Corante

About this Author
Ernest Miller Ernest Miller pursues research and writing on cyberlaw, intellectual property, and First Amendment issues. Mr. Miller attended the U.S. Naval Academy before attending Yale Law School, where he was president and co-founder of the Law and Technology Society, and founded the technology law and policy news site LawMeme. He is a fellow of the Information Society Project at Yale Law School. Ernest Miller's blog postings can also be found @
Copyfight
LawMeme

Listen to the weekly audio edition on IT Conversations:
The Importance Of ... Law and IT.

Feel free to contact me about articles, websites and etc. you think I may find of interest. I'm also available for consulting work and speaking engagements. Email: ernest.miller 8T gmail.com

Amazon Honor System Click Here to Pay Learn More

In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

The Importance of...

« Copyright Office Report on INDUCE Act (IICA) Complete But Not Yet Public | Main | SAVE BETAMAX »

September 14, 2004

CBS Memo Defense: Richard Katz Is Wrong About Ones and Els

Posted by Ernest Miller

I've been following the debate concerning the CBS memos because I believe it has a lot to teach about the future of media, authority, journalism and blogging. It is also a fascinating story in its own right. In any case, there are worse places to start reading about the beginning of the controversy than ABC News' The Note (The Note: Sep. 10, 2004). Of course, the most recent Washington Post (reg. req.) article on the issues will get you up to speed quicker (Expert Cited by CBS Says He Didn't Authenticate Papers). See also, Seth Finkelstein (CBS (60 Minutes) Forged Memos Comparison Evidence and More concerning story that 60 Minutes documents on Bush may be fake).

In any case, despite mounting evidence that the memos are unlikely to have been created in the early 1970s, CBS News continues to defend the story vigorously, if not very successfully. For example, last evening's defense of the CBS memos is, once again, quite problematic. On September 13th, the CBS Evening News included a report where they introduced two new experts who claim to find that the memos are consistent with memos of the early 1970s. One of these new experts is a software designer who claims that the documents use a lower case "L" in place of a numeric one. From the transcript:

DAN RATHER: Richard Katz, a software designer found other indications in the documents. He noticed the lower case "L" is used in documents instead of the actual numeral one. That would be difficult to reproduce on the computer today.

RICHARD KATZ (Software Designer): If you were doing this a week ago or a month ago on a normal laser jet printer, it wouldn't work. The font wouldn't be available to you.

This brief exchange isn't quite clear, but I think what Katz is getting at is that many typewriters of the era "cheated" by having people use a lowercase "L" in place of a numeric one. There was no key for a numeric one on many typewriters of the day. Katz claims that a numeric one isn't being used in the documents, but a lower case "L" is. This would be evidence that the documents were written on a typewriter. At least, I think that is what the argument is.

However, it is not only wrong, it actually contradicts the claim that the documents were typewritten on a proportionally spaced typewriter. Why? Glad you asked.

First, I have no idea how Katz can claim to distinguish between a numeric one and a lowercase "L" on these documents. They look exceedingly similar to me. Given the degradation of the copies CBS has deigned to let us see, how can Katz be so sure?

For example, let's take a look at lowercase "L" and numeric one in Microsoft Word's Times New Roman. I've provided examples in both 48pt type and 12pt type:

Example_One.JPG
There are subtle differences between the shapes of the two characters, but I doubt highly that one could easily distinguish them in 12pt type on a poor copy. The most important distinction between the two characters is their spacing. The numeric one has space on either side so that it is the same size as other numbers (in other words, all the numbers are monospaced). This makes numbers line up in nice columns. The lowercase "L" on the other hand, is proportionally spaced so that it is quite narrow. The difference in spacing is subtle but can be seen in the documents provided.

To see the distinction, consider this line (which was highlighted in the CBS Evening News report) from one of the CBS memos (Bush National Guard Memo August 18, 1973 [PDF]). I've extracted the line that was partially highlighted in the report. The comparison text is 12pt Times New Roman from Microsoft Word. The text has been reduced in size to 96% of the original in order to match the CBS memos.

