To put a wooden stake in the heart of the vampire that is the IBM Selectric Composer typewriter theory, we've put-together a few graphics to illustrate one of the many reasons why this particular unit could not have produced any of the suspect CBS memos.
On an IBM Selectric Composer typewriter, all characters - regardless of font or point size - have fixed "unit values" assigned to them. This means that, while the actual physical space taken-up by a character will change with font and point size, the relationships between characters within that given font and point size are always the same.
The three examples we have chosen are:

If this memo had been typed on an IBM Selectric Composer, "have" and "time" would be the same length (21 units), "File" and "he's" would likewise match each other (18 units), and "drill" and "that" would pair-off as well (19 units). As you can see, they do not.
This is one of the many reasons why this particular typewriter is not a valid explanation as a source for these documents.
In case anyone else wants to play around with this same idea, here is the list of word lengths from the May 19, 1972 memo that I've put together (it doesn't cover every word, but should be good enough):
call 5 + 5 + 3 + 3 = 16 units
will 8 + 3 + 3 + 3 = 17
also 5 + 3 + 4 + 6 = 18
File 7 + 3 + 3 + 5 = 18
he’s 6 + 5 + 3 + 4 = 18
drill 6 + 4 + 3 + 3 + 3 = 19
told 4 + 6 + 3 + 6 = 19
says 4 + 5 + 6 + 4 = 19
that 4 + 6 + 5 + 4 = 19
unit 6 + 6 + 3 + 4 = 19
date 6 + 5 + 4 + 5 = 20
how 6 + 6 + 8 = 20
May 9 + 5 + 6 = 20
now 6 + 6 + 8 = 20
have 6 + 5 + 5 + 5 = 21
with 8 + 3 + 4 + 6 = 21
Says 6 + 5 + 6 + 4 = 21
staff 4 + 4 + 5 + 4 + 4 = 21
time 4 + 3 + 9 + 5 = 21
been 6 + 5 + 5 + 6 = 22
Bush 7 + 6 + 4 + 6 = 23
from 4 + 4 + 6 + 9 = 23
1972 6 + 6 + 6 + 6 = 24
fixed 4 + 3 + 6 + 5 + 6 = 24
three 4 + 6 + 4 + 5 + 5 = 24
come 5 + 6 + 9 + 5 = 25
flight 4 + 3 + 3 + 5 + 6 + 4 = 25
other 6 + 4 + 6 + 5 + 4 = 25
think 4 + 6 + 3 + 6 + 6 = 25
could 5 + 6 + 6 + 3 + 6 = 26
about 5 + 6 + 6 + 6 + 4 = 27
wants 8 + 5 + 6 + 4 + 4 = 27
Memo 9 + 5 + 9 + 6 = 29
Phone 6 + 6 + 6 + 6 + 5 = 29
would 8 + 6 + 6 + 3 + 6 = 29
talked 4 + 5 + 3 + 6 + 5 + 6 = 29
accept 5 + 5 + 5 + 5 + 6 + 4 = 30
follow 4 + 6 + 3 + 3 + 6 + 8 = 30
coming 5 + 6 + 9 + 3 + 6 + 5 = 34
options 6 + 6 + 4 + 3 + 6 + 6 + 4 = 35
Posted by: Deus ex Macrame at September 13, 2004 12:39 AM
Might be useful to link to this, too, in which a Selectric Composer was used to actually try to duplicate part of a forged Killian memo.
Posted by: Ed Jordan at September 13, 2004 08:08 AM
Your examples of the functioning of unit values for calculating type width are not true for the IBM Composer according to IBM's website. The reason why is that that model of typewriter had the capability to adjust the unit value to justify or adjust the type on the document. This would be true for non justifying typewriters but not the IBM Composer. Also, that typewriter was actually a typesetter for the production of press ready documents, so the type would show up as almost word processing - laser print quality.
J. Smith
Posted by: John Smith at September 14, 2004 08:08 PM
True, J. If you had a cold type typesetter in your National Guard office and you had training on it, and extra type balls, you could do proportionate and superscript.
But you'd still have to wait a decade or so for the font face with the same spacing as the memos.
Other than that - no problem.
Posted by: John at September 15, 2004 03:36 AM
John:
A couple of points: the IBM Selectric Composer did not achieve justified margins by varying the spaces within words. It varied the spaces between words:
"To justify a line of type printed with the SELECTRIC Composer, it is necessary to vary the amount of space between words (but not between characters within a word) to maintain the proportionality of the characters."
That's from this site.
The spacing for characters on the IBM Selectric was determined mechanically, which is spelled out in great detail here. There were fixed units assigned to characters, with those "units" having the ability to be varied for a given font. It did not have the ability, as Word does, to make alterations to word spacing on the fly (unless you switched type balls, of course).
I hope this makes sense. Let me know if it doesn't and I'll take another crack at it. If I've misunderstood your point, or am otherwise incorrect, let me know and I will endeavor to address things.
Posted by: Deus ex Macrame at September 15, 2004 12:49 PM
Thanks for signing in, . Now you can comment. (sign out)
(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)