Protected to Death

The two numbers

Put these next to each other

In 2024 the global anime market hit a record ¥3.84 trillion, which is $25.1 billion, up 14.8% in a single year. That is not a soft recovery or a hopeful projection. That is the industry printing money at a rate it has never managed before in its entire history, while everyone involved continues to insist that times are hard. When you break the number open, the interesting part is not its size. It is where the money came from.

56% of it came from overseas. $14.1 billion. Foreign revenue grew 26% year on year while the domestic Japanese market grew 2.8%, which is close enough to flat that you could call it flat and nobody would argue with you. The overseas market has now been larger than the home market for three consecutive years, and the editor of the industry’s own annual report says plainly that the gap will only widen. So the growth story of Japanese animation, the entire growth story, is foreigners. It is us. It is people who found this stuff through a friend or a clip or an argument on the internet at two in the morning.

Now the other number, and I want you to hold both of them at once. In December 2021, Toei Animation filed copyright claims against more than 150 videos belonging to one YouTube channel. The channel was Totally Not Mark, run by an Irishman called Mark Fitzpatrick who had 700,000 subscribers and whose entire catalogue amounted to a multi-year love letter to Dragon Ball. Not uploads of episodes. Not clip farming for ad money. Analysis, the kind where a grown man pauses on a single cut of animation and spends four minutes explaining why its timing is remarkable and why the people who made it deserved better than to be forgotten.

Overseas fans are now the majority of this industry’s income. This industry’s response to an overseas fan explaining, at length and for free, why their animation is beautiful was to try to erase him from the internet. That is not a contradiction I am constructing for dramatic effect. That is the two facts, sitting next to each other, touching.

Who actually pays

The record year did not reach the people drawing it

Before I get to the copyright fight, something has to go on the table, because “record $25.1 billion year” means nothing until you know who is holding the money. It is not the animators. It has never been the animators, and the figures on this are neither vague nor contested nor a matter of interpretation. They are published, they are official, and they are worse than most fans realise.

For comparison, because comparison is the only thing that makes a number mean anything: the average animator in the United States earns somewhere between $65,000 and $75,000 a year. The average Chinese animator earns around ¥5.2 million, roughly 2.7 times what their Japanese counterpart makes. Japan invented this medium, exported it to the entire planet, and now pays the people who make it worse than the countries that learned it from them.

This belongs in a piece about copyright for one specific reason. When a studio talks about protecting its intellectual property, the implied moral argument is that it protects the workers, that takedowns exist to defend the people who poured themselves into those frames. That argument would carry real weight if a meaningful share of the record-breaking $25.1 billion were reaching the person who drew the cut being defended. It is not. The enforcement protects the committee’s asset rather than the artist’s livelihood, and anybody using animators as a human shield in this argument should be made to explain the ¥1.97 million first.

What we lost

The thing that actually breaks my heart

I am not angry about a business dispute. Business disputes are boring and I would not spend a weekend writing about one. I am angry about what disappeared while nobody was allowed to talk about it, and I am angry that the disappearance happened slowly enough that most people never noticed it happening at all.

Anime was the place you went for the questions nobody else would touch. That was the whole proposition, the thing that separated it from every other animated tradition on earth. Western animation had decided, more or less permanently, that drawings were for children. Japan never made that decision, so anime became the one place where you could take an adult problem, hand it to an artist, and let them build something enormous and strange around it without anyone asking whether it might upset a parent group in Ohio.

Akira, 1988, is a film about a government that built a weapon, buried the evidence under a stadium, then watched a damaged kid inherit that power and dissolve into something nobody could reason with or forgive or save. It is about institutional rot and adolescent rage, and about what happens when the people in charge quietly decide that certain lives are acceptable losses. It ends with a boy becoming a universe, because the film would genuinely rather show you something incomprehensible than lie to you about a clean resolution, and it trusts you to sit with that. It cost a fortune. It had no built-in audience. It was adapted from a manga that was still being written, by the man writing it, at a budget that was insane for the era, about a subject nobody had ever proven would sell a single ticket.

Fist of the North Star looked at a world after the bombs and asked what a strong man owes a weak one, then answered that question seriously, inside a genre where the honest answer is usually “nothing, and here is a fight scene instead.” Ninja Scroll is ninety minutes of gorgeous, filthy, entirely uncompromised craft that never once stops to check whether the audience is comfortable, and it is a better film for refusing to check. Perfect Blue took the machinery of idol fame and turned it into psychological horror in 1997, decades before anybody in the West was prepared to admit that the machinery deserved horror.