The first line is Microsoft Word text using both numeric one and lowercase "l" as appropriate. It matches quite nicely with the memo's text, which is on the second line. The third line is what happens when you use lowercase "L" as numeric one in "l87". The difference is quite clear:

Comparison_One.JPG

In this second example, the first and second lines are the same. In the third line, I've used lowercase "L" followed by a space such as "l_87". The difference is more subtle but it is still rather clear, given the degradation of memo CBS has allowed to be made public:

Comparison_Two.JPG
From the above two examples I conclude that it is most likely that the device which produced these documents had both a numeric one and lowercase "L" and that the typist actually used them as appropriate, in general.

Finally, another comparison from the typed letterhead of one of the other memos (Bush National Guard Memo May 4, 1972 [PDF]). The first line is what happens when you use all lowercase "L"s. Clearly, that is not what happened. The second line uses numeric ones, the third is from the memo, and the fourth is lowercase "L" with spaces between letters. Althought the difference is quite subtle, to my eye the numeric one is a closer match:

Comparison_Three.JPG
However, this image is really to make another point. The reason many typewriters "cheated" in not having a numeric one and only had a lowercase "L" was because they were monotype. The shape of the characters was extremely similar, and in monospaced type the spacing was identical. So, it made sense to have a cheat. However, in proportionate spaced type, lowercase "L" and numeric one have very distinct spacing and lowercase "L"s look odd in many cases when used as "ones." And, if you were typing some columns of numbers, the use of proportionately spaced lowercase "L" would really mess up the alignment. Thus, though I am no expert, I would imagine that most proportionate spaced typewriters would have distinct characters for numeric one and lowercase "L". Of course, one might argue that the typewriter used for these memos, although proportionately spaced, did not have separate characters for numeric one and lowercase "L". I consider that improbable, but if there is evidence that proportionate spaced typewriters continued to leave out numeric ones, I would be interested in seeing it.

Of course, if proportionate spacing typewriters had both "1"s and "l"s, why would the typist only use lowercase "L"? After all, the typist of the memos was clever enough to insert a superscript special character (the infamous "th") and do some really nice center aligning. Why wouldn't the typist have used both numeric one and lowercase "L"?

Thus, I conclude and assert that Katz really doesn't know what he is talking about and his defense of the authenticity of the memos is really quite weak.

Btw, this is a tentative conclusion. I'm no expert and the documents provided by CBS are quite poor quality though they claim, according to the Mercury News (reg. req.), that they have first-generation copies (CBS stands by story on Bush's service, defends memos' authenticity):

CBS spokeswoman Sandy Genelius said Monday that the network possesses what it believes to be so-called first generation copies duplicated directly from the original documents.

But the copies posted on its Web site are somewhat blurred and speckled, suggesting repeated copying.

Genelius said she could not explain why the versions posted on the CBS Web site appear to have been repeatedly copied, while the copies the network relied on for its reporting were not.

Sure would be nice if CBS would make high quality scans of the documents available. Might help CBS' case.

Nevertheless, I believe my demonstration here casts serious doubt on Katz's claim.

UPDATE 0930 Pacific Time

More detailed "1" vs. "l" analysis by Joseph Newcomer here: The Bush "Guard memos" are forgeries!.

Scroll down to "Additional Update 13-Sep-04 - The L with it!"

Comments (34) + TrackBacks (0) | Category: Blogging and Journalism


COMMENTS

1. Crush T. Velour on September 14, 2004 08:06 AM writes...

Thanks. This is just the information I was looking for. When I heard Katz's argument about the 1s versus lower case Ls, my head was spinning trying to make sense out of it. I pulled up Word and immediately saw that 1s and lower case ls were virtually identical except for the spacing but that 1s on the alleged documents looked more like 1s than Ls.

Now it is obvious. Richard Katz is a liar being propped up by CBS News to confused the credulous and naive.

I recommend that you immediately print this out and offer it to the WSJ for publication.

Permalink to Comment

2. A.W. on September 14, 2004 04:36 PM writes...

Nice post. Just a tech issue. It is coming out on my browser to that the stuff on the left side of the screen is badly overlapping the text. i had to phyically cut and paste the text from the document into word, and then save the pics on the computer and then read them in microsoft picture viewer, to see your demos.

I am working on my own post over at freespeech.com. I take every L and 1 (one) in the memo and show you that they don't look like each other. I mean the L's don't look alike, the 1's don't, etc. So how can they say that they are the same?

Also let's not assume for one moment that this computer guy saw better copies than we did. the other guy on the show, the typewriter repairman admitted he was only shown a printed pdf for about 15 minutes.

