Raindrops: December 2024

A spineless new publishing world and some good old-fashioned animation magic

It’s raining right now here in the Bay Area. Snowing up in the Sierras. Guess where I’d rather be right now.

The image above is a photo I took right after a snowstorm in Lake Tahoe a couple of years ago. Coming from Central Europe, some of my earliest memories are of snow-covered pine and spruce forests, the soft crunch of fresh powder underfoot, and that warm stillness as everything around you sleeps deep…

Raindrops explores the fields of Art, Music, Writing, and Film. In this month’s edition, we raise a well-plucked eyebrow at yet another attempt to disrupt the imperturbable industry of publishing, and pop the cork for some exquisite French-American magic.

Let the snow fall where it may.

Writing: Publishing

What’s shiny and new, can process natural language, and takes gigawatts of energy to spit out text and images in just minutes?

Why, generative AI, of course.

What’s not so shiny, starting to show its age, and sucks in millions of VC capital to pump out ocean-size volumes of hype?

Why, genera— er, tech startups trafficking in AI.

If you’ve got half a fin dipped in the publishing world, you’ve likely heard of Spines, the latest self-proclaimed disruptor of the publishing industry. The irony of the name is too much for me to bear,¹ so let’s move along to commentary. Normally I wouldn’t want to give an outfit like this more press than they’ve already paid for, but it is useful to put this startup and its ambitions into some perspective. The company’s About page begins with an aspirational albeit inaccurate observation:

“The world of publishing has changed little since the days of Gutenberg and beckons for transformation.”

That line alone beckons for some internal transformation, namely, hiring someone who actually knows about the evolution of publishing. They’re also about a decade behind the last crop of publishing disruptors. I met one of them in person here in Silicon Valley back in 2012, and that onion got peeled real fast. (Reminder to self to cover that story in a future installment of Raindrops.)

The Spines “How it works” video intentionally horrifies: Writers beware! It takes 6-12 months to publish, and can cost you up to $15,723! As an indie publisher myself, I can confirm that it does not have to take this long or cost this much. At the same time, 6 months isn’t all that long—wouldn’t you want to take the necessary time to do it right? You know, work with a professional editor, get the cover design done by an actual artist, craft a proper marketing strategy, line up reviews and interviews…

Quality isn’t the point here though. Speed is. The faster you can get your words out there, the sooner you can call yourself a “published author.” You don’t really need to be a skilled writer to publish a real book—Spines will help you succeed “regardless of whether you’re taking your first steps in writing [my emphasis] or are an established author.” I guess that means first graders can sign up. Note that nowhere is “success as an author” specifically defined on the website. Is simply laying out a manuscript in physical book form “success” regardless of the quality of the writing, sales, or critical reviews? Apparently yes—they’ve even got a handy guide for turning any manuscript into a “published masterpiece.”

The engine powering this stunning machine of innovation is AI. We’re not told what type of AI it is exactly, but we’re assured the AI can generate a cover design, proofread our manuscript, format it, and if we want an audiobook version, it’ll do that too. And it will produce the meta data needed to distribute your precious masterpiece throughout the world, “primed to mesmerize readers worldwide.” In short, the AI will drive “cost efficiency in production.” In their words:

“What if we could combine the old standard publishing process with today’s advanced technology platforms, and simplify the process?”

Mind blown. Why didn’t anyone think of this?

This type of operation is, in very simple terms, just another vanity press. A vanity press turbocharged by generative AI. At its core, there is nothing groundbreaking about Spines.

The very fact that the startup raised $16 million in Series A funding—and who they raised it from—should speak volumes to their underlying ambitions. Pay attention on where the focus of the marketing lies—you won’t see any love lost between their business model and the authors whose work they package up. The $16 million isn’t going to build a new type of publisher to rival the likes of the big houses in New York, or to publish a particular genre or style of stories that might, you know, do silly things like make a meaningful impact on society. It’s VC money going for scale. The kind of scale, the more automated the better of course, that will plump up the ROI the investors are looking for. The more authors paying the publishing fees, the fatter the profits. 😋 Whether or not the authors make any money from their books, is the authors’ problem, not Spines’.

The vanity press gets its fee and its “customer,” and the author gets a book made—that they have to market and promote themselves (Spines glosses over the small detail of book promotion when it talks about distribution). The vanity press could not care less about the quality or impact of the book it produces; it’s performing a service in exchange for payment. The value of a startup, in investors’ eyes, lies in how much revenue it can generate from its customers, not how much revenue the startup can help generate for its customers. See the way this river flows?

If it wasn’t clear before, it should be now. They tell us who they are:

“Spines – the same old story. Retold.”

Indeed.

Film: Animation

This would have shipped with last month’s Raindrops, but I had to finish watching the show before writing about it. It’s an animated series inspired by the massively popular video game League of Legends. Not the sort of production that would typically catch my eye, but…

Arcane is BRILLIANT.

Hands down.

The Netflix trailer doesn’t do it justice.

I’ve got my 14-year-old daughter to thank. My first time, her third. We’re both obsessed. If you’re human, highly recommend it. If you’re a bot, don’t bother (the Arcane will glitch you out of existence).