None of those get funded today. Not one. The talent did not leave and audiences did not get stupider. What changed is that anime used to be an industry where somebody could walk into a room, say “let us make the disturbing thing,” and a person with money could say “fine, do it.” What replaced that person is a production committee, and a production committee’s documented, stated, structural top priority is to avoid failure. Not to succeed. To avoid failing. Those are different objectives, and they produce entirely different art.

Around 90% of all anime produced now adapts an existing manga or light novel, because a manga’s sales history is a spreadsheet a committee can point at during a meeting, while an original idea is just some person’s conviction that a thing ought to exist. A veteran producer whose credits include Tokyo Godfathers and Serial Experiments Lain, a man who was in the room for two of the strangest things this medium ever produced, summarised the situation in four words. “Everything has become shallower.” That is not a fan complaining on a forum at midnight. That is one of the people who made the good stuff, watching what came after.

The empty chair

Satoshi Kon died and nobody took the chair

Satoshi Kon made Perfect Blue in 1997, and it is routinely described as the first animated film to carry a genuine psychological suspense story rather than borrowing the outward shape of one. He followed it with Millennium Actress, then Tokyo Godfathers, then Paranoia Agent for television, then Paprika in 2006, a film about entering other people’s dreams that half of Hollywood has been quietly borrowing from ever since. Four features and a series. Every one of them original. Not one adapted from a property with a proven sales chart.

He died on 24 August 2010 of pancreatic cancer, aged forty-six. He was working on a film called The Dream Machine. It has never been finished. It remains listed as indefinitely postponed, which is how this industry says something is not happening.

Here is the part that genuinely keeps me up. In the sixteen years since, nobody has taken that chair. Not because Japan ran out of talented people, which would be an absurd claim about the country that trains the most skilled animators alive. The pipeline that produced a Satoshi Kon simply does not exist anymore. There is no funding path for a director with a strange original idea and no sales chart to wave at a committee. The gap he left is not a gap in ability. It is a gap in permission, it has stayed open for sixteen years, and every year the industry announces another record and another thirty isekai.

Permission, not quality

Where is my Akira

I keep asking that question and I keep getting isekai, and I want to be careful here, because “anime is bad now” is a lazy thing to say and it is also flatly untrue. Extraordinary work is being made right now. There are shows from the last five years I would defend to anybody, made by people at the absolute peak of the craft, animated with a technical sophistication that would have been impossible in 1988. What vanished is not quality. It is permission.

Permission to make something that might not work. Permission to be ugly on purpose, to be slow, to end without resolving anything, to be so specific and strange that no committee on earth could forecast the merchandising. Permission to spend real money on an idea whose only supporting evidence is that the person pitching it will not shut up about it. That is the ingredient every one of those films had in common, and it is the single thing a record-breaking year cannot buy.

Anime was supposed to be the alternative reality where adults could go and work through the questions that real life gives you no room to ask out loud. What you owe people. Whether you are complicit in something. What happens when you are handed power you did not earn and are not ready for. Whether the person you are frightened of is actually yourself. That is what the medium was for, and it is why it mattered so intensely to the people it mattered to, who were frequently people with nowhere else to put those questions. It was not escapism. It was the opposite of escapism. For some of us it was the only place the hard thing got said out loud.

Now, increasingly, it is a delivery mechanism for competence porn about a guy with a stat sheet who is good at everything and never loses anything that matters to him.

The controlled experiment

Anime and BookTok ran the same test and chose opposite answers

Everything above reads as nostalgia unless you can demonstrate that the alternative actually works. So let me demonstrate it, because there is a controlled experiment sitting in the industry next door and almost nobody in anime appears to have looked at it properly.

Publishing had precisely the same disease. Old, conservative, structurally allergic to risk, absolutely convinced that young people had stopped reading, and run in large part by people who regarded the internet as a threat best handled by the legal department. Every complaint you could make about production committees in 2018, you could have made about publishing houses, and plenty of people did.

