Piranesi

Since I’m still awake, I’ll write another review, this one of something I liked: Susanna Clarke’s Piranesi.

If you know nothing about it, it’s a fable, something of an urban fantasy. The title character lives in another world, which he calls the House– because it’s all one house or mansion. It extends for thousands of rooms in all directions, and despite years of exploration Piranesi has never found an end to it. The lower floors are filled with oceans; the upper floors with clouds. He lives off fish, shellfish, and seaweed he finds in the ocean, and he keeps an obsessive journal– the novel purports to be its 10th volume.

For company there are a few skeletons, and a mysterious man he calls the Other– the only living human he knows. The Other shows up twice a week for meetings. The Other is pursuing what he calls Great and Secret Knowledge which he thinks is hidden in the House; he does not search for it himself, but encourages Piranesi to explore. (He provides him with notebooks and pens.)

The first part of the book explores the House, the strange narrator, and his strange friend. Piranesi, oddly enough, is completely happy with his life. He has excellent recall of everything he’s seen in the House, and he’s satisfied with his daily routines and occasional longer journeys. Every room is full of statues, and he knows them all. He talks to the birds which fly through the halls, and leaves offerings to the skeletons. He regards the Other as a friend, though he is skeptical about the Great and Secret Knowledge.

This idyll is threatened by new knowledge– starting with a visit from another living person, who Piranesi calls the Prophet. The Prophet tells him things that put in question what he knows about the House, and the Other, and himself. He begins to explore these clues…

I won’t say here what he finds, except for what the book jacket reveals– that there is another world besides the House. And magic is involved.

The obvious comparison is to Borges’s Library of Babel– though the House is more a catalogue of the visual than the literary arts, and was more densely populated. It’s also reminiscent of Schuiten & Peeters’ brilliant and gorgeous French graphic novel La Tour (The Tower), which also depicts a near-infinite architectural monstrosity with few human residents and a mysterious origin.

Reviews usually mention C.S. Lewis’s The Magician’s Nephew, probably because it contains a few explicit references to it, and because the House also recalls the huge palace of Charn. But the comparison is not very illuminating. Piranesi’s House is not Charn: it’s not lifeless, it’s not the sinister end of an evil civilization, and there are no traces of Aslan or the Witch here. There is some human evil, but the House itself is– at least as Piranesi experiences it– a peaceful and even joyful place.

What it’s not much like is Clarke’s first novel, Jonathan Strange & Mr Norrell. I liked that one, but it takes a long time to get going, while Piranesi is just 245 pages. One thing they do share is that in both novels, Clarke commits to the bit, as comedians say. Strange is written in the form and diction of long 19th century novels, while Piranesi showcases the oddness of its main character, who learns all sorts of things during his adventures but never becomes what we’d call normal.

I liked the book a lot, and finished it in one long binge. Most everything gets explained eventually, at least one of the bad guys is dealt with satisfactorily, and Piranesi manages to adapt to his new knowledge without entirely losing the calm he found in the House.

It doesn’t explore its magic very thoroughly. I don’t think that’s a flaw, but it’s good to set expectations: if you like your sf ideas teased out in all their complications, this isn’t that.

I do think that– as with The Tower— the first half of the book works the best, when we are most exposed to the awesome and strange world of the House and don’t understand it yet. Schuiten & Peeters couldn’t really make the mystery pay off. I think Clarke does better at that, but some of the magic does leak out. One disappointment, perhaps (in white to avoid spoilers)… the Great and Secret Knowledge turns out to be a dud. Which is realistic, but it might have been more interesting if the Other’s project was more of an actual threat.

Sable: ugh

Negative reviews are kind of annoying; but I’m a bad mood and might as well express it, and maybe draw some game design lessons.

I’ve been looking forward to Sable for years. I tried it tonight and bounced off it so hard I got a refund.

From pictures, it looked like it took a lot of inspiration from Moebius. In-game, it’s not quite as impressive: it’s all flat colors and the overall effect is to turn a complicated 3-D scene into simple flatness. And the cutscenes for some reason are low-FPS. For what it’s worth, Moebius usually modeled his shapes with meticulous linework and subtle coloring. But eh, not a big problem.

