamophobic" hashtag? "We are being overrun by Muslims with their barbaric customs. #‌islamophobic" or "Hate crimes against Muslims are on the increase. #‌islamophobic"?

Because I am in favour of free speech (as a principle and not just as a law), and opposed to misguided and imprecise crackdowns on disapproved speech, a part of me is mischieviously pleased that these measures will probably end up excluding the people who were trying to use them to exclude their opponents.

But another part of me would quite like to work on this problem properly, and apply data analysis techniques (which are more accurate than thoughtful intuition, which in turn is more accurate than thoughtless intuition that assumes groups tweet about the name of their group all the time) to see what keywords actually are most used by what groups.
pic#57300910 triangle

The gap in the abortion debate is wider than I thought

This is how I used to think the abortion debate worked: Everyone agrees that abortion means killing a living being of some sort, but they disagree on whether it's morally equivalent to killing a born human, a pet, an insect, or a bacterium.

But I've seen a few things that made me question the "everyone agrees it means killing a living being" part.

The first was an open letter written by a pregnant woman to the fetus she was aborting. The gist of it was: I'm sorry, I love you, but I'm not in a position where I can be a mother to you at the moment, but I look forward to meeting you in a few years when my life is more stable. I thought: no, you won't meet them in a few years, you'll meet their younger brother or sister. And maybe the writer was just using the word "you" in an imprecise, poetic-licence kind of a way. But then I saw some other things along the same lines.

The second was an article about a woman who had an abortion and went on to get married and have three kids, and it asked the question of whether it would have been better if she'd gone through with the original pregnancy, probably never met her husband, and thereby aborted the three kids she has now. As if "aborted" just meant "prevented from coming into existence". I'm pretty sure they didn't even put it in quotes or say "effectively aborted" or "as it were, aborted" or anything.

The third was someone using the word "aborted" in reference to a hypothetical future superintelligent AI if the biological species that would invent it is wiped out before they get that far. (I guess there's scope to debate whether they meant "aborted" like a fetus or "aborted" like a process, but I think they meant the former. And even if you abort a process, that usually means it's already started; it's not about preventing some hypothetical future thing.)

We don't really have a single succinct word for "prevent from coming into existence". I guess we don't need it very much outside of sci-fi, since usually in real life there's no occasion to refer to a specific entity that is prevented from coming into existence. (Either sci-fi like with the superintelligent AI mentioned above, or time travel; time travellers sometimes go back to eliminate Hitler or whoever, and they could do this by killing him as a child, or causing his mother to miscarry him, or keeping his parents apart so that he is never conceived.) But I think to some people "abort" does mean "prevent from coming into existence". So if the letter-writer in the first example believes she is preventing her hypothetical child from coming into existence, by the same logic there's no reason why the child she does bring into existence a few years later shouldn't be the "same" child.

So I used to think "life begins at birth" was a practical statement about the point where we as a society legally recognise life beginning, confer rights on a person, start counting their age, etc - but now I'm wondering if it's actually meant as an ontological statement about when life literally begins, and if, in the view of the people saying it, a fetus is as much a nonexistent, hypothetical person as a sperm and egg currently inside two people who haven't met.

And I used to think that the debate was between one side saying "Yes, we recognise that a woman's bodily autonomy is important, but we think an unborn baby's life is more important, so sadly we have to sacrifice bodily autonomy to save it" and the other side saying "Yes, we recognise that a fetus's life has some value, but we think a woman's bodily autonomy is more important, so sadly we have to sacrifice the fetus to preserve that." But actually maybe the second side might be saying "A woman's bodily autonomy is important - why on earth do you want to sacrifice it just to avoid preventing a hypothetical future person from coming into existence, when quadrillions of hypothetical future people fail to come into existence every day?"

(I don't have any clear-cut answers myself. It's a horribly difficult topic. I would definitely say it's killing a living being, and I think I'd tentatively say morally equivalent to a pet: not to be done lightly, shouldn't be done just because it's inconvenient to you, probably OK if done to spare it suffering, and people who are considering it are probably in a painful position already and should be treated with sensitivity. I'm somewhat reluctant to give my own views at all, even tentatively, because they might cause controversy and detract from the main point of the post, which was a new understanding of where some other people are coming from. But I was worried that if I didn't say anything about my own views, I would be taken as saying I had come round to the view that it's just preventing a hypothetical being from existing, whcih is not the case.)
pic#57300910 triangle

Coordination without communication

A topic I've become interested in recently is coordination without communication – in other words, two or more people managing to settle on the same answer to a question, when it's in all their interests to do so, without getting to communicate with each other about it.

This is closely tied to the idea of a Schelling point, which is a potential answer that stands out in some way, so that if you have no other information and no way to communicate with the other person, it makes sense for you to pick that answer, because the other person probably will too, for the same reason.