Permalink to Comment

3. Raj on September 14, 2004 04:39 PM writes...

see Additional Update 13-Sep-04 on page http://www.flounder.com/bush2.htm by
JOSEPH M. NEWCOMER

Permalink to Comment

4. Ernest Miller on September 14, 2004 04:42 PM writes...

A.W.,

Thanks, it was due to a non-standard trackback. It has been corrected.

Permalink to Comment

5. giad on September 14, 2004 04:59 PM writes...

The way I'm reading this, all you've proved is that the memo typist did not use the lower case "L" in MS Word. Should the memo be authentic (which seems to me highy unlikely), we'd need to know the spacing for lower case "L"s for an IBM selectric or other possible machines of the era. My recollection of typing from that era is that, even if a tywriter were equipped with numeric ones, it was a conceit among experienced typists to use the lower case "L".

Permalink to Comment

6. kd on September 14, 2004 05:07 PM writes...

giad is right. I have worked with many experienced typists who, once they had trained to use the lowercase "L" on their old typewriters, had a hard time converting to their new computer keyboards, which, of course, had the full complement of numbers and letters. And yes, you could actually see the differences in their work.

Permalink to Comment

7. David Gillies on September 14, 2004 05:11 PM writes...

The Occam's razor explanation: the guy who forged these 'memos' used lowercase 'L's on occasion in a clumsy attempt to defeat Microsoft Word's auto-superscripting feature. That's consonant with the author of these documents being generally clueless from a technical standpoint.

Permalink to Comment

8. Call me snake on September 14, 2004 05:12 PM writes...

Look back at the title of this post...

What does Ernie Els have to do with this story???

I kill me.

A poor attempt to inject humor into a well done, meticulous item. Perhaps Hurricane Dan's collapsing seawall has something to do with my happiness.

Permalink to Comment

9. Ernest Miller on September 14, 2004 05:22 PM writes...

giad and kd,

The expert from CBS's claim seems to be that the typist used only lowercase "l"s instead of "1"s. I believe that I have shown that to be unlikely. In other words, I believe I have shown that the typist used both "1"s and lowecase "l"s, whether in MS Word or not. "1"s and "l" will have different spacing on a proportional spaced typewriter. In this case, similar to Microsoft Word, but any proportionally spaced typewriter will show similar spacing distinctions.

I've also shown, once again, that the evidence is quite consistent with a modern forgery.

Permalink to Comment

10. Brian on September 14, 2004 05:49 PM writes...

Ernest is right – the definitive analysis, confirming Ernest's and showing that these are indeed “ones” and not “ells” is by Joseph Newcomer which you can (and should!) see at:

http://homepage.mac.com/cfj/newcomer/index.htm

--Brian

Permalink to Comment

11. Elder of Ziyon on September 14, 2004 05:58 PM writes...

The IBM Selectric HAD a numeric 1 key, and I'm pretty sure the Composer would have also.

http://www.mytypewriter.com/IBM_Selectric_closeup.jpg

Permalink to Comment

12. aejc on September 14, 2004 06:01 PM writes...

A further 1/l note, bolstering giad: many typewriters had a 1 (one) key, but the one and the lower case l were identical as far as the actual character on the strikeplate. The Selectric ball may have even combined the two, using the same imprint for both keys. Accomplished typists would often use the short cut of typing a lowercase l with the right hand, rather than a 1(one) with the left hand, and many older typists still use this shortcut. So, the Katz-Rather exchange excerpted is spurious.

Permalink to Comment

13. Michael on September 14, 2004 06:02 PM writes...

Katz is wrong about "undoing" the superscript "th" as well. He makes it sound complicated but in MS Word 2000, it's as simple as pressing CTRL+Z immediately after Word applies the superscript. Word reverts to what you originally typed. This works for all of the auto-format elements as far as I know.

Permalink to Comment

14. Vik Rubenfeld on September 14, 2004 06:33 PM writes...

"If you were doing this a week ago or a month ago on a normal laser jet printer, it wouldn't work. The font wouldn't be available to you."

Ahhh..... how about by pressing the el key? Katz can't possibly claim the lowercase el key isn't available.

Permalink to Comment

15. Chris on September 14, 2004 07:08 PM writes...

If anyone wants to send a fax to CBS News calling for Rather's resignation, the fax number is (212) 975-1998

Permalink to Comment

16. jwniii on September 14, 2004 07:16 PM writes...