The big problem is the terrible UI. This is foreshadowed by the moment you get control. You’re in some sort of temple, facing a big sculpture of a face. You can walk around, climb the face, stand on top of it looking outside through a hole in the roof you can’t get out of. Turns out you’re supposed to ignore all that and walk out the other way. Lesson 1: when you highlight something with details and lighting, players will think it’s important and spend time on it. Don’t waste those cues on nothing.

I walked out and ran into a child who offered something I was told I wanted, in return for some beetles. Fine, it’s My First Fetch Quest. You can talk to other people and they’ll give you a vague hint where the beetles are (“go east”). Fine, only…

  • There’s no indication of where east is. It turns out you need a compass, which you get later on.
  • I went east and saw no beetles. The compass highlights points of interest, but there was nothing to indicate where the beetles are. It’s a big world, you can lose insects in a lot of places.
  • Everyone you talk to will give you that same option to talk about the beetles, which you can’t skip.
  • Since I’m in a bad mood, I might as well complain that no one explains why I want the thing the child is offering. Is it a side quest, or something I need to advance the story? No idea.

Lesson 2: Playing hunt-the-pixel was tedious even back in the ’90s. Don’t be mysterious about what you want from the player.

Next, I talked to someone who was supposed to have a glider for me. Cool, supposedly this is the key to the whole game. It’s not ready yet for some reason… fine, it’s a multi-stage quest. But the first step is, he wants you to ride a beater glider to test it out (i.e. preview the skills). The conversation implies that it’s right next to him.

The glider isn’t there. All that’s around him are a few boxes. You can wander around camp or the surrounding desert… there’s nothing that looks like a glider. You can talk to people… no hints. You can bring up the compass… nothing points to a glider. You can talk to Glider Guy again, and he explains where the beetles are, then tells you he won’t talk to you till you’ve flown the glider which he won’t tell you where it is. There is no option to skip the beater glider and get the real one instead.

I looked around for awhile and gave up. If it’s a bug, it’s pretty bad that it makes the tutorial fail. If you’re supposed to wander around the big area available to you until you find it… come on. Lesson 3: don’t hate the player that much. When you’re obviously gating further progress to a task, don’t fucking hide the thing in some non-obvious location.

I did read a few reviews, and obviously some people are progressing easily enough. That’s nice. But to me, the time I spent with the game told me one of two things.

  • Maybe the developers really do hate the player– it’s supposed to be a frustrating grind. In which case, I’m glad I found out within the refund period.
  • Maybe the developers didn’t test their game. Like most developers, their idea of testing was “I ran it once and it didn’t crash.” They know where the damn glider is, so they don’t see a problem. Did they try watching someone else play their game? Even for an indie studio, that level of non-testing is not acceptable.

I’m mildly curious, but very mildly, where these things were. But that’s lesson 4: put the frustrating bits later in the game. Once I’m committed to a game, I’m willing to put up with grind. (It’s astonishing how much time I’ll put into grinding in Minecraft.) But the first hour or two of a game is key to making me feel committed. At that point, you’re still selling the game. Make it interesting and don’t make it impossible.

One more thought: a game can decide that a particular quest item shouldn’t just be highlighted on the map, but there are alternatives besides “randomly hide the thing in a large area and don’t show any clues at all.” E.g. the minimap in Borderlands will highlight you the area where you should hunt. Dishonored had the Heart which gives you the distance to the sought item; by heading to where it beats faster you get the direction. Minecraft has you locate strongholds by throwing an item which will point in the right direction.

The Psychology of Time Travel

I found this book, by Kate Mascarenhas, more or less by chance. It was on the next shelf from where Arkady Martine’s next book should have been, and it was about time travel– I’m a sucker for anything about time travel. And in fact it’s really good! Once I was into it I had to keep going till I was done. And now I’m out of books.

The basic setup: four women– Margaret, Barbara, Grace, and Lucille– invent time travel, in 1967. One of them, Barbara, is overstressed and has a sort of breakdown in front of the media, and she’s kicked out of the group. The rest form an organization to manage time travel, the Conclave.

In July 2017 Barbara receives a cryptic note which turns out to be the report of a death six months in the future, and discusses it with her granddaughter Ruby. In January 2018, a girl named Odette finds a corpse in a locked boiler room… locked from the inside.

So, it’s a mystery, and an sf story, and true to its title it really is about the psychology of time travel: how it might mess with your head. What happens if you know the day you die? Or the day your loved one dies? More strangely yet… how do you grieve, or do you grieve, if after they die you can, whenever you want, take a trip into the past to see them again?

Fitting the subject, the structure of the book is all over the chronological map. Most of it centers on the murder and its aftermath in 2017-18, but chapters are set in the previous or following decades. There are quite a few characters, though the chief ones are named above.

Mascarenhas’s version of time travel is deterministic, and also future-oriented: time travel requires a receptive apparatus, so you can’t travel back before the device was invented. (You also can’t travel more than 300 years in the future. It’s hinted that the 24th century is pretty nasty, and perhaps all the machines are destroyed.)

Most of the focus is on the Conclave itself. Its structure is a baffling, because time travel sort of collapses its 300-year timespan. Any given agent may be given an assignment at a future or past time; people get intimately familiar with their past and future selves; you can even make a phone call to any Conclave employee at any time. There’s an extensive Conclave slang, which never changes since it’s shared over that entire time period. There are objects called “genies” which are acausal: a future you hands it to a younger you, so it exists uncreated in a years-long time loop.

A major subplot is a romance between Ruby and one of the pioneers, which has the brain-busting peculiarity that the lifespans of the two characters barely overlap. Romance is weird for time travelers: if they end up with a partner, they know who it is, often before they’ve met. Also, is it infidelity if, while you’re partnered, you also hook up with yourself?

The Conclave also turns out to be kind of a nasty thing. It’s located in London, but it’s outside the jurisdiction of British law, since the agents present at any one moment in time may be from anywhen, and it’s not clear what set of laws should apply. And it reflects the heavy hand of Margaret, at the top, and her determination that psychological problems like Barbara’s never recur. Naturally, worrying so hard about one problem leads into a set of opposite problems.

The book must have been hell to plot. It’s a lot of fun to explore all these concepts, and almost all of the characters are interesting to be with. (All the viewpoint characters are female; from an interview, it seems that the author tried male characters, and found that readers took the male characters as more important. So she just made everyone important be female.)

Mascarenhas works out lots of weird side-effects of time travel as the Conclave practices it, though I’m not sure they’re worked out enough. To try to explain without spoilers: information about future events is a phone call away. Sometimes the characters use this information; and to make the plot work, some of the characters (Barbara, Ruby, Odette) spend much of the book outside the Conclave and thus have to plod through normal time like regular humans. But some events proceed as if the Conclave weren’t using its own mechanisms. (Though, the timeline being unchangeable, perhaps the ultimate argument is “things happened that way because they did.”)

An example with spoilers: Odette joins the Conclave to dig up info. Because she had therapy, she is ineligible per Margaret’s rules, and she hides this for a time. When it’s revealed, she’s kicked out. There’s a testing process for entry, lovingly detailed; why isn’t part of it calling the future to see if she’s still employed in a month? There may be answers in this particular case– e.g. Odette is hired as a sort of internal detective, and perhaps policy is to not to mess with them. But the same issue comes up with larger plot points. E.g. after 2018 everyone knows that Margaret is a bit of a nutter. How could this be kept a secret before then, when travelers are constantly going back into her tenure? I don’t think these are flaws, it just worries me a bit.)

Buggy book

At the library I picked up a book on history which I hoped would be dull but informative– perfect for reading at the gym or just before bed. Almost the first thing in the book is a map of the ancient Middle East, which shows Sumer… as a city.

Oops! But maybe this was a singular error and the text would be OK. But no, it turns out to be full of little errors.