As a practical example, suppose you arranged to meet a friend "in Cambridge", but your phone died, so you can't contact each other to work out where in Cambridge to meet. Where would you go to maximise your chances of actually finding them? With no other information than that, I might go to King's College Chapel, as it's probably the most famous building in Cambridge and exemplifies "Cambridge", and is easy to find. If, instead, I had the additional information that we were specifically meeting to go shopping, I might pick the entrance to the Grand Arcade or John Lewis. Or if it was a friend I knew from a specific Cambridge-based context, I might head to the place where we first met or where we used to hang out.

More simply, if you're meeting in a big mostly-homogenous park and you haven't specified a meeting point, and there's a statue somewhere in the park, you would probably head there.

For a different example, consider a contrived game where you and your partner each have to pick a number from a list, without communicating, and if you both pick the same number, you both get a reward. For example: 2, 5, 9, 25, 69, 73, 82, 96, 100, 126, 150.

There are a few contenders: 100 because it's a nice round number, 2 as the only even prime, or 69 because it's a bit rude and also because of the Bill and Ted reference. The best choice might depend on who the other player was and what you knew about them - or you might not even have that information, which would make the choice harder. The number 7 is notably absent, otherwise I'd probably pick that, because it's the number people most commonly choose when asked to pick a random number - so picking it would maximise my chances of successfully coordinating, regardless of whether the other player was picking 7 naively or as a deliberate reference to the knowledge that it's the most commonly picked "random" number.

Part of the reason I find this interesting is that it's similar to the idea of currencies or commodities that have value only because everyone thinks they do, or everyone thinks everyone else thinks they do. There is nothing inherently superior about the number 69, but if you credibly believe the other player will pick it, it becomes the superior choice by virtue of that.

Coordination without communication often comes up in the excellent game Just One[*], which we've been playing a lot at GamesEvening recently. (Obviously there you're trying to pick different answers, rather than the same answer, but the principle is the same. If you prefer, think of it that the answer you're trying to converge on is "Alice gives clue A and Bob gives clue B".) In the game, it would be cheating and spoil the fun to communicate in advance and agree something like "I'll give the more obvious clue" or "You give the clue that comes first in the alphabet". But sometimes, very satisfying coordination-without-communication solutions emerge organically.

The best example was this: the word was "Cocoon", and Alex and I each independently went through this same thought process:
Chrysalis and Pupa are both the same sort of thing as Cocoon.
We each want to write a different one of those.
Chrysalis is also a My Little Pony character.
Alex wrote a fanfic about Chrysalis, so in some sense he has more "affinity" with her and therefore more "right" to give that clue.
Therefore Alex shall clue "Chrysalis" and Rachael shall clue "Pupa".

We were both delighted at our "telepathy" when we realised afterwards that we'd successfully followed this same train of thought independently.

A similar, but slightly less satisfying, example: the word was "Churchill". Based on the same "Chrysalis principle" above, all players agreed without communicating that it made sense for the players not named Churchill to give clues like "Winston" or "insurance", and for Alex and me to give clues relating to our family. But the principle didn't go far enough to specify which of us should clue "Bethany" and which "Zoe". (Thankfully neither of us has a favourite child.) I chose to write "Zoe", partly out of the vague reasoning that she's younger than Bethany and I'm younger than Alex, but it was a bit tenuous. As it happened, the two of us did coordinate and not clash, but I think that part was mostly luck.

A fun story that explores the concept of coordination without communication is The Demiurge's Older Brother by Scott Alexander, about a superintelligent AI negotiating a treaty with hypothetical other AIs it might meet in the future, by discussing precommitments with a simulated version of one of them.

It's also related to Douglas Hofstadter's idea of hyperrationalitysuperrationality in the Prisoner's Dilemma: the idea that, even though defecting is the "rational" choice (because if they cooperate, you're better off defecting, and if they defect, you're still better off defecting), cooperating is actually better if you model the other player as going through the same thought processes as you (including the bit where they model you as going through the same thought processes as them), and cooperating is the best outcome for two players both thinking in that way. It's as if you get to communicate, and credibly commit to both cooperating, but without actually doing so. (Rationality can then object: OK, but having reached that conclusion, I'll now defect, now that I know they've decided to cooperate, and then I'll get the best possible outcome for me. But hyperrationalitysuperrationality can counter: but, according to my model of them which thinks the same way I do, they'll now defect too, and we'll be stuck in defect–defect.) It's a weird kind of mental gymnastics, because it almost requires you to think that your choice actually influences theirs - but not quite. It's more like reaching an equilibrium that you and your opponent-who-thinks-like you will both be willing to stop at.

See also: one-boxing in Newcomb's Problem (that feels intuitively very similar to cooperating in the PD: two-boxing and defecting are more naively "rational" if you assume the other entity's actions are fixed and unalterable at the point where you make your decision, and doing so will get you a higher reward if that is the case, but one-boxing and cooperating make more sense if you assume that the other entity can predict you at least as well as you can predict them, and will make their decision based on their prediction of yours); timeless decision theory.

