Reflecting on Marta Blicharz’s Exhibition – Self-Referenced Abstract Art

By chance, I come across Marta Blicharz‘s very cool & interesting MFA exhibition “Breaking The Spell – Exhibition“. Check out the Exhibition if you haven’t seen it yet.

I got more curious about abstract art especially after my 2008 interview with abstract artist Christine Cheung. I try to keep an eye of whats new and exciting. I am happy to say I found something that interested me in Breaking The Spell. Here is how Marta describe an interesting abstract image “A Sultry Blonde in a Red Dress Touching Her Lips Seductively” (emphasis and link added),

“One of the exhibited images included a display of the ASCII symbols from a single JPEG file, 4 by 6 inches in size, and 300dpi in resolution. Of course that JPEG image was not shown, but in it’s place I have put up 5.5 panels, each 4 by 6 feet in size, with a 9.5 point Monaco  typeface. I had to break down these panels into 11 by 17 inch pieces to print them affordably and store them easily. It was titled: ‘A Sultry Blonde in a Red Dress Touching Her Lips Seductively.’”

Sultry Blonde got me into a bit of Matrix-ish thinking. I want to try to extend Marta‘s idea a few steps further and define the new idea (to me anyway) of a “Self-Referenced Abstract Art”.

Here is my preliminary idea of how to create a piece of Self-Referenced Abstract Art.

Step 1) Take a JPEG image “J” (e.g. a 5MP, 4×6 jpeg) and convert it into an ASCII text file “aJ” using a method similar but slightly different from Marta‘s Sultry Blonde. (note: for example, I need a monospaced font that I can map the extended ASCII set of characters to it.)

Step 2) Calculate the total number of ASCII characters, say “N”.

Step 3) Based on the total number of ASCII characters “N”, and the aspect ratio (e.g. 4×6), then calculate the number of ASCII characters per line (e.g “n”) and the total number of lines (e.g. “m”) to try to shape the ASCII characters into a rectangle/box/picture of the similar aspect ratio (in this case, 4×6). The last line may not have n characters but thats ok. (note: Again, I need a monospaced font in order to shape the resulted picture without too much distortion).

Step 4) Lay out each of the ASCII character in the ASCII text file “aJ” from left to right (“n” characters each line), top to bottom (for “m” lines).

Step 5) Treat the resulted ASCII Text grid (an “n” characters x “m” characters grid) as a virtual/abstractcanvas“. (Note: I think I am indirectly influenced by Chuck Close‘s method here.)

Step 6)Highlight/Colour/Paint” each ASCII character using the idea of mosaic (using an automated free or commercial program) to create a new “coloured ASCII picture” “caJ” that closely resembles the original JPEG picture “J” itself. Thus completing the self referencing process.

Note 1) The following are two different ways to “Highlight/Colour/Paint” the ASCII characters.

Note 1a) One method is painting the font colour of individual characters,


Note 1b) Another method is painting the “whole rectangle” the individual characters are in,


By the way, I don’t know if it will look good, but the above two methods can even be combined, and each “character” can have two colours, one for the box, one for the letter. But then it is getting a bit too complex.

Observation 1) The higher resolution (big file size) the original JPEG image “J” is, the closer the resulted “coloured ASCII picture” “caJ” will, obviously, more resemble the original.

Observation 2) In terms of personal taste, I like the “Highlight/Colour/Paint” method described in Note 1a better. It seems “cleaner”. The thing with the Note 1b method, it looks too much like Pointillism.

Concluding thoughts

I wish I can paint but I can’t. So while I can’t honestly calling myself a “painter” but may be I can call myself a “theoretical painter“, kinda like the difference between physics vs theoretical physic.

Comments are closed.

%d bloggers like this: