Every Frame A Painting

Postmortem: Every Frame a Painting – Tony Zhou – Medium 03.12.
17 19:10
Tony Zhou Follow

Creator/Editor/Narrator: Every Frame a Painting
Dec 2 · 16 min read
Postmortem: Every Frame a Painting

Every Frame a Painting is o1cially dead. Nothing sinister; we just
decided to end it, rather than keep on making stu?.
The existing videos will, of course, remain online. But there won’t be any
new ones.
The following is the script for what was supposed to be the Fnal episode,
voiced by both Taylor and myself. We were never able to make it. But we
think it may be useful to some of you making your own work on the
Internet, so we’re publishing it here.
. . .
OPENING WORDS
(TONY) Hi, my name is Tony… (TAYLOR) and my name is Taylor… and
this is the postmortem for Every Frame a Painting.
(TONY) As many of you have guessed, the channel more or less ended in
September 2016 with the release of the “Marvel Symphonic Universe”
video. For the last year, Taylor and I have tinkered behind-the-scenes to
see if there was anything else we wanted to do with this YouTube
channel.
https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 1 von 25

Postmortem: Every Frame a Painting – Tony Zhou – Medium 03.12.17 19:10
The last video we ever made for the channel. At the time, we didn’t know it would be the last.
(TAYLOR) But in the past year, we’ve both started new jobs and taken on
other freelance work. Things started piling up and it took all our energy
to get through the work we’d agreed to do.
When we started this YouTube project, we gave ourselves one simple

rule: if we ever stopped enjoying the videos, we’d also stop making them.
And one day, we woke up and felt it was time.
. . .
A CORRECTION, FOR THE RECORD

(TONY) “Every Frame a Painting” was not the product of a single person,
but of two people working, thinking, and disagreeing with each other.
This crediting issue was my fault. I Fnished the very Frst video with the
words “Edited & Narrated by Tony Zhou” and ever since then, it has been
hard to get people to notice that it now says “Written & Edited by
Taylor Ramos and Tony Zhou.”
So just to make it clear: these videos were made by the both of us.

Pictured: The both of us
. . .
PART I: ORIGINS
(TONY) Every Frame a Painting started because of two things. The Frst
happened in March of 2013. The second happened in April of 2014.
The Frst reason was that Taylor and I were dealing with the same
problem: at work, we had to discuss visual ideas with non-visual
people. I’m an editor, she’s an animator, but the problems we
encountered were largely the same.

If you’ve ever tried to talk about visual ideas with non-visual people, you
know how di1cult it is. Many people can’t understand something until
you show it AND explain it to them. Eventually I would just go to a
computer, pull up a YouTube clip, sit next to them and point out speciFc
things (this is how the Spielberg oner video started).
Video based on a real discussion I had with a director in 2013, a year before the video
One day I thought: “I wish someone would make video essays this way —
like someone sitting next to you, demonstrating visual literacy.” And that
was how the idea of Every Frame a Painting was born.
But once I had the idea, I didn’t make it. All I did was talk about it for a
year.
Instead, I would go to work, be frustrated by my boss, then come home

angry. By this time, Taylor and I had moved to San Francisco, and while I
went into the o1ce every day, she worked from home. Some days, I was
the only human contact she had. And for weeks and weeks and weeks,
all she heard was my anger.
I came in one Tuesday and started spouting o? about my boss. She

turned around and cut me o?. She said that I should stop complaining
about other people, and instead spend my energy on more productive
things. Maybe I should make those videos I keep talking about, instead
of bitching about work, or else I’d wake up in ten years and realize I’d
gotten nowhere. Then she closed the door.

Artist’s storyboard of himself being told to stop blaming others for his problems
Everything she said was true. Harsh, but true.
So naturally, I got mad at her. I stomped into the living room, turned on
my computer, and — partly out of inspiration, partly out of spite — I
started typing the Frst words: “Hi my name is Tony and this is Every
Frame a Painting.”
And that was the second reason the channel began.
. . .
PART II: PROBLEMS AND SOLUTIONS

(TAYLOR) It’s tempting to look back on decisions we made three years
ago and say that we knew which choices would eventually become
important. But the truth is we made all those initial decisions based on
instinct, and they ended up just working out.
1 (TAYLOR)
Choosing form over content

