The Story of Mel

April 7, 2001 10:30 PM

The Story of Mel - Almost everyone's seen the Story of Mel on USENET or via email... the story of the guy who wrote programs for a particular ancient drum computer by using the characteristics of the drum to handle memory allocation and time delays. In a footnote on the Jargon File, it seems that his last name is known... An interesting footnote to an interesting and probably true story.
posted by SpecialK (5 comments total)

I swear I'll never whine about Visual Basic or C again! *grin* That was beautiful. I knew a man years ago who could write machine code. This was long before I got "into" computers. One look and my brain cramped.

Thought about posting this as a separate thread, but somehow they seem related. It's a song/video (requires RealPlayer or Windows Media Player) entitled Every OS Sucks by Three Dead Trolls. Absolutely hysterical.

Thanks to userfriendly.org for this.
posted by NsJen at 10:47 AM on April 8, 2001

The Mel story is a relatively old urban legend-type story. In Jan Harold Brunvand's latest book, "The Truth Never Stands in the Way of a Good Story," in an article written by his son Erik, the original posting of the Mel story is dated as May 21, 1983. He writes that "Mel is most likely a fictional programmer, although the story is told as a true account and uses real machines and real languages as examples," and goes into the different themes in the verse that makes him suspect it's hacker legend (as opposed to an honest-to-god account). For what it's worth...
posted by amyscoop at 12:04 PM on April 8, 2001

Here's a version of the Story of Mel, annotated by Erik Brunvand, a Computer Science professor at the University of Utah and the son of Jan Harold Brunvand.
posted by waxpancake at 12:31 PM on April 8, 2001

Note that Brunvand's annotations are dated 1996. The Jargon file note confirming his name is dated 1999.

The "legend" is specific as to time, place, and equipment, which is often an easy way to disprove them. In this case, the attributions seem real. Here's an LGP-30, for example.

I'm surprised nobody's followed up with Mr. Kaye, or (presumably) his heirs, to find out his side of the story.
posted by dhartung at 2:37 PM on April 9, 2001

It's fun to read about folks who know their machines the way Stradivarius knew fiddles. But even hexadecimal is one step up from rock bottom. You programmed the Altair 8800 directly in binary, one byte's worth of 1s and 0s at a time, from a row of front panel toggle switches.
posted by jfuller at 2:41 PM on April 9, 2001