Come here Watson, I need you

About midway through the game Horizon: Forbidden West the character Sylens does something that appears on the surface so utterly stupid it feels out of character.

And immediately you are pushed into contemplating the Watsonian vs. Doylist question. Is there a in-world explanation for this baffling choice? Or is this something that was necessary for game purposes, necessary enough to supersede the lack of proper motivation?

Not that it really matters for that game. You can be as paranoid as you like, try to predict where the inevitable betrayal is coming from, but that betrayal is going to be on the other side of a cut scene -- you won't even be able to lay traps in preparation.

And, really, it doesn't matter so much in reading a book or watching a movie, either. That you know what is coming won't change the story that is presented. It will, however, change your experience of it. There is that perfect balance, sometimes referred to as "the expected surprise," when you can look forward to something happening with pleasurable anticipation and then get that jolt of satisfaction as it finally unfolds.

I say delicate balance because if you guess too early or too fully, then the revelation feels pointless when it finally happens. If you guess too late or not at all, there's less power in that key moment. The goal as an author is to ramp up the anticipation, whether it is a revelation or a long-overdue retribution, until it drops with the greatest possible impact.

Back to patterns. There's people (I've met several) who can watch a Perry Mason episode and know by the second reel who did it. There are two main strategies here. One is Watsonian; the desk clerk did it because nobody else would have had an excuse to open the cloakroom door after ten o'clock. The other is Doylist, but comes out of an instinct for structure. The murderer is the maid because she is the third suspect introduced, and the only suspect with a perfect alibi.

I'm calling this story patterns until and unless I find a better term. These are more than tropes or genre conventions. These are basic ways that story tends to get told, patterns that can be recognized.

And they are on a continuum. There are story patterns that come out of the needs of the media; such as recognizing a character will be important because he is being played by a well-known actor. And there are patterns that are part of the language of media and intended to be understood; like the soldier displaying a picture of his girlfriend. Story-telling shorthand, in other words; ways to inform the audience about the structural shape of the story without spelling it out.

The go from the near-universal expected to be grasped by all audiences, to the more subtle that require experience with that particular genre to read (those Perry Mason guessers were people who had seen a lot of episodes). So they aren't always read by all audiences. Not with equal ease.

***

That same day, I also started reading a new urban fantasy set in Paris. Almost immediately I had two Doylist realizations; the writer was not American. And the writer was also not French (turns out she's German born and now living in New Zealand). It was also pretty obvious the writer was female.

The European attitude is more subtle and harder to boil down to specific observations. It just didn't feel like the way an American author would approach it. The French thing was...it was a little too "look, here's the Eiffel Tower!" Things that were distinctly French and communicated that idea to the reader, but that aren't what a French person would think of as what was important to the story.

This one is a woman who can talk to ghosts (and unlike for Hotspur, the ghosts answer back). It opened in Pere Lachaise and I was already hooked. When it was revealed her day job was at the Pantheon...that's when I bought the book (the main action of the book, however, takes place in the Catacombs).

I was also admiring the experienced way the writer was building the story. The complications (the protagonist's relationship with her family, a suspicious cop, a veiled warning from Victor Hugo) were dropped at exactly the right places in page count and the rhythm of the story. This is another one of those expected surprises that come out of a well-established structure. You don't have to drop the complications -- you don't even have to have the body drop, or any of the other big ones -- but it is so satisfying to a reader when they are happening just when you anticipate they are going to happen.

It is like the experience of listening to music, when the chord sequence is pointing you towards a cadence that finally falls. After reaching all the way to the 9th or the needle tension of the dominant 7th, to finally drop back to the root. (Or to go somewhere completely different, if you are Sondheim...!)

***

I like reading the first book of a new series, but I think I like watching the opening of a television show even more. Because those guys are really, really good at the job. Introducing the world, the cast, the conflict. 

I got a few episodes into Continuum. There's a difference between an older series like Bonanza and a more recent series; the long form. Something like The Expanse doesn't have a status quo. You aren't expecting to find the same cast and the same situation. Take the Enterprise away and Star Trek stops being the same show, but take the Rosinante away and that story continues without a problem.

This means that, as in the self-contained form of the novel, questions are being continuously raised and answered. This also means that not all of the world-building is front-loaded, because those are some of the questions which are reserved to be answered slowly, as the series progresses.

***

It is good I am getting some reading done (and watching) because, since at least my nasty bout of COVID at least (and possibly since finishing the Paris book) I've been unable to write. Not at all.

More on that later.

Hugin and Munin were drones

Yes, that was actually said by someone. Ran into that "theory" at the APN (Archaeology Podcast Network), which I finally remembered to add back to my podcast feed.

I don't think I can pivot the Athena Fox stories. They've gone four volumes and given how slow I work, I'm not about to rewrite all of those from scratch. And, oddly, several of the comments I've gotten say that pivoting into more of an adventure/thriller direction is not what those commenters would want. A friend -- an American ex-pat I visited in Paris -- didn't know what to make of what she called the "James Bond" stuff in the last book. Good thing I didn't try her on the Japan one!

The books want to be mystery books spliced with travel books. They'd be better if they were less dense, and if I did rewrite from scratch, I'd make changes to my protagonist and her background to take it away from going quite so in-depth in culture and history.

Which would also be plausible with a bad idea I had just today.

The latest 'cast of Writing Excuses talked about opening a story with a thriller. Not necessarily a body drop; more like a Call to Action. I have known for a while, talked about on Quora, and made a conscious chose to start my first book with stakes. Only there weren't really stakes; Penny started the book with a goal, but the obstacle wasn't immediately obvious.

The other aspect that podcast got into is that this opening thing is usually a shaking of the status quo. It isn't "Here we are in a YA dystopia, things suck, we should do something about it." It is more like, "Here we are in a YA dystopia and oops -- the secret police are banging on the door." Meaning, we have joined the story at that moment where the characters literally can not go on as they have been.

Thing is, even starting with a body drop (as the New Mexico book will) is hardly the same as having the clear and present danger of a destructive force just uncovered and our heroes the only people in place to stop it. That is really so very much easier with fake history. Not just fake, but a particular kind of fake, where ancient aliens or items of power or a pharaoh's curse or whatever are, well, real.

Which I was loathe to do, which is why the Paris book never gets higher than the stakes of an idiot would-be treasure-hunter about to take a crowbar to one of the grotesques on Notre-Dame de Paris.

The bad idea is...what if they are real now?

As in, history is taught in that world the way it is taught in ours, and as with ours, it is largely correct. But something happened. A mad wizard did it (or in SF circles, Alien Space Bats). Now both versions are correct. Imhotep stacked a bunch of mastaba to invent pyramids. And Grey aliens beamed down to awe the puny humans with the ability to stack a bunch of rocks.

This way, we don't have to insult working historians. And our protagonists can declare their astonishment without looking like idiots who never noticed the actual suit of armor worn by King Arthur is on display in the White Tower.

But two big problems. One is on me. Not just that it is too easy to have the protagonists snark about how stupid the idea seems, but that it is very, very hard not to get dragged into all kinds of related story tropes.

There's a fun little two-book series by Seanan McGuire (Indexing) where fairy-tales are coming true in the real world. Warts and all; these are the Grimm versions indeed. But her hand-wave is like Sir Pterry's "Narrativium," where Story (as Seanan puts it) is an almost conscious and alive force that really wants the fairy tale to happen and to play out properly. So all the trophic elements happen -- weaponized by both the good guys and the bad guys, even.

Having a mystery thing happen and now Atlantis was real and divers can reach it is far too much temptation for a writer like me to have protagonists and others start reading the Evil Overlord list and weaponizing being genre-aware.

https://www.giantitp.com/comics/oots0763.html

The other is...

The really well-known pseudo-history and magical artifact and lost city stuff is, well, usually not good story. There's no internal consistency. Arthur above; even Mallory, combing every single Arthurian epic he could get his hands on and combining them into a coherent whole, was only somewhat successful at it.

As with conspiracy theories in general, the point is never about how Atlantis actually works and how it was hidden -- it is about how "mainstream science" are all poo-poo heads. The most common stuff is along the von Danieken ilk where it is a bunch of "you can't explain this!" thrown at a wall in hopes some will stick. There's no consistent through-line, no single underlying theory. 

(The Apollo Program deniers and the Creationists are so very much like this.)

Sometimes you do get a good story. If you are talking about the modern and specifically against-the-mainstream creations, the full phantasmagorical story of the land of Mu, for instance, has all the right cast of characters and geography and deep history and all of that.

As do older myths and legends, as inconsistent as they are.

There's pretty much shit-all for a well constructed story of how everybody managed to mislay a continent the size of Asia. 

Which suggests to me that the "suddenly all the myths are true" isn't a good way to construct a fictional universe that one can have adventures in. I think you need a spoof explanation for why many myths may have a basis in truth. Like the Stargate universe does.