The Frst choice Tony made was to focus on Flm form, rather than
content. At the time, most YouTube videos seemed to focus on story and
character, so we went in the opposite direction.
This made the videos really straightforward and more of a learning tool
than anything else. Instead of showing a clip and talking about the plot,
we were showing the clip and talking about the clip. A person could
watch almost anything we made without having seen the Flms.
In fact, the videos often played better if you didn’t know what the Flms
were. That way, all you could do was react to the interplay of images and
sound.
The Satoshi Kon video, which (hopefully) works whether you’ve seen his work or not
2 (TAYLOR)
Making a cohesive channel
The second choice was mine. After watching the Frst video, I argued that
Tony should make a channel instead of individual videos.
Making it a channel meant creating a uniform tone and style, something

that Tony initially resisted. But I argued that a person who watched one
video would be more likely to come back and watch another if there was
a uniform style — even if they weren’t interested in that speciFc topic.
This meant we could get away with talking about less-known subjects
and plenty of people would still watch because the format was the same.
To his credit, Tony now admits I was right. (Tony’s note: “I refuse to
admit this”).

Though if the channel got one more person to watch Lynne Ramsay Slms, then we’re very glad
3 (TONY)
Creating a style from our biggest restriction
In order to make video essays on the Internet, we had to learn the basics
of copyright law. In America, there’s a provision called fair use; if you
meet four criteria, you can argue in court that you made reasonable use
of copyrighted material.
But as always, there’s a di?erence between what the law says and how
the law is implemented. You could make a video that meets the criteria
for fair use, but YouTube could still take it down because of their internal
system (Copyright ID) which analyzes and detects copyrighted material.
So I learned to edit my way around that system.
Nearly every stylistic decision you see about the channel — the length of
the clips, the number of examples, which studios’ Flms we chose, the
way narration and clip audio weave together, the reordering and cipping
of shots, the remixing of 5.1 audio, the rhythm and pacing of the overall
video — all of that was reverse-engineered from YouTube’s Copyright ID.

Style also reverse-engineered from watching this Slm.
I spent about a week doing brute force trial-and-error. I would privately

upload several di?erent essay clips, then see which got cagged and
which didn’t. This gave me a rough idea what the system could detect,
and I edited the videos to avoid those potholes.
So something that was designed to restrict us ended up becoming our

style.
4 (TONY)
The limits of our choices
And yet there were major problems with all of these decisions. We
wouldn’t realize it until years later. But by creating such a simple,
approachable style that skirted the edge of legality, we pretty much cut
ourselves o? from our most ambitious topics.
For instance, we’d always wanted to talk about Tarkovsky, but it’s
impossible to talk about how he moved the camera without talking
about why he moved the camera — and that meant playing very long
shots (impossible due to Copyright ID) and discussing religion.
Similarly, we could never Fgure out how to tackle Agnes Varda, because
our favorite idea required shooting new footage that didn’t mesh well
with what existed.

Though we did eventually do a video about the music in “Cléo from 5 to 7” for FilmStruck
The script you’re reading right now is the ultimate example of our
failure. We could not Fgure out how to turn this script into a video essay.
Half of the things in this script required Flming new footage, or using
still images or extensive re-writes to Ft other people’s footage. In other
words, this script shows the limits of the style we forged three years ago
— and frankly, it was pretty constricting.
. . .
PART III: GOOD HABITS

Here are some habits we picked up while making essays. Take whichever
you like, ignore the rest.
5 (TAYLOR)
Keep a notebook
Whenever you have an idea, jot it down (along with the date), then
forget about it.
The most important part of the process is to forget. Every idea seems
amazing at the moment of inception, but once you sleep on it and check
the notebook weeks later, you’ll Fnd that your brain has already
forgotten the weak ideas, but still thinks about the promising ones.

Notes and thumbnails for “Memories of Murder — Ensemble Staging” from 6 months before
the video
Every idea we’ve ever had for a video started in one of these notebooks,
and many of them gestated for months. A few of them were written
down in 2011, three years before the channel even began.
The end result is that you (the audience) are only seeing the stu? that
passed through our Frst Flter: are we still thinking about this idea long
after we Frst came up with it?
6 (TONY)
Research oEine
A huge percentage of the Internet is the same information, repeated over

and over again. This is especially apparent on Flm websites; they call it
aggregation but it’s really just a nicer way to say regurgitation.
So go to the library. Read books. We cannot emphasize this enough: read

books, read books, read books. Your work is only as good as your
research, and the best research tool we have is the public library (or a
Flm archive/research library if you’re lucky).

