The Cousins of ASCII: A Field Guide to Art Made of Text

Jul 24, 26

Ask what “ASCII art” is and the honest answer is: one member of a much larger family. The instinct behind it - suggest a picture using nothing but a grid of characters and a limited alphabet of marks - has been reinvented on typewriters, teletypes, home computers, Japanese forums, and printed poetry. Each reinvention kept the core idea and changed one variable: the character set, the color, the font, or the century.

This is a field guide to that family. For each cousin you get the one-line definition, what makes it different, a small example where the medium allows, and whether it already lives in this gallery. If you want the deep history instead - Bell Labs, the BBS artscene, the demoscene - that is a separate post. This one is a map of the living relatives.

There are roughly a dozen close ones. I have grouped them into three families: the character-set cousins (same idea, different glyphs or hardware), the ASCII dialects (built directly on the ASCII grid), and the literary cousins (the pre-digital ancestors that had the idea first).

Family one: the character-set cousins

These share ASCII art’s DNA exactly - a fixed grid, a fixed alphabet - but reach for a richer alphabet or add color.

ANSI art

Plain ASCII is monochrome text. ANSI art adds two things: sixteen colors (via ANSI escape codes) and the block and shading glyphs of IBM’s code page 437. Those four shades were essentially low-resolution pixels, so artists could build smooth gradients and shaded portraits that plain text never could. This was the dominant computer folk art of the BBS era, and it still carries the .ANS file extension and the .NFO release-note tradition.

 full: █    dark: ▓    medium: ▒    light: ░    halves: ▀ ▄ ▌ ▐

Block and half-block art

The colorless sibling of ANSI. Using only the Unicode block elements above - especially the half-blocks ▀ and ▄ - you can address two “pixels” per character cell, doubling vertical resolution. Modern converters like chafa lean on this trick to turn photographs into shockingly detailed text.

Braille art

The highest-resolution text art there is. A single Unicode Braille cell (U+2800 and up) encodes a 2x4 grid of dots, so one character holds eight independent pixels. Pack a bitmap into Braille and you get startling detail out of pure text:

⣿⣿⣷⣦⡀   ⢀⣴⣾⣿⣿
⠈⠻⣿⣿⣷⣄⣴⣿⣿⠟⠁

This gallery has one: Three Candles Rising, three candles that climb in height from left to right, the first Braille piece on the wall.

PETSCII and ATASCII

Home computers shipped their own extended sets tuned for graphics. The Commodore 64’s PETSCII and the Atari’s ATASCII each added dedicated line-drawing and block characters, so a kid in 1983 could compose a detailed screen with no bitmap at all. Both have devoted revival scenes today.

Teletext and videotex

The broadcast cousin. Teletext (the pages hidden behind your old TV) and systems like France’s Minitel used chunky mosaic block graphics - a 2x3 sub-grid per cell - to draw weather maps and menus. Same idea, delivered over the airwaves.

Shift-JIS art

The most visually ambitious cousin, and the one that breaks ASCII’s cardinal rule. On Japan’s 2channel forums, artists composed with the proportional Shift-JIS character set - meaning characters are not all the same width. That sounds like a bug, but skilled artists exploited the varying widths to draw smooth curves impossible in a monospace grid, producing beloved mascots like Mona and enormous, expressive figures. Its smallest form is the kaomoji, a face you read without tilting your head:

(^_^)     (T_T)     \(^o^)/     ¯\_(ツ)_/¯

Family two: the ASCII dialects

These are built directly on the plain ASCII grid - no new glyphs required - but they are distinct enough crafts to have their own names and tools.

Turning a single word into a large heading assembled from smaller characters. FIGlet (1986) is the classic tool and is still typed into terminals every day:

 __   __
 \ \ / /__ ___
  \ V / -_|_-<
   |_|\___/__/

Box-drawing and diagram art

The practical, everyday cousin: flowcharts, tables, and UI mockups drawn with line characters (│ ─ ┌ ┐ └ ┘ ├ ┤). Every architecture diagram in a README is a small piece of this tradition.

┌───────────┐      ┌───────────┐
│  editor   │─────▶│  gallery  │
└───────────┘      └───────────┘

Halftone, or image-to-ASCII

The oldest computer technique in the family, descended directly from Bell Labs in the 1960s: sample the brightness of each cell in a photo and map it onto a ramp of characters ordered from empty to solid.

 .:-=+*#%@

It is deterministic and, for photographic subjects, unbeatable. It is exactly what this project’s image import does in your browser.

Emoticons, kaomoji, and emoji mosaics

The smallest text art has had the biggest reach. Scott Fahlman’s :-) in 1982 was the first emoticon; kaomoji made faces that need no rotation; and emoji mosaics tile colored glyphs into larger pictures. The shrug survives in group chats worldwide, and almost nobody typing it knows it is descended from forum culture.

Family three: the literary cousins

These predate the computer entirely. They are proof the idea was never really about ASCII - it was about arrangement.

Calligram and concrete poetry

Words arranged so their shape is the thing they describe: a poem about rain set falling down the page, or a mind that branches into a tree. Guillaume Apollinaire named the form in his 1918 Calligrammes, but pattern poems go back to Greek technopaignia around 300 BCE.

This gallery leans on the calligram hard. The Thinking Tree grows the words dreaming, musing, brewing, becoming, growing into an actual tree - a mind, left to grow, becoming a forest. A Good Omen and The Crooked Candle pour a poem into the silhouette of a candle. These are made with a tool that fills a shape with flowing text, which is the calligram translated into code.

Typewriter art

The direct mechanical ancestor. Before the computer, people made pictures on typewriters - the earliest known piece is a butterfly composed by the stenographer Flora Stacey in 1893 - using overstriking (printing several marks in one position) to build up shade. Change “typewriter” to “line printer” and you have the first computer ASCII art, unchanged in spirit.

Which cousins live here

If you are browsing the wall, here is the current lay of the land:

And the notable gaps - cousins the wall has never seen:

Those three are the obvious next frontiers if you want to stretch what “ASCII art” means on this site.

The thread that ties them together

Strip away color, glyph set, and font and every cousin here reduces to the same two ingredients: a grid, and a small alphabet of marks. That constraint is old enough to predate electricity and flexible enough to survive every display technology we have invented since. It asks the maker to suggest a great deal with very little - which is a fair description of art in general.

Want to add to the family? Make one in the editor, or import a photo and let the density ramp do the Bell Labs trick for you. Drop the result on the wall and the collection grows by one.