The series Go Ogle by photographer Meggan Gould takes the first 100 responses to a Google image search, then overlays those 100 images into a single photographic “average.”
As Gould herself writes:
“The results, a visualization of intersections between Boolean logic and the popular imagination, are more often than not a hopeless jumble of unidentifiable pixels – but occasionally a recognizable form does emerge.” See animation.
“Word choice, spelling, and textual hints are all critical to conducting an effective search, and the averages reflect their importance: a search for coke+can reveals a crisp, almost legible average, whereas coca+cola+can is muddy and barely recognizable. Truly iconic imagery is elusive, particularly considering the glut of computer graphics through which internet spiders and archivers crawl daily; only a small fraction of searches retains any degree of legibility through the averaging process.”
But some of the most recognizable averages, I have found, are architectural.
Leaving me to wonder what has to be at least one other BLDGBLOG reader’s first thought: what about world+trade+center?
If BLDGBLOG was richer, in fact, it could probably keep Meggan Gould in business for several years, producing more and more – and more and more – architectural averages: stone+henge; san+andreas+fault; berlin+wall; yucca+mountain; space+shuttle; taj+mahal; falling+water…
PS: big+ben; forbidden+city; chrysler+building; trump+tower…
3 thoughts on “Architectural averages”
Check out the python script that does averages with Flickr photos…
check out Jason Salavon’s work:
Anonymous, check out this post about Salavon’s work. But thanks for the tip!