Research for Buster Keaton video — half purchased, half borrowed from San Francisco Public Library
It’s very tempting to use Google because it’s so quick and it’s right there,
but that’s exactly why you shouldn’t go straight to it. By taking your
research to the library, you’re immediately breaking out of the online
cycle of repetition, and your work will improve immediately.
7 (TONY)
Test ideas out loud
We like to turn each essay’s premise into a very simple statement or

question: Can you hum any musical theme from a Marvel Flm? How do
you show a text message in a movie?

Sometimes we keep that question as the opening line of the essay
Then we take that simple question and we test it out on real people —
usually our friends (who are also Flmmakers), usually over drinks or
dinner. The key is to ask the question casually, as if we aren’t planning to
make an essay.
The goal is to see if people react to an idea without us trying to sell it.
Because if the basic idea is already interesting to people, it’ll only get
better once we sharpen and hone it into a proper argument. But if
nobody bites at the early stage, that usually means the idea needs more
time to develop, or maybe we’re asking the wrong question.
8 (TAYLOR)
Focus on the fundamentals
Jerry Seinfeld has a quote we really like. He was talking about joke
writing and he said:
“…you can put all kinds of furniture in, but you gotta
have steel in the walls.”
He meant that a comic can dress up their act with whatever gimmick
they want, but underneath it all, there still has to be a joke.

We feel the same way about video essays. You can put whatever window
dressing you want: you can play a creepy serial killer, you can have nice
motion graphics, you can even throw Nujabes on the soundtrack. But
you’re still making an essay, which means it’s gotta have an argument.
Tony and I do something quite old-fashioned. We write all of our points

down on cash cards and then put them on a table. Then we rewrite, re-
arrange, and try them out in a number of di?erent combinations until
we Fgure out the tightest, strongest through-line for the argument.
Early ]ash cards for this script you’re reading
Once we Fnd this spine, Tony has to memorize all the cash cards in
order. He has to be able to say the entire argument, beginning to end,
without stopping. If he screws up, I make him go back to the beginning
and do it all over again.
Flash cards may or may not work for you, but we encourage you to take
Seinfeld’s advice: put steel in the walls.
9 (TONY)
Keep your shit organized
Many people falsely believe that I have some sort of crazy memory for
clips and movies. Not really. I just keep everything organized. I’m an

editor; this was beaten into me.
For “Vancouver Never Plays Itself” we used footage from 85 di?erent

Flms, but there were actually over 200 Flms in the project Fle. Keeping
everything organized and annotated was the only way I could Fnd
anything.
Every Frame a Painting was edited entirely in Final Cut Pro X for one
reason: keywords.
The Frst time I watch something, I watch it with a notebook. The second
time I watch it, I use FCPX and keyword anything that interests me.
Keywords for Chuck Jones video, all along the left column
Keywords group everything in a really simple, visual way. This is how I

Fgured out to cut from West Side Story to Transformers. From Godzilla
to I, Robot. From Jackie Chan to Marvel Flms. On my screen, all of these
clips are side-by-side because they share the same keyword.

This cut was discovered because of keywording
Organization is not just some anal-retentive habit; it is literally the best

way to make connections that would not happen otherwise.
. . .
PART IV: UNACKNOWLEDGED TRUTHS

We feel there are a lot of misconceptions or ideas that people believe
about making stu? on the Internet. Here’s the truth, as we see it.
10 (TAYLOR)
Work with a partner
There’s a common myth in the arts — the lone genius, usually a man,

creating everything by himself. For the most part, neither of us has
found it to be completely true. Look up most cases of a lone genius, and
you’ll Fnd a footnote about some unacknowledged helper.
Here’s how we work: Tony usually researches, writes and edits alone.
But I do everything else: I edit every draft, watch every version, watch
all the clips, do the cash cards, and build the thesis. I am the Frst and
last audience that sees everything before it goes out. And the closest
description we’ve ever come up with is that he is the editor, and I am the
editor’s editor.
If you look at the picture below, you’ll see Tony edited version 1 of the
Chuck Jones video by himself. Then we worked together for 7 days to
create version 7 (the Fnal). The yellow boxes are the only parts that
stayed the same, and even those sections got moved around.