  • The author says that both Egypt and Mesopotamia tamed their rivers by means of elaborate waterworks. That’s more or less true of southern Mesopotamia; not so much Assyria or Egypt. Egypt barely needed waterworks: the entire Nile valley flooded every year, and the Egyptians just had to recover the locations of their fields after the flood.
  • He has Egypt founded by Menes, not Narmer. He describes the Nubians as being too weak to resist Egyptian domination, though in fact Nubia conquered Egypt at one point.
  • He describes the Amarna letters as showing the supremacy of Egypt over all other kingdoms, and to prove this quotes a letter from an Egyptian vassal. In fact the letters show that the major states were and presented themselves as fully equal to Egypt. A little clue that Egypt wasn’t top dog is that the letters are written in Akkadian!
  • He has the Sumerians as successors to the Akkadians, and declares that the Sumerians ruled from Ur. In fact the Sumerians came first, and were notoriously divided into fractious city-states. He talks about the Assyrians inheriting culture from the Babylonians– he doesn’t seem to realize that they spoke the same language.
  • He says that Babylon never recovered after the fall of Hammurabi– though in fact it was a major player for another thousand years, usually able to stand up to Assyria, and finally triumphant over it. He mentions the Kassite takeover of Babylon but doesn’t seem to know that the Kassites ruled peacefully for almost 500 years.
  • He summarizes Hinduism as the worship of Lakshmi and Vishnu, which not only erases Shiva but forgets that there was a whole different pantheon in Vedic times.
  • He says that India’s major external threat, until the Europeans arrives in the 16C, was the nomads of Central Asia. Which, um, forgets the Persians, the Greeks, and the Arabs.
  • He has the early Chinese surrounded by exotic peoples named Xirong, Beidi, and Nanman. This is a confused and truncated reflection of the Chinese giving a name to the barbarians in each direction: Róng to the west, Dí to the north, Mán to the south, and Yí to the east. The author’s versions redundantly add the Chinese direction name: Xī-Róng = ‘west Róng’. None of these names should be taken as ethnic groups or nations.
  • He uses temple names for (medieval) Chinese emperors, which leads to absurdity when he mentions a Khitan emperor crowning himself Taizu. First, temple names are posthumous; second, Tàizǔ is the traditional temple name given to the founder of a dynasty; it’s more of a title (‘great ancestor’) than a name.

Naturally, my upcoming book won’t have these errors. (Of course, in my less self-confident moments I worry that it will be riddled with other errors. But hopefully they’ll be more interesting and better-informed ones.)

I’m not going to name the book, because my intent is not to shame the author. I don’t expect to finish it, anyway, because he’s lost my trust. I expect he’s reading way out of his field, and he is trying to talk about all of history, which implies a pretty overwhelming research load. But still, didn’t he have any specialist friends who could give the manuscript a quick read?

Oh, one more complaint– he talks about a Middle Eastern “dark age” from 1200-1000. That’s not an error exactly, but it’s kind of an outdated way of thinking. In many regions historians focus on strong, high-profile states… largely because these are the places that left impressive ruins and lots of documents. It’s fair to say that these were nice times to be alive if you were in the elite and lived in the capital. It’s far less accurate to say that these were the best times for the rest of the population.

Take, say, the Old Kingdom of Egypt. We can trace the rising power of the leaders from predynastic times, culminating in the near-totalitarian power of the 4th dynasty (which built the Giza pyramids). Centuries later, the power of the kings declined; tombs of provincial leaders increased in size; there was a tendency toward regional art styles. Finally the central power collapsed.

Was this a “dark age”? For the royal family, sure! But arguably prosperity was spread far more equally at the end of the dynasty and beyond it, and the common people were perhaps better off when they weren’t being forced to build pyramids.

We get the “dark ages” concept from European history, and European historians’ eternal disappointment at the fall of Rome.

  • The prosperity of the Roman empire is frequently exaggerated, especially in the West. Urbanism didn’t have deep roots in Gaul or Spain, and largely collapsed when central authority did.
  • Rome was really pretty badly managed. There was never really a golden age of Roman republicanism, and the emperors operated at about the level of Third World warlords.
  • The collapse in power and technology we see in the western half of the empire should not be imputed to interregna elsewhere, e.g. China and India. A set of small kingdoms ruling where there was once an empire is not inherently bad.

The Collapsing Empire

This, by John Scalzi, was another of the books recommended by NPR, but it turns out to be very similar in theme to A Memory Called Empire. It’s also a space opera about (spoiler) a collapsing empire, and even explores the same idea of a sentient brainscan. What I learned: it’s best not to read two space operas in a row about collapsing empires.

It’s not bad, mind you. The basic setup: humanity lives in the Interdependency, a network of colonies dominated by guilds (basically megacorps with monopolies), linked by a para-space called the Flow that provides FTL travel. The problem is that the Flow is disappearing, which will be particularly bad because human society has been designed to be interdependent– so the colonies will probably die off on their own. Only one colony is an actual planet, called End.

