yet more at boing bing:

http://www.boingboing.net/2004/09/09/did_the_white_house_.html

Did the White House release forged documents about Bush's service record?
Charles at Little Green Footballs presents a persuasive argument that the memos recently released by the White House about President Bush's National Guard service are forgeries.

Of course, Karl Rove could have pulled such a nasty nasty trick, but...

Why the (expletives deleted) didn't CBS and the Boston Globe check the sources?

Do you really want to be on the side of the clueless? Ask them to produce the originals. Ask them to track them down the source. Ask them to find the perpetrators. If you truly believe Karl Rove is behind this, there should be no fear in your heart. But damnit, don't assume the people who carry your flag speak for you. Check it out.

And M E-L, on a less impassioned note, I have to think whoever typed these documents must be thinking now:

"Idiot! Notepad! Stupid! Stupid!"

AAAAAAGH! You guys are going to make my head explode! Twenty years of courier being called the "typewriter font" and the wide popularity of the cheap Selectrics has apparently made people think no one tried to make decent typographic machines before Apple. This is like judging the power of all computers in the 80s on the C= 64.

"Everything done in the memo was certainly possible -- without too much sweat -- at the time.":http://www.thudfactor.com/textpattern/index.php?id=905
.

It is possible that the documents were forged, but the "ooh, the superscript is a dead giveaway" or "ooh, the font is a dead giveaway" just flat isn't true.

From the NYT:

CBS News executives also produced a document released earlier by the White House about Mr. Bush's service that was clearly from a typewriter and had a superscript "th'' in it. CBS said it proved that some typewriters did indeed have superscript keys.

Also, the font is not the modern Times New Roman on Windows. Look at the numeric characters, e.g. 187 on Windows and compare with the memo.

LGF's analysis looks pretty damning, unless he's lying and toyed with the kerning using Quark or some such thing. Still, I think the burden of proof is now on CBS. I SO want these docs to be real, but if CBS is being so unprofessional as to pass off obvious forgeries, CBS needs to suffer.

Corrente talks about a typewriter of the era which includes "th" and "st" superscripts.

Can someone fill me in? Who would forge these memos? Aren't they damaging to W.?

Y'know, when we watched this the other night, I kept noticed that when they showed the docs on screen, they were little itty-bitty snippets of them that you couldn't really see. So I just embroidered and listened to the voiceover. Hmmmmm.

Also, WHY are the signature lines way over on the right? Did anybody bothering tabbing over there before there were computers? Wouldn't you just sign on the left?

Also, doesn't the second line of the subject look like a clumsy tab-over in Word?

OTOH, there are two places where "1st" appears with neither a superscript nor a space. Maybe there was a space like in "147 th" but they remembered to delete it?

I've got to say, I really thought this was good news, but at first close glance they look like total MS Word.

Here's a great roundup of typewriter theories. Main points: PC typefaces were explicitly designed to look like those used in typewriters, and there was a common IBM typewriter in use with proportional fonts and a "th" key.

But there are reasons why someone might make transparent forgeries.

These may be forgeries, but they're not transparently forged. We know that there were typewriters with these capabilities during that time period (we can see other national guard documents with the superscripted th), and we know that the Air Force had auditioned some of them years before, although they weren't common.

The bit I find most informative is that letters sometimes fall above and below the line, which is characteristic of typewriters. It doesn't mean that these were generated on a typewriter, just that it doesn't actually immediately resemble something that was printed on a computer.

Ennis, it may not have been obvious to you, but as someone who worked on a lot of different systems in the mid-80s the near-typeset quality docs at CBS' site set my alarms off immediately. Frankly, back then to get documents to look that good took a lot of work and some esoteric technology.

I applaud those who want these documents to be real for digging up scenarios explaining how they might have been produced, but to (possibly mis-)quote one type pro I read today "you'd stand a better chance of being eaten by pigs."

Here's a good explanation of just what you'd have to do to get these memos using technology available at the time (and please note, the guy doing the experimenting never did get a great match):
http://shapeofdays.typepad.com/the_shape_of_days/2004/09/the_ibm_selectr.html

Whereas if you just type the same memos into Word 2000 and print them out, you get exact duplicates.

Someone at CBS should have caught this, and they should have consulted an expert on typography and not just a signature expert. CBS is looking very bad here, and as a major media outlet they have a responsibility to do better.