The Adventure Continues

Horizon: Forbidden West went on sale during the massive and ongoing black-cyber-frimonday sales and I've started it. It looks gorgeous. But something has slipped in the character acting. Even compared to the pre-remastered Horizon Zero Dawn the characters feel less lifelike and less interesting.

A lot of this is intensity. Sun-King Avad is a bit of a dork in the original. He is young and inexperienced and doing his best. But the weight of his position gives him gravitas, he has charisma and he's hot. But above all that, there's intensity.

The HFW version he just comes across as a bit of a goof. Intensity is the same complaint with several other returning characters, even Varl (despite growing a beard). I've got a sneaking suspicion at the moment that this was intentional. That they told the voice actors to dial it down a little, possibly to make the later original characters look better.

***

Anyhow.

The thing about an adventure archaeologist character is that there are always ideas for them. Especially if you have already decided not to go the way of the greatest hits, because there's only a dozen big name artifacts and even fewer big-name locations to discover (and once you've done the Atlantis story, it is hard to top it with somewhere even more big important and magical.) There are, however, millions of ways a student archaeologist can get in trouble, even if they end up being more cozy murder mystery plots and the archaeology is tangential.

I know the things I don't like about the previous books, but I don't know how to pivot. Making a big pivot in a series is tough because unless you want to re-write all the early books, you are setting up for one audience then changing to a different audience. How can you introduce the new reader you are after when the first books aren't like that at all? And what about the readers that got interested in what was happening in the older books?

My version today of what went wrong is that the absurd detail and the way that detail is presented is baked in, and will be there as long as I continue with the format of throwing an inexperienced traveler into a new culture.

That's the biggest part of it. The first book was largely about Penny being overwhelmed, and gradually coming into confidence in navigating strange places. And I've kept that, at least as long as she is still going into places that are fully immersive; where she doesn't speak the language, where she is having to eat locally and sleep locally and otherwise deal with the unfamiliar culture 24-7. 

I backed off a little in the Paris book. There are several unusual things about the Paris book. She is largely in tourist areas and most of her conversations are with a fellow American. And her dip is less into modern Parisian culture and more into history -- and at that, it is art history, so further divorced from her Japanese experience of finding herself in a wooden room with tatami mat floors and going, "What am I supposed to be doing here?"

Thing is, I am also doing classic mysteries. My read is that the Cozy Mystery genre introduces a cast of characters with issues and that is the Gordian Knot that needs to be unravelled (cutting isn't usually allowed in those stories).

I don't know if there is a name or even a recognized genre, but I am writing mysteries where the place and the culture are the thing that has to be understood. The solution to the mystery in each of the Athena Fox books is reached through gaining a gestalt of a place and people. And the process of gaining that gestalt is through being a sponge. Learning everything she can because she doesn't know what the important stuff is yet.

Come to think, Asimov's Caves of Steel and some of Niven's Gil the Arm stories also hinge on grasping subtle elements of culture. Many is the case in those stories where someone says, "Belters don't do that because on a single-ship..."

This may change. The things that are at the top of my list right now for new Athena Fox stories have several that are a local sub-culture that can be experienced in small doses with a ready retreat back to the familiar. And Penny is gaining confidence and experience to where she isn't intended to be overwhelmed but instead has the tools to pick up what she needs and keep her cool.

***

So what's at the top of the list right now? I mean, I want to do underwater archaeology, and do the Holy Land, and visit Antarctica...but my list of plausible and might do them soonish is rather smaller.

The White Sands one. I've already backed off on trying to work in Old West stuff, or ghost towns, and I might have to put the UFO stuff further back in the mix. More and more, it is about that specific bit of geography and the various peoples that have inhabited it. Three in particular; the neuvomexicanos, who are connected to the pueblo -- mostly Tewa. The Los Alamos group. And the hominid who may have left footprints well before the Clovis peoples.

The Darien Gap one. Archaeological tourism, some mayincatec stuff (whatever seems appropriate) possibly the fake artifact trade, and a survival story.

The Minnesota Vikings one. Penny revisiting a different life-path by getting hooked up with folk music and Viking re-enactors, plus of course some pseudo-archaeology like the Kensington Runestone.

The one on a boat. The private yacht of a billionaire collector is in international waters and an eclectic group of feuding experts (and ringers and spies) are gathered to try and figure out which artifacts should be repatriated, and to whom. 