Then TikTok arrived and publishers made a decision that looks obvious now and absolutely was not at the time. They let people read the book out loud. Not a licensed excerpt. Not a pull quote vetted by marketing. People filmed themselves reading entire passages to camera, screaming about a chapter twist, holding the physical page up to the lens, spoiling endings, arguing viciously with strangers, crying about a paragraph at one in the morning with the lights off. Publishers did not send takedowns. Several of them started actively seeding it.

Print adult fiction sales rose 8.5% in 2022 and BookTok was identified as the single biggest driver of that gain. BookTok author sales then grew nearly 20% again in 2024, a fifth consecutive year of growth. They more than tripled their volume in two years by doing nothing at all except declining to stop strangers from showing the receipts.

The part that matters more than the sales figure

Look at which books won. Not the safe ones. Not the algorithmically optimised ones. Weird books. Enormously long books. Books about grief and obsession and relationships that make you deeply uncomfortable. Backlist titles that had been commercially dead for a decade came back from nowhere, because one person made a compelling forty-five second case to an audience of strangers and the strangers went and bought it.

That is the mechanism, and it is not complicated. When people are free to point at the specific thing that is good, the specific thing gets rewarded, and “specific” is another word for “not a clone.” An ecosystem full of people arguing loudly about what works is an ecosystem where being distinctive becomes the profitable strategy. Suppress the arguing and what remains is the spreadsheet, and the spreadsheet says with total confidence: make the isekai.

Now compare the two revenue models, because this is where it gets damning

When BookTok converts somebody, the conversion is the product. Person watches video, person buys book, publisher receives money for the book. The path from exposure to revenue is one step long, and crucially, the thing being sold is the thing being discussed. Every new person is a new book sale. The ceiling is the population of earth.

Anime does not work like that, and the industry’s own accounting proves it. Of that record $25.1 billion, the production market accounts for only ¥466.2 billion, roughly $3.06 billion. Somewhere near $22 billion of the total is downstream: merchandise, events, concerts, licensing, bundled rights deals. The AJA’s own methodology counts all of it, which is legitimate, and it also tells you exactly what kind of business this has become.

So the two industries had the same problem and picked opposite solutions. Publishing bet that letting people talk would grow the number of buyers. Anime bet that it could keep growing by selling more objects per existing buyer, and it protected that model by suppressing the discovery mechanism that would have grown the base instead. One of those strategies compounds forever. The other one runs into a shelf that is already full.

And notice how neatly the creative outcome follows the revenue model. If your money comes from selling objects to people who are already converted, the safest product is one whose characters are designed to be merchandised and whose premise is already proven. If your money comes from converting new people, the safest product is one that somebody feels compelled to tell a stranger about. Thirty-four isekai a year is not a mysterious failure of imagination. It is the correct output of the incentive structure the industry built for itself.

The business math

The only language a committee speaks

Let us do the rest of this in the language the responsible parties actually respond to, because moral arguments have not worked and arithmetic is worth a try. You are an anime rightsholder. 56% of your revenue is now foreign and compounding at 26% a year, while your domestic base grows at 2.8%. Every strategic document in your building says the same thing about where the future is.

Meanwhile an Irishman with 700,000 subscribers makes forty-minute videos explaining why your 1989 fight choreography is a masterpiece. He is running a permanent, free, extraordinarily effective advertising campaign aimed with sniper precision at exactly the demographic your entire growth curve depends on, and he is paying for the equipment, the editing time and the electricity out of his own pocket. What is the correct business response?

It is not a trick question, and every other entertainment sector answered it decades ago without needing to be asked twice. Film studios send press kits and screeners and fly critics to junkets. Games publishers ship review codes before launch, bundle capture tools into the product, and openly beg people to make content about it. Music labels put the official video on the exact platform where the fighting happens, because the fighting is the marketing. None of those industries collapsed. All of them worked out that somebody spending forty minutes explaining your work is not competing with you for an audience. They are assembling one on your behalf, for free, on their own time, with their own equipment.

Toei filed 150 claims. Later they copyright-struck Uncle Roger, a comedian, over One Piece. So this is not one overzealous week in a legal department. It is a policy, and the policy appears to be that the growth strategy for a company whose growth is entirely foreign should be fewer foreigners saying its name out loud.

All publicity is free publicity, including the bad kind, and that is the part that always gets missed by people who have never watched a fandom form. A furious video essay about how an adaptation butchered its source material still puts that source material in front of a hundred thousand people who had never heard of it. Somebody in those comments buys volume one purely out of spite, reads it, and is a genuine fan by volume four. That is not a theory I am floating. That is how the fandom you are currently in recruited you.