Poasted elsewhere as well:

Either Rather plus his producers are colossally stupid, really, or they understood what they were saying and hoped to fool some colossally stupid people. I go with Rather is colossally stupid, as follows:

We bloggers say they are forgeries because we say someone went out of their way to make them look original.

Here is what Rather said tonight:

1. "We checked the signature handwriting on the copies!"

We say, sure, but I can myself easily go out of my way to make documents with copies of Killian's signature. So could any forger. What's your point?

2. His expert said, "The documents show that the same key was used for a character one and a character ell. If you used different keystrokes in a Word Processor a laser printer would never print them the same!"

We say, sure, but I can myself easily go out of my way to make documents with the same key used for a character one and a character ell. So could any forger. What's your point?

3. His expert said, "The documents have a superscript 'th' and same-size-lower-justified 'th'. Someone using a word processor would have to go out of their way to duplicate that!"

(Uh, gee, can't even brain-dead Dan Rather see through that one????)

We say, sure, but I can myself easily go out of my way to make a superscript 'th' and same-size-lower-justified 'th'. So could any forger. What's your point?

(Not to mention that the Blogs are full of the fact tonight that clearly lowercase ells were used to turn off auto-superscripting!!!")

4. Rather says, "Look here, some of the dates in our documents match some other dates known in the public domain. That contributes to their authenticity."

This is really getting tiresome. Dan Rather really is as stupid as he looks.

We say, sure, but I can myself easily go out of my way to find public domain dates to put into a document. So could any forger. What's your point?

5. I wasn't going to add this because it doesn't have the same idiotic obvious flavor of the others, but it bears saying.

That "expert" Glennon (let's hear it for 15 minutes of fame!) says that all of the features in the memo were existent at that time and could be ordered as special features. Rather said essentially the same thing in his first defense on Friday.

We have no interest in a general assertion of that nature. Show us the machine that had ALL of them simultaneously, period. And I add, that machine must have the pseudo-kerning of the documents, it must tuck the "r" under the "f" in "from" and it must do it in a reasonably automatic way that would make us think a 1972 Lt. Col. would use it to dash off memos.

(All of those type capabilities probably existed SOMEWHERE in 1950, 1925, 1875? I'm just guessing. The issue is not did the type capabilities exist but did they exist in one machine that a 1972 Lt. Col. would use to dash off memos?)

So, really, multiple times tonight Rather and his experts used the defense "you would have to go out of your way to get those results in a WORD PROCESSOR!!!!!!!!!!!!!!!!!!!!!!!!!!!!"

Uh, yes, that is by definition what a forgery is.

So, I reiterate,

1. if Rather could not see that then he is monumentally stupid (my explanation) or,

2. it was so painfully obvious that it was extraordinarily transparently bad logic that I had to consider the possibility that he knows better and just figured it would totally dupe his stupid listeners. I admit I had to re-play the part about the 1's & l's to fully grasp the total stupidity what they were saying, so maybe Dan is right and can get away with it.

Respectfully submitted,

jwniii

Permalink to Comment

17. Yancey Ward on September 14, 2004 07:41 PM writes...

The argument of els vs ones is a silly one, in my opinion. The text of the memos is clearly proportionally spaced and a lower case el would be given a narrower space than most of the other letters in the alphabet. However, any proportional-spacing capable typewriter would have had a key for the numeral one since typewriters have to be able to produce aligned columns of numbers (typewriter designers weren't stupid like bloggers). The only scenario in which Katz could be correct in his argument is where the typist chose not to use the key for the numeral one, not that the typewriter did not have one. While this isn't implausible, his arguments for these being lower case els are unsupported by the physical evidence. This was just another smokescreen put up by be the increasingly desperate CBS News Division.

Permalink to Comment

18. js on September 14, 2004 07:56 PM writes...

aejc is correct -- I took adminstrative training in 1971, and was taught to use the lowercase L to substitute for the numeral 1 because both speed and accuracy improve if you stay one row of keys either side of the *home* row. With proportional type, this practice ended -- the lowercase L no longer works, always because of its allocated letterspace and often because a given font will include two very distinctive characters. I would expect that some of the *real* TANG documents would use lowercase L to substitute for 1, but a younger person -- anyone who learned to type on the computer -- typing the forged documents would use the numeral.