And last place is split between hanging with warbird fliers and the kind of WWII buffs who dance to big band music at the Hornet Museum, with of course an experimental flying machine unwisely named Icarus in the mix. Or, one about a brand-new science museum with a living exhibit on the space race and early visions of the future; L5 society, plus maybe work some Lustron Houses in there somehow.

Impatient Inpainter

I started a quick throw-away artwork just as a demonstration of some of the ways img2img and inpainting works within Stable Diffusion, via the Automatic1111 webUI (off-line installation).

I tried to pick something that was fantastic enough that it wouldn't get too hung up on trying to make the tech look plausible or the coke bottles the right shape, something that didn't have obvious ambiguities the AI language parser would get hung up on, and that wasn't layered or a complex pose or something that would be difficult to stack (like my first idea, which was a sort of typical golden age SF cover of a guy in a space suit with a blaster in foreground, a spaceship and barren planet/moon surface behind him).

So went for a retro, well, pretty specifically Love and Rockets sort of flying bike thing.

Here's what the AI is spitting out when I tried to do it with prompts alone (aka the txt2img mode):

Nice details and a lot of surface gloss, but what the hell is that thing? It's also not the classic racing motorbike pose that these flying motorbikes seem to attract. 

So better to rough out the composition in paint first:

That is actually too detailed. Really, the less you put in, the better the AI is at working things out. Among other things, your way of drawing a shape won't be its way of drawing a shape, and even to get at the same end point you are better off handing it less to begin with.

Similarly, the first prompt was pretty bare-bones:

(That checkpoint is a fork off SDXL, but doesn't require the weird two-passes approach of the latter.)

First pass of img2img, with a denoising of .4 or thereabouts:

Yeah, I ran it about a dozen times, keeping the one that preserved more of the details I cared about, and sacrificing others. That's the trouble with doing things in this mode, with a single pass.

The more powerful mode is inpainting and selecting just part of an image. For which you also get to change the denoising level, plus you can edit the prompt to reflect just the part of the image you are trying to iterate on:

Also an example of prompt engineering. Things like automobile grill or air intake wasn't leading the AI in the right directions, Fortunately the model I was using was sensitive to vintage stuff, so a reference to the famed Shure-55 (not by name), triggered the look I was after.

I also canned dieselpunk and Cadillac pretty quickly. The latter kept adding a caddy logo to things. The former turned out not to be in the language model but it did seem to be hauling "punk" out of it and was starting to add mohawk and piercings.

At various stages bits had gotten too far off what I was after. Like that fin -- that one, I used the built-in paint window in Automatic1111 to slap red paint over the hood ornament and forced a re-render of a different look.

A couple dozen iterations as I fixed some details and changed my mind about others, and it was more-or-less what I was thinking when I drew it. A final beauty pass at a denoising of .3 to get it to all gel together and, well, good enough for the exercise.

But...what would happen to it if I stayed in img2img and rolled the denoising up past .5 ? (as a very rough guide, and it depends quite a bit on the model and less, but still there, on how easy that specific image is for the AI to interpret, up to .3 is a "clean up," at .4 it begins to change things, and somewhere just before .7 it "snaps" and gives up completely on respecting the original image, making up something new that only vaguely resembles the colors and masses.)

Ooh, an unexpected prompt-ambiguity lesson. "Riding" a flying bike, indeed. Yee-hah! (Plus, where the hell did the AI get those feathers from?)

But this does show the basic ideas are imbedded into the training data; there are stereotypical elements like the hotrod paint job or the fluffy white clouds that the AI puts in there even without me specifying those (the first image, after all, did not include the cloud background in the prompt. The AI put it in because that's what this sort of image usually gets).

So plausible. Try a few more runs and see what happens:

This one had promise. You can see both the bane and the boon of AI here; not just the details, but the angle/perspective that gives it such a better look...but yet, the perspective is messed up, and some of the elements are contradictory. 

So back to inpainting. That fin wasn't working, the jets had gotten just stupid...but I knew those were easy fixes. The "magic words" to fix the air intake, after I'd roughly painted out most of it with blue sky and sketched in a better shape, was "P-51." For the jets, "contrail" and "afterburner," even if I did have to make a second pass to remove a Blue Angel that snuck in there.

And that is where I really stopped.

(This is AI, and furthermore, the modified SDXL and one LoRA I used during a few trouble-spots -- not specific to flying rockets or retro SF in particular, just one that was good at this sort of illustration look -- are via Civitai therefore even more copyright-violating than the original SD training data. I present these under the shade of "academic or criticism" as my sop towards fair use.)