A generative virtual blossom algorithmically constructed in real time from society's representations of flowers.
Data Flower (prototype I) explores the possibility of creating unpredictable and ephemeral synthetic flora within the deterministic constraints of the digital realm. The 3D structure of the artwork is produced by a set of VRML (Virtual Reality Modelling Language) files that define the core geometry of the artificial flowers. A series of algorithms instigates and directs an endless cycle of emergence, growth and decay of the virtual blossoms. Randomisation of certain parameters at the onset of every new cycle causes subtle mutations within the petal formations and ensures that each flower develops in a different manner.
Unlike conventional artificial life systems which are solely based upon unchanging internal factors, the artwork integrates an external, non-deterministic element directly into its creation process. The surface textures of the synthetic blossoms are programmatically constructed each day by an online Java application that parses the image repository Flickr and selects one hundred of the most recent photographs which have been uploaded with the tag 'flower'. The sampled pictures are then algorithmically prepared and stored as a temporary database that is linked to the artwork's VRML component. On each loop of the flowering cycle, a randomly selected image from the database is applied across the growing virtual geometry, thus completing that flower's ephemeral form.
As in real life, every virtual blossom the artwork generates is unique since its internal 'genetic' code exists in a perpetual state of flux and its external 'developmental' influence is derived from an ever-changing pool of user-generated media.
Drew Baker [ VRML programming ] . Erik Fleming [ Java programming ]
Oliver Gingrich [ curation ] . Hugh Denard [ discourse ]
Data Flower (prototype I) v1.0 was created in 2010 with generous support from King's Visualisation Lab, Centre for Computing in the Humanities, King's College London. Version 1.2 was produced in 2014 for the Living Data solo touring exhibition.
[ gallery ] High-specification Windows 7/8/10 computer system capable of real-time high definition 3D rendering (VRML); multi/single-channel high definition video system; 5.1/2.1 audio system; and HCI device for user interaction.
[ online ] Windows 7/8/10 computer system with Firefox or Internet Explorer; the Cortona3D Viewer plugin; and stereo audio. A high-specification CPU/GPU, colour display with ≥1024x768 resolution and high-speed Internet connection are recommended.
* The Cortona3D Viewer is not available for OS X and Linux. On these systems FreeWRL can be used to render most aspects of the work.
* In April 2015 Google deprecated support for NPAPI plug-ins (including Cortona3D Viewer). As such, VRML works can no longer be viewed using Chrome.