Art is good compression
It is a rare skill to make something compelling, without using the full dimensionality of reality
It is fitting that the writers of Silicon Valley chose insanely good compression to the be the backbone of their TV show.
It would be easy, especially if you are of the MBB-HBR-MBA bent, to assume that technology is simply a lever you can “turn up” - like money, or manpower. The compression algorithm is good? Excellent. Add moar money 💰 and moar engineer 🐵. Make it good-er, and we’ll all get richer.
Engineers know that the world doesn’t work this way. The first lesson of ECON101 is that there is no such thing as a free lunch: technology can’t simply be “dialed up” to infinity.
For example, compression is either lossy or lossless:
Lossy compression discards the least meaningful data: JPEG, or more accurately, JFIF
(Source: http://pi.math.cornell.edu/~web6140/TopTenAlgorithms/JPEG.html)
The downsampling algorithms used by JPEG take advantage of human biology:
Our eyes are more sensitive to illuminance (light bouncing off a surface) than chrominance (the colors of things). This is why black and white photographs still look meaningful to us.
Our eyes are not very sensitive to the high frequency components of an image, so those get removed after a Discrete Cosine Transform (read: represent an image as a bunch of added-up waves).
tl;dr: we lose information after this compression, but we remove the information that the human eye finds the least meaningful, so the image still looks fine to us - to a limit.¹
Lossless compression requires work from the consumer: Huffman encoding
(Source: https://www.programiz.com/dsa/huffman-coding)
The last step of JPEG compression (as well as .zip, .png, .mp3…) is Huffman encoding. Tldr: use the frequencies of letters in a string to construct a tree, and store the way that you’d have to walk the tree to get the string, instead of the string itself.
This is lossless (as in, you can always recover the original string). But this requires the use of a previously agreed-upon external construct: the idea of the tree, the structure of the edges, the fact that left is 0 and right is 1, the understanding that the nodes represent letters, etc.
Which leads me to…
Media, hot and cold
In his paradigm-defining 1964 classic, Understanding Media: The Extensions of Man, Marshall McLuhan defines two kinds of media - hot 🔥 and cold ❄️:
There is a basic principle that distinguishes a hot medium like radio from a cool one like the telephone, or a hot medium like the movie from a cool one like TV. A hot medium is one that extends one single sense in "high definition." High definition is the state of being well filled with data. A photograph is, visually, "high definition." A cartoon is "low definition," simply because very little visual information is provided. Telephone is a cool medium. or one of low definition, because the ear is given a meager amount of information. And speech is a cool medium of low definition, because so little is given and so much has to be filled in by the listener. On the other hand, ho, media do not leave so much to be filled in or completed by the audience. Hot media are, therefore, low in participation, and cool media are high in participation or completion by the audience.
And in that charming kind of 1960’s way that we can no longer get away with in this day and age without getting canceled, McLuhan also explains why sunglasses are compelling while normal glasses are not, sexy librarian tropes notwithstanding:
The principle that distinguishes hot and cold media is perfectly embodied in the folk wisdom: "Men seldom make passes at girls who wear glasses." Glasses intensify the outward-going vision, and fill in the feminine image exceedingly, Marion the Librarian notwithstanding. Dark glasses, on the other hand, create the inscrutable and inaccessible image that invites a great deal of participation and completion.
The reason why arthouse films are more compelling to certain people than the latest piece of Michael-Bay-box-office-visual-overkill is because, well, some of us actually like doing some mental work when we enjoy art (or experiences…or other people’s outfits). If it’s all laid out for us already, the experience can actually be an incredibly passive and uninteresting one - because it doesn’t actually contain us.
McLuhan knows this too, when he points out that the myth of Narcissus isn’t really about a man falling in love with himself, but about seeing himself extended into another medium (water):
Now the point of this myth is the fact that men at once become fascinated by any extension of themselves in any material other than themselves. There have been cynics who insisted that men fall deepest in love with women who give them back their own image.
We could replace “men” with something less gendered here, and replace “women” with any form of media - art, movies, music, professions - and the phrase would still stand.
Two ways to make compelling art
Taking lessons from JPEG, we conclude that there are two ways to make media (art, movies, outfits, buildings) compelling without using the full dimensionality of reality²:
1: Remove things that provide little value
“Before you leave the house, look in the mirror and take at least one thing off.”
~ widely attributed to Coco Chanel
Every building, every room, every garden is better, when all the patterns which it needs are compressed as far as it is possible for them to be. The building will be cheaper; and the meanings in it will be denser…You may think of this process of compressing patterns, as a way to make the cheapest possible building which has the necessary patterns in it. It is, also, the only way of using a pattern language to make buildings which are poems.
~ mathematician-qua-architect Christopher Alexander, A Pattern Language (1977)
2: Allow the consumer to do some work in reconstructing meaning
In other words, the best movies actually make you think, and the best outfits actually leave something to the imagination. 👀
Footnotes
1- Examine the following crops from Christina's World by Andrew Wyeth, 1948.
JPEG compression at 100% quality in Photoshop (868kb):
JPEG compression at 1% quality in Photoshop (46kb):
JPEG compression after upload to a potatoFacebook (108kb - how?! 🥔):
2- This blog post was inspired by a conversation with Christopher Masurek from the EF Berlin office, who finally shot down my insistence that VR can never be compelling because it will never be able to convey the full dimensionality of reality - by arguing that we still find movies compelling even though they are just sounds and projected light on a flat, 2D surface.
This also explains why gimmicks like 3D glasses and dumb movie theatre seats that shake and spray water at you have failed to provide any genuine elevation of the movie-going experience: they are adding dimensions that aren’t really meaningful.
Choosing the correct dimensions to include is a craft that movie directors and cinematographers literally spend a lifetime perfecting.
If you enjoyed this, find more writing at casey.li.