Some people have pointed to the selectric typewriter as evidence that it could have been used to produce the proportionally spaced text.
This is impossible and easy to show.
Page 108 of the below doc gives the width of each character in units independent of point size.
http://www.ibmcomposer.org/docs/Selectric%20Composer%20Operations%20Manual.pdf
Now take a look at the 18 August 1973 memo. Compare the length of the words "Bush's" and "period".
With the Selectric, "Bush's" should be 7+6+4+6+3+4=30 units long and
"period" should be 6+5+4+3+6+6=30 units long. So they should be equal in length.
It clear, however, that in the CBS document, "Bush's" extends beyond "period". Hence the Selectric could not have been used to create the memo.
What I find compelling evidence is the lack of erasures. It's almost impossible to type a full page - especially on a variable spaced typewriter - and not make an error. People forget, in the days before computers errors and their associated erasures were commonplace. The woman who invented White-out made millions. Why no errors in these memos? Because they weren't typed on a typewriter.
Now, that's detail.
Ooooh, what an elegant proof! You don't even need access to an IBM Selectric Composer to see that it could not possibly have been typed on one.
It shows how the documents were created using MS Word 2002
hey hombre,
tried hitting the ibm composer link, prompted me for a password/domain - don't have them.
Could you be kind enough to post a copy of the page in question here on FR?
Tx,
CGV
This is indirect proof of kerning, something that can only be done by a computer that has knowledge of kerning pairs (or by a professional typesetter). The kerning algorithm for word processing by computer was published in 1981.
Excellent work.
Thanks.