Promotional poster for Arcane. Source: League of Legends Wiki

For me, Arcane’s brilliance is embedded in the characters. No one is black & white evil or good. They are all complex, nuanced people, with intertwining back stories and histories and inner struggles and relationships, from the platonic to the professional, from the romantic to the spiritual. Even the things that feel cliché, like the Christian archangel-god references, turn out to be surprisingly profound once you realize your perception of the cliché was intended precisely so for a reason, and there’s a deeper layer of meaning you have to dig up.

The art of Arcane is a blend of 2D watercolor & oil and 3D modeling, all done up steampunk slash Art Deco style. Every frame is a painting. The details on everything, rich but not overdone. You can watch it with the sound off and you’ll still feel like you attended a banquet. It can at times feel like sensory overload—the art, the movement, the sound, the dialogue, the narrative tension—but that’s the point. You get dropped into Arcane’s world from Frame One and you’re only released at the end of Season Two. By then you’ll have morphed into a more evolved version of yourself, forever transformed.

What fascinates me is the process that took characters from a video game and crafted a narrative for an episodic series. I’m not a video game player, but I have no trouble dropping into the world Riot and Fortiche have created and hitting that ground running. Game or animated series, story is story. Perhaps it’s a different experience when you’ve played the game, and then you watch the series. Or vice versa—although my sense is that it would be stranger to watch the series first and then play the game, which might make for more of a narrative-cognitive dissonance.

I’ve been hearing rumors of a Season Three, but those rumors have been put firmly to bed by Marc Merrill, Riot Games co-founder on Reddit [emphasis mine]:

The confusion is because internally there was a budget conversation about “approving 5 seasons” - which simply means we were setting aside a bunch of money for lots of development and is completely irrelevant to the Arcane creative. …

Just because that is our intent, doesn’t mean it’s easy or that we can pump these things out like a factory - just like making games that are worthy of players time / love is really hard, so too is telling great stories and making incredible shows / films (especially when we are building it from scratch for the first time).

He then adds a brutally honest riposte [again emphasis mine]:

Some other notes that may be helpful for everyone:

- All of Arcane’s budget goes to talent. animators. concept artists. voice actors. sound designers. story boarders. composers. etc.

- The character animation alone makes up about 80% of the budget, because we hand animate every frame to hit this type of quality.

- Anyone complaining about the development budget being high is actually complaining about us paying talent, and complaining about us putting art first. Arcane is a great thing for every person working in animation as it helps point towards the value of higher budgets.

Heard that, Hollywood execs?

These days, anytime a studio releases an animated film or series, the question isn’t whether they used CGI. The question is whether they used generative AI. It’s fairly obvious Arcane was not animated using gen AI. The art is too organically beautiful. But don’t take my work for it… let’s hear directly from Pascal Charrue, one of the co-founders of Fortiche Production, the studio that worked with Riot Games to create Arcane—see the video above. The interview is in French (that in itself makes the interview worth a listen!), but, subtitles in English, so you’re good. At 08:36, Pascal says:

“This question [of AI] is always a little… we get asked this a lot. It’s a bit problematic, this question, this topic. I remember a few years ago when Arcane first came out, the first question we were asked was, did we use AI. I said, ‘Absolutely never!’”

Yet he remains open, as long as it’s used to support his artists, not replace them.

“For us, our vision, the way we make films, it’s ‘artistic first.’ The image can be reproduced, but the intention behind it, the thinking behind it, no. Even if the prompts are so good that someone else can do it, that person will end up spending the same amount of time as our artists. I’m not saying we don’t want to take on the topic of AI at all. We’ll try to use AI as a tool, much like we were able to use 3D when there was really only 2D. It’ll be a revolution.”

“But for us, our goal will be to enable our artists to save time. … We’re working on trying to… reduce the time our people have to spend working. If tomorrow we’re given a tool our artists can use to achieve the same result that I want, with a little less time … I’d put my bet on that. We want to use AI to relieve our artists of laborious work, of retakes—in other words as a starting point that allows them to continue to use their talent. There’s strong demand for our teams to go part-time. We’re very close with our teams so we do everything possible to try and listen to them … Part-time is challenging to implement, but I’ve always said that if AI would enable our artists to work part-time and achieve the same result, then hey let’s go for it. That could be a solution.”

“That being said,” he continues, “the reality is that today we’re doing our Season 2 post mortems. AI notwithstanding, the intent of the post mortems is to identify the parts of the production pipeline and the workflows that can be improved.”

They ask the teams questions like “What’s taking time?” and “What are the boring repetitive things you do?” These are the questions, Pascal says, that all studios should be constantly asking. Not “How can we fatten up our profit margins by using fewer people” or “How much AI can we get away with before audiences start leaving us.”

When the founding team was building up Fortiche, their first thought was that everyone else was doing the same thing—3D animation, as photorealistic as possible. They wanted to do something different, something more artistic. They built their art using a mix of 2D and 3D animation (and this was before the Spider-verse). In other words, Fortiche found their main jam by doing the human thing—working together, ideating, bringing together everyone’s experience, ideas, and creativity. They’re sure as hell not going to let generative AI drag them down the slippery slope of median pixel. You still need to be a real artist no matter what tool you’re using.

As their Italian neighbors would say, forza Fortiche!

Your turn… have you seen Arcane? What do you think?? By all means disagree fiercely as appropriate. Of course, I might need to put on my hextech body armor to respond… all’s fair in love and war 😜

1. According to a blog post on their site, the company’s previous name was “BooxAI.” So much there to unbox… or should I say, unboox. Happy belated Halloween everyone.