There’s a number of viewpoint characters: Kiva, a roguish and foul-mouthed Owner’s Representative on a trading ship at End; Cardenia, the new and unprepared Emperox of the Interdependency; Marce, a scholar from End who has the best insight into how the Flow is failing.

Let’s start with the positive: the prelude, which sets the tone and the theme. It details an attempted mutiny on yet another spaceship. It’s fun and showcases what Scalzi seems to do best: tough asshole characters, tense but witty confrontations, quick reversals, and a good helping of comedy. If you like that sort of thing, you’ll like the book. And the series; it’s a trilogy.

The overall situation is good too, though maybe it hits a little close to our little problem, the looming climate change apocalypse. The Interdependency feels like a bunch of overgrown Renaissance city-states, everyone trying to screw each other over without quite violating social norms. Throwing an existential threat at these people is an excellent way to see what they’re made of.

So, I like a lot of it but I also find it tiring. The comparison to Arkady Martine does not help it. I think the things that rub me the wrong way are these:

  • The characters are divided into good assholes and nasty assholes. (Not everyone, but close enough.) I actually like Kiva a lot– she’s lively and inventive. But, I dunno, there’s a reason Star Wars didn’t consist only of Han Solo and Boba Fett. It’s nice to have some actually likeable people in there somewhere.
  • The multiple viewpoint characters are a good fit for space opera, but I miss the focus provided by a single protagonist.
  • The Interdependency as Scalzi portrays it is hard to like– Martine is far better at explaining why an empire could be both dangerous and attractive. But the deeper problem is that the systemic problems are sidelined in favor of building up one particular clan as the archvillains. That is, Scalzi understands that the Interdependency is hopelessly corrupt, but still writes a story where shooting three people would pretty much fix everything.
  • Come on, Scalzi, you literally give your exposition of the Flow in the form of a lecture for schoolchildren?
  • The structure is weirdly digressive and repetitive. E.g. something like a third of the book is devoted on getting Marce from Point A to Point B. He’s threatened at home, then kidnaped, then they send assassins after him, then pirates. It’s competent thriller-plot, but did we need all four of these inconveniences, especially when the bad guy is the same one each time and we learn nothing new about him? The Chandler plot, careening from one danger to another, doesn’t work merely because it puts the hero in danger, but because it deepens the plot and provides surprises.
  • One more unfortunate comparison: Martine is far better at showing how it feels to be involved in apocalyptic events. It’s fun when a resourceful character like Kiva keeps sidestepping the enemy, but it’s more affecting when a character sometimes feels overwhelmed and panicky.

I’m afraid this sounds more negative than I meant it to. When something feels off about a book or game, I want to analyze that until I feel I understand it. But if I’d read this book first, or been in a slightly different mood, I’d probably have liked it more.

FWIW, re-reading my review of Scalzi’s Redshirts, I note that I had a very similar reaction: great idea, weird structure, and a certain difficulty in raising the emotional temperature.

Sales Report 2021

I haven’t done one of these in awhile. The kobolds in the Accountancy Wing of the Zompist Fortressplex have produced a chart on book sales for the last ten years. (Kobolds are cheaper than goblins, and easily replaceable after infestations of first-level adventurers.)

Total sales are just over 41,000 units. Of that, 25,000 are paperbacks, the rest are e-books. The LCK has sold about 18,000, which is quite respectable for a book. That probably means that it’s inspired at least 30,000 conlangs, which is a little frightening. (Obviously, many readers never produce a conlang, but Conlangs Georg, an outlier, creates one every day.)

The to-date chart is skewed a bit toward older titles. Month by month, the Syntax book does as well as the other linguistics books.

If you’ve ever wondered why authors keep writing what they’ve always written, look at the smaller slices. The best-selling novel, Against Peace and Freedom, has just 384 sales. I feel like I just can’t write more fiction yet. The history/language books do better but not great, but I can’t write an endless series of linguistics books.

(Oh, if the order in the pie chart is baffling, it’s chronological.)

If you’ve bought any books, thank you! Buy some more, they’re priced to move!