Mark -- take a look at Kos's roundup on this topic. If you blow the images up, you'll see that the letters are above and below the line, like a typewriter. Sure, this can be faked, but now we're talking a sophisticated forger who forgot about proportional spacing.

Secondly, the more likely suspect here was the IBM Selectric Executive Model D. As Kos said:

http://www.dailykos.com/story/2004/9/10/34914/1603

...the [IBM Executive] Model D can produce those documents, not only did it do proportional spacing, you could order any font that IBM produced AND order keys that had the aftmentioned superscripted "th." Also you could order the platen, thats the roller that grabs the paper, in a 54 tooth configuration that produced space, space and a half and double spacing on the line indexing, this BTW was popular in legal offices. The Model D had to be ordered from a IBM salesmen and was not something that was a off the shelf item, typical delivery time were 4-6 weeks. Also, typewriter keys were changed in the field all the time, its not that hard to do.

Again, I don't know if these were forgeries. But they're certainly not obvious forgeries.

Lastly -- take a look at the breakdown where people enlarge the two images rather than shrinking them. You'll see that Word hardly produces an exact copy, there are small but consistent and significant differences.

Eaten by pigs, Ennis.

As I pointed out on my site, the only place where I consistently ran into proportional-spaced typing during my summers as a Kelly girl was in legal offices. And all I can say is, the output from those machines still looked considerably less sophisticated than what I produced in college using TROFF off an AT&T 3b20.

It's also interesting to note that the Kos story takes pains to point out how much effort both Apple and Microsoft put into getting Times New Roman to render true to the 1930s version. The reason it was difficult to do is because, frankly, it's difficult to do. Back in the days of real typewriting, an individual typewriter could be positively identified by the variations each one posessed due to imperfections in the manufacturing process and through normal wear and tear. And when recreating old fonts, fontographers get pretty zealous about which print samples represent the "true" version of a given typeface. It's actually a very cool field, in an extremely esoteric way.

If it turns out LtCol Killian moonlighted in legal typing pools, or in custom print shops, let me know. Otherwise folks are still trying to prove the world might really be flat. I mean, you can't see the curvature or anything...

On a slightly less dismissive note, the docs received (or at least presented by CBS) had obviously been run through a few generations of photocopying and/or faxing. That's why they're so smudged. Both processes introduce distortions to the original text (hence the smudging). Try it yourself: Take an origninal, photocopy it, photocopy the copy, repeat a few times, and compare the n-th generation child document to the original. You'll see changes in registration and clarity.

BTW, CBS's expert Matley wrote in 2002:

In fact, modern copiers and computer printers are so good that they permit easy fabrication of quality forgeries. From a copy, the document examiner cannot authenticate the unseen original but may well be able to determine that the unseen original is false. Further, a definite finding of authenticity for a signature is not possible from a photocopy, while a definite finding of falsity is possible.

Of course, all CBS gave him was a copy, and apparently a copy of only one document....

I'm sorry, I'll grant that to people without a background in printing, or to anyone not very familiar with old-school typing, the documents might not "on the face" look like forgeries. But CBS is going to end up losing major face over this, and they should have been more careful.

Let's examine the screen grab and the MS Word mock-up above:

The font in the "typewriter" memo is obviously narrower than the one produced in Word (is there a "Times New Roman Narrow" font?). Even the first word, "Memo", is crushed together on the typewritten memo but not in the Word version. Come on people - one word into the text and already alarm bells are ringing!

Even more blatantly, the superscript "th" on 187th is much higher up and narrower in the first case than the second. These parameters can of course be set in the Format-Font dialog in Word (including the width of superscript text? I doubt that!) but since the default Word settings produce perfectly good and plausible superscript, why would somebody bother adjusting that? The alleged "forger" stuck with all the default Word settings, supposedly.

Then there is the odd spacing between the first period and "Harris". Why is that different if the memo was produced in MS Word? Again, this can be adjusted but if that was done it suggests staggering inconsistency in the methods used by the "forger".

On the one hand he is supposed to have opted for all the default settings in Word (hence the apparent ease of "replication" - a broad use of the word, I fear), on the other he is supposed to have adjusted things like the height and width of superscript text and the spacing between sentences!

All of that said, it looks more and more likely that these are not original documents, but some sort of recreation, whether in word (or, as I think is more likely) a later typewriter.