The law

They were already allowed to do this

Here is where I stop being measured, because the genuinely maddening thing is not that critics lack legal protection. They have it. It simply does not help them.

United States law, 17 U.S.C. §107, fair use. The very first words in the statute’s list of protected purposes are criticism and comment. Not buried in a subsection, not read into it by a court decades later. First. The people who drafted that provision sat down to decide what mattered most, and the first thing they wrote was that people must be able to criticise things. Four factors decide any given case: how transformative the use is, the nature of the original, how much was taken, and whether the market for the original was damaged. A review clipping thirty seconds to make an argument passes all four so comfortably that walking through it is almost boring. It is transformative because you added the entire analytical apparatus the original never contained. The amount is thirty seconds out of a twenty-four minute episode. The market harm is negative, because you are advertising the show to people who might go and pay for it.

That is not a clever loophole somebody discovered. It is the textbook central case the doctrine exists to protect, and it has been litigated and settled twice in ways that should have ended this argument years ago.

Now hold that ruling against what actually happened. Toei filed more than 150 claims against a channel whose entire reason for existing is criticism, the single most explicitly protected category in the American statute, then failed to provide justification for any of them when challenged. They did not engage with the dispute process at all. They manually claimed and blocked each video individually, one after another, 150 consecutive times, violating YouTube’s own copyright policy and voiding the entire batch on procedural grounds. I would sincerely love to see the good faith fair use analysis performed for video number ninety-four.

Fitzpatrick did not win on the merits. He got his life’s work back on a technicality, because the rightsholder could not be bothered to fill in a form correctly a hundred and fifty times in a row. Everybody treated that as vindication. It was luck, and it required a public outcry loud enough for a platform to notice, which is a resource almost nobody in his position possesses. The videos came back everywhere except Japan, because YouTube applied American law and reached one conclusion, then applied Japanese law and reached another. Same footage. Same man. The legality changed at a border.

Being fair about Japan

The clause nobody bothers to quote

The lazy version of this argument, which I have made myself and want to correct on the record, is that Japan has no fair use and therefore criticism is simply illegal there. That is not accurate, and I would rather be right than satisfying.

Japan does not have a general fair use doctrine. What it has is an enumerated list of specific exceptions, Articles 30 through 50 of the Copyright Act, an approach far closer to fair dealing than to the American model. A 2018 amendment, adjusted again in 2020, introduced more flexible provisions that balance factors in a way that rhymes with fair use, but the underlying architecture remains a closed list rather than an open principle.

Critique. Named explicitly, in the statute, in the same breath as news reporting and academic study. Japanese law does contemplate quoting a work in order to criticise it, and anybody telling you otherwise has not read the provision.

What Japanese law does not provide is the elastic, purpose-driven balancing that makes American fair use workable for modern media. The phrases “consistent with fair practices” and “within a scope that is justified” carry an enormous amount of unspecified weight, and Japanese practice has traditionally read quotation narrowly. The quoted material must be clearly subordinate to your own work, clearly delineated from it, and no more extensive than strictly necessary. That framework was built for a scholar quoting a paragraph in a journal article. It was never built for a video essay in which the footage and the argument are inseparable by design, because the argument is about the footage and cannot exist without showing it.

So both sides behave rationally inside their own rulebook. The American critic does something their statute names as its first protected purpose. The Japanese rightsholder enforces rights that, within their own legal tradition, are genuinely being infringed. Nobody in this story is a cartoon villain. The rulebooks simply do not reconcile, and the person standing in the gap between them gets flattened by both of them at once.

The machine

Content ID cannot read

That unresolvable legal gap would still only be a stalemate, and stalemates can be negotiated. What turns it into a rout is the enforcement layer sitting on top, which converts a genuine legal disagreement into a slot machine with a lever only one party can reach.

Content ID is pattern matching and nothing else. It finds footage. It cannot watch your video, observe that nine of its eleven minutes are storyboard analysis, and conclude that you are protected, because it has no concept of protection. It has a fingerprint and a match, and the match wins by default. The strike system does not adjudicate anything either. It processes claims, and the burden of contesting one falls entirely on the accused, every single time, by design.

People say “just file a counter-notification” as though that were a free action, which tells me they have never read the form. Filing a counter-notification requires you to consent to the jurisdiction of a federal court. That is not a metaphor or an alarmist reading. It is a line you sign. In order to assert a right you already legally possess, you must formally agree to be sued by a Japanese animation conglomerate employing more lawyers than you have subscribers. If three claims stick, the channel is deleted. Not the videos. The channel, the archive, the subscriber base, the decade of work, all of it, gone.

So the operative question was never “is what I am doing legal.” It is “can I afford to find out,” and for essentially everyone doing this work the answer is no. The law never ordered this silence. The cost structure produced it independently, and the cost structure does not answer to §107 or to anyone else.

The tangent

Doujinshi is fine but a microphone is theft

I am taking this tangent because it is the same hypocrisy wearing a different hat, and once you see it you cannot unsee it. Doujinshi. Fan comics. Frequently unlicensed, routinely featuring copyrighted characters, often explicit, sold for real money at Comiket inside a convention centre, in public, twice a year, at genuinely industrial scale, with the full knowledge of every publisher in the country.

The industry looks the other way deliberately. Not because it lacks the legal power to act, which it obviously possesses in abundance, but because publishers understand perfectly well that the doujinshi scene functions as a farm system. It builds fandom, it trains creators, it keeps properties alive between official releases, and a significant number of professional mangaka came up through it before going legitimate. The tolerance is decades old, entirely intentional, and load-bearing for the whole ecosystem.

So fans redrawing your copyrighted characters and selling the results at a table for cash is ecosystem, while a guy with a microphone explaining why your key animation is brilliant is theft. I have turned that over for a long time looking for the coherent principle that separates them, and the only explanation surviving contact with the evidence is depressingly mundane. Doujinshi happens in Japanese, at home, inside a culture with unwritten rules everybody already understands, where enforcement would be expensive and socially costly and bad for business. YouTube happens in English, in public, on a platform that helpfully installed a button.

It is not principle. It is friction. Where enforcement is cheap, it happens. Where it is expensive and awkward, we suddenly discover a great deal of nuance and cultural sensitivity. That is a bleak conclusion to reach about your own hobby and I have reached it anyway.

The quiet part

The blogs died too, and that mattered more than anyone noticed

There is a second collapse running underneath the copyright fight, and it gets almost no attention because nothing dramatic ever happened. Anime fandom used to run on blogs and forums. Long episodic write-ups, week by week, with comment sections where people argued for three hundred replies about whether a director had earned an ending. Those places are largely gone now.

The 2010s hollowed them out as social platforms centralised everything. Once Reddit replaced forums and Discord replaced IRC, a new generation of fans had no particular reason to go looking for independent sites, and independent sites stopped acquiring the new writers who would have kept them alive. Well-known anime blogs that once pulled serious traffic now sit at a fraction of it. AnimeSuki’s forums, one of the genuinely storied fan communities, recorded 44,830 visits in October 2025, down 55.56% from the previous month. That is not a slow decline. That is a floor giving way.

Stack that on top of everything else and the picture gets bleak fast. Reading is down generally. The independent written fandom that used to metabolise a series week by week has largely dispersed into platforms that do not archive, do not rank, and do not surface anything older than a few days. And the video format that replaced all of it, the one thing with genuine reach left, is precisely the format sitting under an automated takedown machine.

So the discovery surface for anime has been shrinking from both directions simultaneously. The written spaces emptied out on their own, and the video spaces are being actively policed. Every Content ID claim against a YouTuber removes another entrance to the building at the exact moment most of the other entrances have quietly been bricked up. A series does not need to be attacked to disappear. It just needs nobody to have any remaining way to hear about it.

What it costs

And it is not the YouTubers

The critics will be fine, incidentally, and that is worth saying plainly because it disposes of the least interesting objection. Fitzpatrick recovered. People pivot, cover something else, rebuild elsewhere. I am not writing thousands of words in defence of the career prospects of video essayists.

A series that cannot be discussed becomes a series that dies, and that is the actual cost, paid by the work itself and by everybody who would have loved it.

It does not die dramatically. There is no moment anybody could point at. It simply stops entering people’s field of vision. No video essays. No stills in a review that stop somebody mid-scroll on a train. No twenty-second clip circulating with four hundred quote-tweets arguing underneath it. No friend saying you have to watch episode fourteen, I am serious, just get to fourteen. The show does not get attacked or defended or wilfully misread or fought over by people who care far too much about it. It just is not there. Then one day it is a thing your friend’s older brother used to like, then it is a title on a streaming service nobody clicks, and then it is gone.