A fair example of Tony’s Srst draft of the edit vs our mutual Snal draft
But most of all throughout this process, I’m a sounding board. Tony and I
often build a thesis by arguing the points with each other. Except for a
handful of videos, that has basically been our process for three years.
We’re not saying that this system will work for everyone, but having two
sets of eyes has worked really well for us.
11 (TONY)
There is no such thing as free content on the Internet
Everything costs something to make. If a person is putting out content

for free, that means they’re not getting paid for their time.
Video essays cost money to make. There’s the cost of the research, the
writing, the assembling of materials, the editing. It adds up to hours and
hours of work for something that takes minutes to consume. My average
just for editing (not anything else) is about 8 hours of editing for every 1
minute of video essay. So the 9-minute Jackie Chan video was around 72
hours of editing. It was probably at least double for research and writing.

Probably around 200 hours of work. Still a favorite, though.
With Patreon donations, Every Frame a Painting was eventually able to

break even. But from April 2014 until December 2015, we were making
video essays at a loss. We were never in danger of not making the rent,
but it got pretty Fnancially stressful on several occasions.
12 (TONY)
Nobody can cheat the triangle
Everyone who works in Flmmaking knows the triangle: Faster, Cheaper,

Better. Pick two. A Flm can be made fast and cheap, but it won’t be good.
Or you can make it fast and good, but it won’t be cheap. Or it can be
cheap and good, but it won’t happen fast.

The famous triangle/Venn diagram
Every Frame a Painting was made after we came home from our day jobs
and paid our bills. That kept it cheap. We also tried really hard to make it
good. Which ultimately meant we had to sacriFce “fast.”
The big danger for future video essayists is that large websites have
started moving away from the written word and towards video, which is
completely unsustainable. Video is just too expensive and time-
consuming to make.

All of this is accurate
(TAYLOR) Unfortunately, no matter how hard you try, nobody can cheat
this triangle. And sooner or later, all of these large sites will bleed
money, at which point some executive will say “We need to make our
content both faster AND cheaper!”
This is why we encourage every person who wants to make something

on the Internet to understand the value of independence. This is not
about artistic integrity or even money. We kept Every Frame a Painting
independent because as long as we could control this triangle, we could
control the end result.
We didn’t care about cheap or fast, we cared about it being good. If we

found a company willing to pay for it to happen fast, we’d work fast (full
disclosure: we eventually found two, Criterion and FilmStruck). If
nobody was willing to pay, then we worked slow.
But if we sold the channel to another company, or partnered with some

network, then we would no longer control the triangle. And guess which
of these three things would get sacriFced Frst?
13 (TONY)
Know thyself and thy audience

Nothing really prepares you for the experience of suddenly having “an
audience” on the Internet. The experience is di?erent for every person,
but I suspect the two major feelings are the same: there’s that initial
dopamine rush of “Oh my God somebody likes me” followed by that
creeping fear of “I better not fuck this up.”
George Orwell has this great quote:
What people always demand of a popular novelist is

that he shall write the same book over and over
again, forgetting that a man who would write the
same book twice could not even write it once.
The Internet, and YouTube in particular, is a massive echo chamber of

people asking you to write the same book over and over again. For your
own sanity, you need to keep those voices at arm’s length. But because of
that combination of dopamine and fear, it seems as though a lot of
people end up leaning into what the audience wants.
We’re not asking you to ignore the audience completely; it’s more about
setting clear boundaries between you and them.
My belief is that if you give the audience exactly what they ask for every
time, they will probably enjoy it, but on some level they’ll lose respect for
you. Hell, you’ll lose respect for yourself.
This tweet, however, did make me very happy. Baby, I got a stew.

(TAYLOR) And I believe that there is a balance that can be found, but you
and the audience should be equally as passionate about the idea. The
key is nuance.
So with Every Frame a Painting, we made it clear that we would not

accept requests; we came up with all the ideas and set up boundaries.
Not everybody needs to set the same boundaries we did, but you need to
have something.
14 (TAYLOR)
Success can be scarier than failure
The idea of failure is always scary. Nobody wants to fail, especially not in
front of other people. But this script you’re reading is a failure.
The reason we‘re putting it out there isn’t as an act of bravery or

anything; we just wanted to be honest. Failure is a fact of life — certainly
a fact of any life in the arts. It’s the only way we’ve ever learned
anything.
(TONY) For us, Every Frame a Painting ended up being both a personal
and a professional success. But over time, I felt trapped by what we’d
created — and also trapped by that success.
Every time I mentioned some Flm, I’d hear, “are you gonna make a video
about it?” Every time I started writing something, even for my own
amusement, a voice in the back of my head would say “how do I make
this accessible to my audience?” I stopped experimenting in my editing,
mostly because it was too far outside the margins of what I was making
on YouTube.
It wasn’t even fun making jokes on Twitter anymore, because they got
taken at face value:

Reaction to a joke I made on Twitter

I’d been a working video editor since I was 19 years old. I’d spent three
years of my personal time editing video essays. And now I could barely
stand to look at my own work. Eventually, the solution became clear: go
do something else.
(TAYLOR) Whenever Tony got really down like this, I would remind him
of a clip that we both love, from the Studio Ghibli documentary “The
Kingdom of Dreams and Madness.” It’s a moment when Hayao Miyazaki
is trying to draw a speciFc airplane. And for some reason, he cannot do
it.
From “The Kingdom of Dreams and Madness” (2014)
For days and days, he keeps trying to draw this plane, but nothing meets
his satisfaction. Eventually, he realizes that he’s spent too much time on
it. So he hands the plane o? to another animator, and he moves on to
something else.
For me and for Tony, this clip is extremely reassuring. If Miyazaki — the

world’s greatest living animator — can admit defeat after trying his best,
then it’s okay for everyone else. If he can let go, then so can we.
. . .
FINAL WORDS

Every Frame a Painting is over, because that period of our lives is over.
We have no idea what’s next, but we can tell you that we’re enjoying the
time we spend at our jobs, and the time we spend on other side projects
that nobody knows about. It also doesn’t hurt that we’re now back home
in Vancouver.
Still playing American cities though, grrrrrr
Thank you all for watching and supporting us over the past three years.
We can never express what an amazing experience this has been and
how much this has meant to us. We hope that this script may help
someone somewhere. Just liked we hoped the videos would. Maybe we’ll
see you for the next project. But for now:
My name is Tony and my name is Taylor, and this concludes Every Frame
a Painting.


Every Frame A Painting

Uploaded by

Copyright:

Available Formats

Every Frame A Painting

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Every Frame A Painting

Uploaded by

Copyright:

Available Formats

Postmortem: Every Frame a Painting – Tony Zhou – Medium 03.12.

Tony Zhou Follow

Postmortem: Every Frame a Painting

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 1 von 25

When we started this YouTube project, we gave ourselves one simple

A CORRECTION, FOR THE RECORD

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 2 von 25

Pictured: The both of us

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 3 von 25

Instead, I would go to work, be frustrated by my boss, then come home

I came in one Tuesday and started spouting o? about my boss. She

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 4 von 25

Everything she said was true. Harsh, but true.

And that was the second reason the channel began.

PART II: PROBLEMS AND SOLUTIONS

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 5 von 25

Making it a channel meant creating a uniform tone and style, something

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 6 von 25

So I learned to edit my way around that system.

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 7 von 25

Style also reverse-engineered from watching this Slm.

I spent about a week doing brute force trial-and-error. I would privately

So something that was designed to restrict us ended up becoming our

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 8 von 25

PART III: GOOD HABITS

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 9 von 25

A huge percentage of the Internet is the same information, repeated over

So go to the library. Read books. We cannot emphasize this enough: read

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 10 von 25

We like to turn each essay’s premise into a very simple statement or

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 11 von 25

Sometimes we keep that question as the opening line of the essay

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 12 von 25

Tony and I do something quite old-fashioned. We write all of our points

Early ]ash cards for this script you’re reading

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 13 von 25

editor; this was beaten into me.

For “Vancouver Never Plays Itself” we used footage from 85 di?erent

Keywords group everything in a really simple, visual way. This is how I

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 14 von 25

This cut was discovered because of keywording

Organization is not just some anal-retentive habit; it is literally the best

PART IV: UNACKNOWLEDGED TRUTHS

There’s a common myth in the arts — the lone genius, usually a man,

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 15 von 25

Everything costs something to make. If a person is putting out content

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 16 von 25

Probably around 200 hours of work. Still a favorite, though.

With Patreon donations, Every Frame a Painting was eventually able to

Everyone who works in Flmmaking knows the triangle: Faster, Cheaper,

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 17 von 25

The famous triangle/Venn diagram

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 18 von 25

All of this is accurate

This is why we encourage every person who wants to make something

We didn’t care about cheap or fast, we cared about it being good. If we

But if we sold the channel to another company, or partnered with some

https://medium.com/@tonyszhou/postmortem-1b338537fabc Seite 19 von 25