Category Archives: Data Bending

Playing data files as if they were audio files.

A Slippery Place

Leave a Reply

This piece was written for Lev Taylor.

When trying to pick a religious name, I raised the idea of “Glitch” with my rabbi, who was initially nonplussed. In Yiddish, it means a mistake or a slip up. It’s in English that it’s come to mean a voltage spike and to go on to name a genre of digital art.

I made this piece out of a photo of Rabbi Lev Taylor, which I found on the internet. I used a script to glitch hundreds of copies of the Jpeg and stitch them into a film. I also played the Jpeg as it it were an audio file. Then I converted it to a larger file format and stretched it with PaulStretch. I used SuperCollider’s PV_RandComb ugen to pick a few frequencies to play at a time. Due to the maths of the FFT, this tends toward high pitches. I stitched everything together and sent him this piece.

Honestly, I like the idea of going by Glitch so much, I’m thinking about what contexts I could use it in. Art, sure! Everyday life?

Play

Podcast: Play in new window | Download (Duration: 1:02 — 2.3MB)

Dick Pic Hymnen

[play] Dick Pic Hymnen (2016)

This piece was commissioned by lesbiancockslut.com [NSFW]

When commissioned, I appealed on twitter for people to send me dick pics. Specifically, encrypted dick picks, to highlight that various national governments are spying on our sexts. In return, I was sent two unecrpyted images and one encrypted one. The unlocked images were of a tin of spotted dick, and a rooster (known to the English as a ‘cock’). The encrypted image was of the sender’s ‘friend, Richard.’

I took these files and played them as if they were wav files. Compressed and encrypted files sound like white noise, so I took the two unencrypted images and converted them to raw image formats. I then converted those to raw audio and used them as samples.

The pitches and rhythms in the piece are taken from a hymn that was popular in American Catholic churches when I was a child.

The images in the video are (obviously) not the images in the audio, as those were of unknown copyright status. The video images come from picasaweb.google.com/lh/photo/hJOtwAhiy6I9de3uFZxJUw and ommons.wikimedia.org/wiki/File:Boxerbriefs.jpg

Play

Podcast: Play in new window | Download (Duration: 1:11 — 2.2MB)

Shorts #33: Birthday Music for Caroline (2014)

[play]Shorts #33: Birthday Music for Caroline (2014)

Commissioned in honour of Caroline’s birthday – “Happy Birthday, love from Lauren and Alistair”

This was a digital piece using a very large number of scripts. First I wrote a SuperCollider script to generate some stochastic sounds. Then I wrote a script to convert audio files into bmp graphics files. Then I wrote a script to convert those files to jpegs, glitch the jpegs and convert back to audio. Then I also did several revisions of drawing more and more on the images in GIMP to transform them. I ended up with several batch processing scripts to glitch audio using visual data processing. I’ve put some of them on github and the rest will go after I fix some bugs I found after finishing this piece.

After generating loads of material, I listened to it and assembled it as a collage in Ardour.

The images are also kind of interesting! 2.aiff.au.bmp.jpg.gl

Play

Podcast: Play in new window | Download (Duration: 1:00 — 1.4MB)

Virtual Memory (Excerpt)

[play] Virtual Memory (Excerpt)
2001

Composed in 2001 using Macintosh Virtual Memory with an AIFF header added, a Jomox AirBase drum machine and a Midiverb. This used to be a 40 minute composition, which I broke into two parts due to limitations on file size imposed by Mp3.com. I released this as an album in 2001 or 2002, but it failed to sell well. To try to get people to listen to it, I renamed the files to “unreleased britney spears,” and logged on to Napster, where I would hang around in the chat rooms encouraging people to download them. In those days, we called that sort of behavior “guerilla marketing.” Now, it’s more commonly referred to as “being an asshole.”

The first 8 minutes were always the best, so here they are.

Play

Podcast: Play in new window | Download (9.9MB)

Headerless Data No. 2

[play] Headerless Data No. 2
2001

This piece is an example of data bending: a technique where a normal computer file is played as if it were a sound file. The file in this case is an image. Data bending, to me, sounds very similar to some fo the sounds created by Xenakis’ Gendyn algorithms, which shows the similarities of mathematical modelling of different phenomenon, whether it be stochastic or generated by a photograph.

Play

Podcast: Play in new window | Download (8.3MB)