Databending vs Glitch

データベンディング / グリッチ

Databending comes from Digital Image Techniques and Glitch from UI Expression. One is technique and the other style. They are reached for with some of the same intent, which is where they get confused.

Databending

Databending — AI interpretation
AI interpretation, not a reference

Databending edits image or sound files with tools not intended for their format, turning a decoder's misreading into the work; file structure, compression and edit position determine its bands and color shifts.

Glitch

Refuses to hide error and corruption, exposing the fragility of the digital medium itself.

DatabendingGlitch
Era2000s–digital
FamilyDigital Image TechniquesUI Expression
KindTechniqueStyle
CuesSections of an image displaced into horizontal bands / Separated color channels and unexpected saturation / Local compression blocks that leave the source visible / Row-like fractures repeating from the edit pointChannel shift / Block noise / Scanlines / Dropout
Best used forShowing the fragility and format-dependence of digital files through the damaged image · Colliding source imagery with decoder error in video, covers and postersExpressing uncertainty and system failure · A jolt of tension in music and video
TypeAdd copy after bending, or displace only part of a duplicate headline along the same axis.Partially corrupt a monospace or plain sans
CompositionPreserve one identifiable point and use the fracture direction as a compositional axis.Build the healthy composition first, then break part of it
MaterialDuplicate the file, protect its header and edit only payload data through another format's editor.RGB shift, compression noise, missing pieces
CautionA glitch overlay is not databending. The way one format was deliberately misread must determine the result.Perpetual glitch loses meaning. Stage the difference between normal and broken.

IndexStyle