Why am I not a skeptic about claims that the Killian documents are forgeries, like Atrios and Joshua Micah Marshall? Because the work I actually get paid to do -- and have done since 1980 -- has always involved squinting at type.
In the early and mid-1980s, I worked for companies where "typesetting" was done (for sci-tech books and journals) on IBM typewriters. I did some spot typing myself back then -- switching from the standard IBM Selectric "golf ball" to the one with all the scientific notation, or, on a non-Selectric typewriter with the traditional long, spindly type elements that swung around to hit the paper, using what I think we called a "dead key" -- a key that typed whatever character was on a removable type element you snapped in and out. (These type elements were kept, if I remember correctly, in a wooden case by the typewriter. If you're much younger than I am, this would all look absolutely Victorian to you. But we used it twenty or so years ago.)
I later saw proportional-font type stumble to its current state -- really good-looking type is now generated on desktop computers, but there was some truly ugly desktop typesetting done about fifteen years ago that was trumpeted as state-of-the-art.
These are issues I've thought about for years. I've had to -- it's been part of my job. I don't know about every way type has been generated since the 1970s, but I really do know a lot about type, and what's being said by the people who believe these documents are fake makes a great deal of sense to me.
As for my Rove theory, I'll say it again -- go see Bush's Brain or read the book. Then decide for yourself whether you can imagine Rove ordering poison to be spread along a promising evidence trail, poison for which Democrats would be blamed.
I can't say for sure that this is Rove's work, but look -- if Bush wins, it really has to be open season on Rove. The press loves to say that Rove can't be caught because he leaves "no fingerprints." I'm sick of hearing that. How hard have major national news organizations tried to find the fingerprints? Have they tried as hard as, say, The New York Times tried to find evidence of chicanery in Whitewater? Clinton was a Southerner perceived as too "slick" to get caught back home, and the national press, far from feeling they'd met their match, took him on as a challenge.
The same thing has to happen to Rove. He needs to be destroyed. I just don't believe it's impossible to find concrete evidence of the immoral and illegal things he's done. He may well elect the next president of the United States if that doesn't happen.
ON THE OTHER HAND: Alicublog reproduces this ad for IBM's proportional-font "Executive," which shows the type quality the machine could produce; the ad is taken from a long post at Rising Hegemon, which, among other things, explains that the ad appeared in 1954. Interesting. The machine wasn't really a typewriter, but the technology began appearing in some IBM Selectric typewriters in 1966. But note this about the justified-font Selectric:
Since it has no memory, the user was required to type everything twice. While typing the text the first time, the machine would measure the length of the line and count the number of spaces. When the user finished typing a line of text, they would record special measurements into the right margin of the paper. Once the entire column of text was typed and measured, it would then be retyped, however before typing each line, the operator would set the special justification dial (on the right side) to the proper settings, then type the line. The machine would automatically insert the appropriate amount of space between words so that all of the text would be justified.
Hard to believe the military would buy typewriters that required typists to type every document twice.