The first digital picture on a computer was, of course, a picture of a baby.

In 1957, Russell Kirsch was working at the National Bureau of Standards, which had something amazing—the first computer that could be programmed. Here's how Kirsch introduced it in a 2011 talk:

Please try to put your thinking back
to when there was a widespread lack
of any tool that you believed
could do whatever you conceived.

When I was young and not a fool
I did have use of such a tool
The tool I had I'll show you now
My best ideas it did allow.

And here's what the computer looked like:

NIST

The computer had plenty of important jobs to do, like, as Kirsch told the Oregonian later, "thermonuclear weapons calculations and things of this sort." But he was allowed to use it to solve other problems, too. And he wondered what it would mean to have a computer look at a picture.

He began with a photo of himself holding his infant son, and he cut himself out of his picture. What was left was a headshot of baby Walden. He and his colleagues scanned the image using a drum scanner, which captured the information from little squares of the image and turned that information into binary 1s and 0s—black and white pixels. This, close-up, is what the baby's eye turned into:

NIST

That first scan inaugurated the standard square pixels that for years dominated the creation and storage of digital images (and that Kirsch has despaired over). But it did not immediately kick off a revolution in computerized images because computers just didn't have the storage capacity. Even after digital cameras started becoming a consumer good, images were sometimes stored on cassette tapes.

It wasn't until storage capacity—and image compression—became affordable that digital pictures really took off. From there, it was only a matter of time before society came to the realization that it needed a tool to block the unceasing supply of digital baby images from view.