Meanwhile every Western property is being clipped and essayed and argued about across every platform simultaneously, because their rightsholders worked out two decades ago that the conversation is the marketing, and that suppressing it is functionally identical to buying no advertising while paying a legal team for the privilege of buying none.

Here is the unfashionable thing I will say plainly, because pretending otherwise helps nobody. We do not read anymore. Not really, not at scale, not the way we used to. That is not a moral failing and I refuse to perform disappointment about it. Attention moved to visual and short and clipped and shareable, and it is not coming back because I disapprove. A review with images gets read. Nine hundred words describing what a fight scene looked like gets closed in four seconds by somebody who fully intended to read it and simply did not.

So the studios guarding their frames most aggressively are, with total efficiency, administering the quietest possible death to their own catalogues. The protection works flawlessly. Nobody discusses the show. Nobody discovers the show. Objective achieved, sir. Please collect your record-breaking annual report on the way out.

The move

Own the site, rent the channel

I am not asking anybody to surrender their copyright, and I want to be precise about that, because studios reliably hear “abolish our rights” whenever somebody says “let people review things.” Keep the copyright. Enforce it against people uploading full episodes. Nobody sensible is defending that.

What I am asking for is the thing every other corner of entertainment already worked out and acts on daily. Criticism is not competition. BookTok proved it with numbers, thirteen million copies rising to forty-seven million in two years, achieved by doing nothing except declining to interfere. Ghibli proved a Japanese studio can act on the identical insight from inside the identical legal system without anything catching fire. The evidence is not thin and it is not theoretical. It is published, in both industries, waiting for somebody to read it.

Own the site. Rent the channel. The real analysis, with the images it needs in order to make sense, lives on ground you control where nothing is scanning. The video becomes a pointer toward it rather than the thing itself, which also means that when a platform changes its mind about you overnight, and platforms reliably do, you still have the work.

The one good sign

You cannot Content ID a conversation

I want to end somewhere other than despair, because something genuinely encouraging is happening and it has been happening quietly enough that the industry has not noticed it or found a way to interfere with it yet.

The old series are coming back by word of mouth. Not through a marketing campaign. Not through a remaster nobody asked for. Through people in their thirties and forties telling younger relatives about the thing that reorganised their head at fifteen. An uncle explaining what Akira did to him in 1990. An aunt describing Perfect Blue badly and incompletely and with total conviction over dinner. Somebody’s older cousin refusing to shut up about Cowboy Bebop until the kid finally watches it out of sheer exhaustion. I keep encountering versions of this, and every version arrives without a single frame of copyrighted footage attached.

That is the thing the machine cannot touch. Content ID scans uploads. It cannot scan a kitchen table. There is no fingerprint for a person insisting that something mattered, no automated claim that lands on an anecdote, no strike system for enthusiasm transmitted directly between two humans who are related to each other. You can Content ID a video. You cannot Content ID free speech.

It is slower than the alternative and vastly less efficient, and it should not have to be the fallback for an industry earning twenty-five billion dollars a year. But it is durable in a way platforms are not, and it explains why the titles enjoying a genuine resurgence right now tend to be the old, strange, uncompromising ones rather than anything from the safe middle of the current catalogue. Nobody drags their nephew across the room to tell him about the thirty-fourth isekai. People do that for the thing that frightened them.

Final take

Now go find me another Ninja Scroll

Thirty-four isekai last year. A record twenty-five billion dollars, of which barely three billion is production. Animators living below the poverty line of their own country while the industry they built posts the best numbers in its history and calls it a triumph. And somewhere in a drawer there is a storyboard for the next Akira that a committee killed because the sales history was unproven and nobody in the room could point at a spreadsheet justifying it.

There is nobody left with a microphone big enough to have argued for it either, and that is the part connecting all of this together. We spent a decade teaching the people who argue loudest and most lovingly about anime that their reward for doing it is a copyright strike and a deleted channel, then we act surprised when the only ideas surviving the funding process are the ones that already worked twice.

Satoshi Kon’s last film is still unfinished. Sixteen years. Nobody took the chair.

Twenty-five billion dollars. Record year. Congratulations, sincerely, to whoever is holding it.

Now go find me another Ninja Scroll.