Permalink to Comment

19. old maltese on September 14, 2004 08:19 PM writes...

Monospaced ones. Eureka moment. Excellent work, Ernest, absolutely excellent.

Permalink to Comment

20. jeremiah on September 14, 2004 09:15 PM writes...

Hrmm. Now that I am seeing the fonts at a large enough size that I can contrast and compare, it does appear to me that there are some pretty significant differences between the microsoft font and the documents. Up until now I was pretty confident the documents were forgeries. Now I am not so sure.
Aside from the differences in the 1's, the e's and the o's look considerably different (in the CBS document they look skinnier). The way the letters bunch up is different between the two. The g's are completely different (look at the lower half). The i's have much more pronounced serif's in the CBS documents and so on.
I came here looking for more support of what I already thought, and instead I find myself doubting more then ever. We should be committed to the truth here, and we can't have it both ways: saying the documents match up perfectly when they clearly don't, and then blaming differences on fax artifacts is not living up to that committment.

Permalink to Comment

21. Dead Parrot James on September 14, 2004 10:13 PM writes...

I'm beginning to feel like a broken record.

There actually are some L's used as 1's in one of the memos, possibly two. Not that it helps their case any. It appears to be more of an amateurish attempt to defeat Word's automatic superscripting since it will only auto-superscript if the preceding text (not just preceding character) is a number.

If you'll notice in the Aug. 1 memo in the subject and 1st paragraph, the character used to represent the one in "1st" sits too close to the "s". You can read Newcomer's analysis to see why (the -1 C width for l vs. the +1 C width for 1).

http://www.deadparrots.net/archives/media/0409of_ls_and_1s.html

Permalink to Comment

22. Lynxx Pherrett on September 14, 2004 11:01 PM writes...

While I think the documents are probably forged based on the totaltity of the discrepancies, the CBS expert, Katz, may be correct about the ones and els.

First, redo your 48pt and 12pt tests in Word, this time alternating ones and els on the same line. The difference in height between the them in MS's Times New Roman font becomes obvious.

When it comes to the PDFs available, we've got artifacts from JPEG lossy compression added to artifacts from scanning, added to artifacts from photocopying, which make it basically impossible to accurately compare individual characters.

On an electric typewriter with individual strike levers, including the proportional spacing IBM Executive, the character strike and carriage movement are separate actions. Say lever #23 had the "l" character.

On a single-pitch typewriter, the typist presses the one key, lever #23 strikes the platten, then the carriage moves one space to the left; when the typeist presses the el key, again it is lever #23 that strikes and then the carriage moves one space left. The ones and els are the same character.

On a proportional spacing typewriter almost the same thing would happen: lever #23 would strike whenever the one or el key was pressed, but the carriage would move differently, a full space for the one key, somewhat less for the el key. The one and el characters would be identical, but the spacing would be different.

Maybe the cleaner copies CBS has are sharp enough for Katz to determine the ones and els are the same character, on the poor PDF images they superficially appear to be the same height but there are so few els it's hard to tell (most of them seem to occur in the word "flight" where their tops merge with the "f" adding to the difficulty).

Katz is wrong about the difficulty in overcomming the autocorrect superscript of ordinals in Word. Besides the Cntl+Z method; another way is to put a space after the number and type the "th", left arrow twice and backspace. That deletes the space and positions the full size "th" next to the number.

The 1 August 1992 "order" does have a space between the number and the "th" and "st" as if (if these were "forged on Word") the writer forgot to go back and remove the space after using the above method. It occurs in "147 th Ftr Intrcp Grp" in the second and third paragraphs, and in "9921 st Air Reserve Squadron" in the second paragraph. That's one of the inconsistencies that has me leaning toward clumsy forgery by somebody who didn't know Word very well, because the best way to do overcome the superscripting is to go into the autocorrect controls and uncheck the 'automatically replace ordinals with superscript' option. Then you don't have to worry about it ever again.

It's the inconsistency that really bugs me. Killian was the commander of the 111th squadron in the 147th group. If he typed, even on an IBM Executive, then those would be numbers he typed often. He wouldn't inconsistently alternate between typing "111th," "111 th," "111" without an ordinal, and then occassionaly throw in using the superscripted-th key on his typewriter. The smae inconsistencies appear in "147th" and in "1st Lt Bush," as well as the really strange "111 F.I.S." followed by the same in the nexe paragraph, only this time with a superscripted ordinal, in the 4 May 1972 "order."

