Saturday, September 09, 2006

My first experience trying to correct a Wikipedia falsehood

For my Day Job at Michigan State we've been working on celebration of 50 years of computing at the university. MSU engineers started building the university's first computer, MISTIC, in 1956 -- fifty years ago.

Click to see full-size screen shot

MISTIC was based on a design by the legendary John Von Neumann at Princeton, a design offered to the world and replicated 15 times. Wikipedia reports that MSU also built a computer called MSUDC. There is no evidence that MSU built such a critter. It's likely that someone somewhere ran across an alternate name, never adopted, for Michigan State University Digital Computer. (MISTIC was Michigan State Integral Computer.)

Wanting to contact the author of this false information, I found links to a sort of Wikipedia screen name, and no evidence for a human's real name. So for my first venture into this world, I signed up to edit a Wikipedia article. The article claims:


MSUDC
From Wikipedia, the free encyclopedia

The Computer Center once housed the early computers
MISTIC and MSUDC.

The MSUDC or Michigan State University Discrete Computer, was a first-generation computer built by Michigan State University

based on the IAS architecture developed by John von Neumann. As with almost all computers of its era, it was a one of a kind machine that could not exchange programs with other computers (even other IAS machines)




This stuff simply isn't true.

There never was a thing called the "Discrete Computer" at MSU. Some people may have called MISTIC some such thing, but MSU built one IAS computer in the late 50s, not two.

It is also most manifestly false that programs could not be exchanged among the IAS computers. Trust me, in June I interviewed the man who led the development of MISTIC, Dr. Lawrence W. Von Tersch, and he emphasized how important it was that the University of Illinois had shared mathematical and statistical software that they'd developed, which helped jump-start scientific computing at Michigan State.

And one of the IAS computers was built in Australia -- Sydney's SILLIAC, like MSU's MISTIC, a close clone of ILLIAC. They developed a diagnostic program that played out "Waltzing Matilda" -- these machines had speakers to chirp out the sounds for diagnostic purposes -- and that program was used every day at 8:00 a.m. circa 1959 to test MISTIC.

Even the statement that these computers were based on the IAS architecture is false. They were based on the IAS design. Von Neumann is famous for the stored program concept -- that a computer would have program code and data in a single shared memory. But the design that IAS offerred the world was a very specific design for a computer. You know, schematics. Each implementation had its quirks, but from what I've gleaned, ILLIAC, MISTIC, and SILLIAC were all quite close to one another in actual function.

So we have some random author of an article in an encyclopedia asserting the existence of a computer that never existed, and asserting that the 15 or so Princeton IAS Von Neumann computers couldn't run each others' programs. Both claims totally false.



In other words, virtually the entire substance of this Wikipedia entry is simply wrong.

This is my concern about Wikipedia: not that they will get "why is the sky blue?" stuff wrong, because a million people will correct that. Instead, they will get small, but important pieces of history wrong, because some well-meaning writer with poor research skills will get an important niche story wrong.

Why can't I find the real name and contact info (real phone number, primary e-mail address) for the person who wrote this article?

Here is my call to the author of the article filled with false information:

It is distressing that Wikipedia gives the imprimatur of authenticity to an
encyclopedia article written by an anonymous person who does not have the facts
straight.
I request that the author of this page contact me at
wikipedia@richardwiggins.com


For stories like this, Wikipedia puts the kid struggling to write a paper for school on the same level as a tyro who wrote a piece about an obscure piece of history. For some people I suppose that's very cool. To me it's scary.

Friday, September 08, 2006

Memo from New York Times: What matters most happens on the coasts

David Pogue, who writes about computers for The New York Times, declares in his most recent column:

"White-collar types on both coasts have become addicted to the BlackBerry, thanks to its ability to display e-mail instantly as it arrives and to synchronize with your computer back at the office over the air."

Oh, so mo one uses Blackberries in the middle of the damn country? Does Blackberry offer no service other than on "both coasts"?

For decades the New York and LA media elites have been jealous of each other. The de facto detente they reached is to refer to, and to cover, "the coasts". This most arrogant formulation omits the middle of the entire country -- including the North Coast, the Great Lakes, home to thousands of miles of coastland and the greatest collection of fresh water on Earth.

Does Mr. Pogue assume that the Blackberry does not function in Chicago, home of Boeing executives? Does he postulate that the Blackberry does not operate in Memphis, home of Fed Ex? Does he forget that big oil executives work in Texas headquarters? Does Pogue suppose that a white collar employee of 3M, based in Minneapolis, is deprived of the opportunity to be among the Blackberry cognoscenti? Does he think that Warren Buffet, a billionaire who made his fortune out of simple propositions, could not have a Blackberry in Tulsa if he thought he needed one? (No doubt Buffet would rather drink a Coke or read the print edition of the Wall Street Journal instead of poking and peeking at a Blackberry.)

Adding irony upon irony: Research in Motion, or RIM, which invented and markets the Blackberry, is a Canadian company, based in Toronto. Not on the coasts Pogue refers to, but near the Great Lakes.

Please, drop your coastal arrogance.

Wednesday, September 06, 2006

Overloaded Google News links to tasteless Steve Irwin story parody

Prominently listed among the 3,462 articles about Steve Irwin, Google News offered this headline:

Stingray Apologizes for Crocodile Hunter Irwin's Death

Click to see full-size screen shot

Of course, it links to a parody news site:



No doubt there are a lot less tasteful Steve Irwin jokes in the pipeline. I just don't understand why Google News includes parody sites in a news service. At the very least, Google should be transparent about which sites it includes and why. What are the criteria for inclusion?

http://www.laist.com/archives/2006/09/06/stingray_apologizes_for_crocodile_hunter_irwins_death.php

Monday, September 04, 2006

80 gig for 80 bucks - and tiny

Once upon a time it was amazing that disk storage cost a dollar a megabyte. Today I bought a disk drive that was the size of a PDA. It cost $90 for 80 gigabytes. That's 80,000 megabytes if you're paying attention.

It's a LaCie drive, USB powered, no need to carry an external power brick.



Dr. Lew Greenberg, retired director at my day job, recently told me that Moore's Law doesn't apply to disk; improvements in disk storage outpace improvements in semiconductors.

You can store an awful lot of stuff on an 80 gig drive that fits in your pocket. Expect much more corporate espionage and identity theft.