Inventor of the Square Pixel Apologizes, Creates Newer, Better Version
"The logical thing was not the only possibility," he told Wired in a recent interview. The technology of the time limited him to 176-by-176 pixels, but those pixels could have been any shape he desired. But he went with square; a decision he has since described as "something very foolish that everyone in the world has been suffering from ever since."
Kirsch is trying to make amends, though. Rather than silently sulk about the world of jagged lines to which he doomed us all back in 1957, the digital imaging pioneer is working on a new technique for rendering images that uses several differently shaped pixels to create smoother edges in images. Essentially, the new technology applies masks to pixels, and divides them into six sub-pixels, splitting them into rough triangles or rectangles based on the needs of the image.
Researchers already perform similar smoothing and compression techniques on images, but they're based on much more complex mathematical formulas. (Scientists rely on wavelet theory to accomplish the task.) Kirsch's simpler, mask-based treatment may or may not have a future in digital imaging, but it's good to know the man who arbitrarily stuck us with the square pixel isn't content to simply watch the world suffer any longer. [From: Wired]