[*]Footnote about Just One: It's a fantastically good game, which I thoroughly recommend. It works with experienced gamers, non-gamers, and primary-school kids, and even mixtures of the above in the same game. We've played it almost every week at GamesEvening since we got it in May, and a lot with our kids, our parents, church people, etc. It's a simple cooperative word game where someone has to guess a word (say "Orange") and the other players each have to some up with a single-word clue to it (like "Fruit", "Colour", "Seville", etc). The twist is that if two or more players come up with the same clue, all copies of that clue get removed and the guessing player never sees it. So you have to come up with clues that are clear enough to convey the target word (perhaps only in conjunction with other clues that you hope the other players will give), but not so obvious that other people will give the same clue.

[Edited to correct terminology]
pic#57300910 triangle

Random thoughts on Frozen II (becoming more spoilery towards the end of the post)

I liked this a lot and am really looking forward to seeing the film again when it comes out on DVD.

Some random thoughts and observations:

I enjoyed the bit in the first song, which is about how things change, when Olaf turns to the fourth wall and comments that we all look older.

I love the charades scene. Olaf is brilliant at it because he can rearrange the parts of his body. Elsa is terrible at it, in a way that rings true with her being a bit uptight and socially awkward. She stands around dithering and making vague hand-waving motions in exactly the way people do when they're terrible at charades.

Olaf later puts his foreshadowed charades talent to use in summarising the events of the first film for the characters' new friends and any audience members who need a recap.

I'm not completely convinced by Kristoff's soppy boy-band-esque love ballad. He's supposed to be more rugged than that.

I also didn't like the farcical comedy about Kristoff unsuccessfully trying to propose, especially as it made Anna have to become the sort of woman who misinterprets every everything in dramatic and negative ways. Luckily she didn't stay that way for most of the film.

Anna knocks on Elsa's door in exactly the same rhythm as at the beginning of "Do You Wanna Build A Snowman?" You know how of you've heard a song a lot, when you hear the intro your brain starts playing the next bit of the song? But instead of the sad song about rejection, there's just a friendly "Come in!" The contrast is a powerful illustration of the change in the sisters' relationship.

I like the guy Kristoff meets who "talks for" reindeer in the same way that he does.

How do you make Elsa even cooler and more appealing to little girls? Give her a shimmery translucent water-elemental horse to ride.

At one point Elsa gets to see magical animated ice sculptures depicting past events, some from her own memory and some from longer ago (revealing plot-relevant information). She punches the statue of Hans and shatters it, and she looks away in embarrassment from the one of herself singing "Let It Go" in just the way that people do when seeing embarrassing old videos of themselves. Her attitude almost conveys that she knows every little girl in the country has been singing that song constantly for the last few years.

Spoilers follow
Collapse )
pic#57300910 triangle

Recent books

Island in the Sea of Time - S M Stirling
Great sci-fi book about the island of Nantucket, which comprises a small town, being unexpectedly transported back into the Bronze Age. The islanders (plus some people who were visiting) have to work together to survive and try to rebuild their infrastructure.
It's an interesting variant on stories about a single individual being transported to an earlier time - for a whole town to survive is easier in some ways and harder in others.
If the event had happened about a hundred years ago it might have been easier for them to be self-sufficient, because all their farming and manufacturing would still have happened locally. As it is, they have to get old industrial machines out of museums and try to get them working again. They also try trading with the nearby Native American tribes, and the ancient European tribes (there was a full-sized Coast Guard ship close enough to the island to get caught up in the event, so they can cross the Atlantic). The cultural and technological differences among the various tribes are well-portrayed: they're not all just rounded off to "primitive". Some have writing and some don't; some are patriarchal, but one is matriarchal and has no concept of monogamy.
Like in books about alien first contact, everyone is a linguistic prodigy. The Nantucketers learn some tribal languages, and some tribespeople learn English and each other's lanugages, and they mostly have little difficulty communicating fluently in all these languages.
Ooh! And I just Googled the title to remind me of the author's name, and discovered that there are two sequels! Looking forward to reading those :)

If I Die Before I Wake - Emily Koch
This is narrated by a young man who's in a vegetative state (which is like a coma but not quite) after a climbing accident. He can hear everything going on around him, including what the hospital staff say to him and about him, and what his visitors say to him and to each other. He can see a little bit, because sometimes his eyes open of their own accord, but he doesn't have control over his eyelids or focusing muscles. Based on the things he overhears, he begins to suspect that his accident was actually attempted murder.
This book was surprisingly reminiscent of Nutshell, the one narrated by an unborn baby. Both narrators are passive observers who have to piece together information from what people say in their hearing, and both are worried they or their loved ones are in danger but can't do anything to influence the outside world.

Little Face - Sophie Hannah
This was a very strange book. It's about a woman, Alice, who goes out for a couple of hours, leaving her new baby Florence with her husband David, and when she gets home she screams and insists it isn't her baby and someone must have swapped it. In the family home, not in the hospital like the last two baby-swap stories I read.
Collapse )