September 10, 2004
An Excuse to Play with Photoshop
So I've been intrigued by the speculation that recently released documents about George Bush's military service might have been forged. The blogospere is in a veritable tizzy. To be clear: I could hardly care less whether George Bush was actually AWOL from the Texas Air National Guard 30 years ago, it's the documents themselves that have piqued my curiosity.
Thus, being a Photoshop nerd, I've decided to check it out myself. Before I started, I checked out best arguments that I could find that lay out the case for and against forgery. Then I went to work.
I've seen claims that simply typing the memo into Microsoft Word using default settings will yield an exact match to the memo. This isn't precisely true. I first had to modify my margins to match the memo (1.5 inches on the left, 1 inch on the right) and then disable the "auto-list" feature that will automatically indent numbered lists as you type them. But otherwise, it came out pretty darn close (click for a larger view):
A lot of speculation has centered around the font used in the memo, and that this font, as printed on the memo, could only have been produced from a modern computer. In fact, Times New Roman is a typeface has been around for quite some time, having been created in 1931 for the Times of London.
Now, a font is a font. Whether it's displayed on a computer screen or typed with a typewriter or printed on a printing press, it's supposed to look nearly the same in every case. When the personal computer and desktop publishing exploded onto the scene, most electronic fonts were created to mimic existing fonts as closely as possible. This includes not only the shape of the letters, but the spaces between the letters. So you can see how, after aligning the margins correctly, a memo typed in Microsoft Word can match a memo typed on an advanced typewriter in 1973.
Still, let's look real close. Here's one chunk of text, selected more or less at random. The top line is from my Word document, the bottom is from the TANG memo. Even after multiple photocopies and scanning, you can tell that the memo looks slightly different. One thing stands out for me, the letters all appear taller and skinnier in the Word document, look especially at the ascenders and descenders in letters like p, k, d, b, etc.
Now look at this chunk of text and you will see a similar difference, especially in the controversial th. What's this prove? Nothing. We won't know anything for certain until someone recreates the memo using a 1973-era typewriter.
Now, a font is a font. Whether it's displayed on a computer screen or typed with a typewriter or printed on a printing press, it's supposed to look nearly the same in every case.
Not true. Charles showed that the same font rendered in a different program will be measurably different. I have done the same thing with the second memo, using the same passage in Word X, AppleWorks 6 and Apple Text Editor. There are differences in how letters and words are spaced and with line leading. If "a font is a font" each of the three would be identical. Considering that three Mac text applications can't even render the identical font file the same way I still find it rather fishy that a memo banged out over thirty years ago on a typewriter can match the Word document closer than the AppleWorks or Apple Text Editor could. (I haven't posted a blog entry about this since its basically a rehash of what Charles posted, but I can post up the images if needs be).
We won't know anything for certain until someone recreates the memo using a 1973-era typewriter.
Jeff Harrell has done just that. The results were close, but still no where near the match of the Word document. But the kicker is to even get that close lots of adjustments had to be made, and at one point one font ball had to be replaced with another (in order to do the superscript). So, again, the question comes down to how possible do you think that a) the TexANG had an IBM Selectric Composer (not your average everyday "bang out a memo" piece of equipment), had it tuned in such a way that it matched up perfectly with a Word document (but not an AppleWorks or Apple Text Editor) and the unit used a completely different typeface and machine to bang out every other memo from that office other than the four CBS dug up?
Posted by: Nathan at September 11, 2004 02:21 AMAlso take a look at Fygar's comments to this post. One of the better typographical overviews and tackles the question of why "a font isn't a font."
Posted by: Nathan at September 11, 2004 02:32 AMThanks for your comments, Nathan. I appreciate the points you are making, but I'm not convinced by the close-but-not-exact match produced by the default Word settings theory. When viewed in detail, the letterforms of the text look different enough to make me suspicious.
However, I have a new theory that may explain this discrepancy, use it as fuel for your fire if you like.
Posted by: Ken at September 11, 2004 03:10 PMBen-Lag
Capitalism, Chinese-Style
Year of the Sleeping Dog
Learning from Each Other
Home at Last
We Are Family
Ladies Man
Feeling Blessed
Traveling in a Pack
January 2006
December 2005
November 2005
October 2005
September 2005
August 2005
July 2005
June 2005
May 2005
April 2005
March 2005
February 2005
January 2005
December 2004
November 2004
October 2004
September 2004
August 2004
July 2004
June 2004
May 2004
April 2004
March 2004
February 2004
January 2004
December 2003
November 2003
October 2003
September 2003
August 2003
July 2003
June 2003
May 2003