Permalink to Comment

23. Ernest Miller on September 14, 2004 11:22 PM writes...

Cleaner copies may show that the characters are identical, but Katz came to his conclusion without access to the cleaner copies (he called the CBS affiliate in LA with his conclusions, apparently based on copies of the document on the internet, unless he, too, has access to the originals). In any case, it sure would be nice if CBS provided cleaner copies (which they claim they have) so that independent experts could judge them.

Permalink to Comment

24. Lynxx Pherrett on September 14, 2004 11:27 PM writes...

Preview would be my friend, if it were available. Not "typeist," but "typist;" not "best way to do overcome," but "best way to overcome;" not "smae," but "same;" and not "nexe," but "next."

That's one other thing that makes the number use inconsistencies stand out -- I haven't seen any mispellings, transposed letters or other typographical errors in those documents.

Permalink to Comment

25. nick foresta on September 15, 2004 12:05 AM writes...

Mr. Katz is a certified expert, other than the go-ahead from the great and powerful Glenn Reynolds, what have you got that qualifies you as an expert? It would seem to me the standards for proof on the blogisphere change depending on whether the story helps or hurts the boy king.
Leaving aside the document story for a second, the facts are that Mr. Bush used his fathers's influence to avoid fighting in a war he claimed to support and did either the bare minimium or not even that to fulfill his duty to his beloved country. When he had the chance to show some courage, he ran the other way. Now, he runs around calling himself a "war president" as if he's some sort of Rambo. Those are the facts that matter most. Instead of expending so much time and energy researching fonts and typewriters and word processing tools, you guys should spend a few minutes researching words like "courage".

Permalink to Comment

26. satermo on September 15, 2004 12:12 AM writes...

Lynxx: review from your friends: even "mispellings" is misspelled.

Permalink to Comment

27. Branko Collin on September 15, 2004 12:51 AM writes...

"Mr. Katz is a certified expert"

Fallacy: Appeal to authority.

"what have you got [...]?"

His wit?

Permalink to Comment

28. politicaobscura on September 15, 2004 02:57 AM writes...


Just a reminder to folks to keep all of CBS' excuses straight and to NOT let the excuses cancel each other out.

For instance, IF CBS claims an IBM Composer was the typwriter used in 1973, then THEY have to explain why it has a 1 key!!!!!

CBS' own explanation are not consistent from night to night...!

http://www.ibmcomposer.org/SelComposer/brochure.htm

Permalink to Comment

29. Roundguy on September 15, 2004 03:08 AM writes...

The lady, now 86, that was Killian's secretary at the time claimed that she used an old Olympic typewriter. She also claims that what is written closely follows how Killian felt about Bush.

Permalink to Comment

30. Matt Brubeck on September 30, 2004 05:58 PM writes...

This analysis makes a very strong case that the memos were set not in Times New Roman, but in the (much older) "Typewriter" typeface, which was indeed used on military typewriters in the 1960s and 70s:

http://imrl.usu.edu/bush_memo_study/index.htm

The analysis also presents other evidence that the memos were printed using an impact typewriter.

Permalink to Comment

31. Ernest Miller on September 30, 2004 07:40 PM writes...

It makes a case, but it hardly addresses many of the issues raised by Joseph Newcomer in his analysis.

Strangely enough, despite having a 20-year archive or military documents and a repository of documents, he fails to provide any full body examples that resemble the memos.

How he is able to definitively exclude computer generated copies baffles me. Given the poor quality of the copies, to definitively claim that minor potential errors in certain keys is obvious is not so clear to me.

His other arguments seem to lack a certain rigor as well: "There is no good way for proving the documents in question are authentic. If I were in the Texas Air National Guard, and I said, “I saw the documents in Col. Killian’s cabinet,” who would believe me? The answer to that question depends entirely on the political point of view of my audience." There is plenty of evidence that would help to prove the documents were authentic, such as contemporary documents using the exact same font from a location readily available to LT COL Killian. Any typewriter of the era that could readily reproduce the documents would also lend much to the documents authenticity.

Despite supposedly being about the authenticity of the documents, the paper claims that CBS was justified in using them based on what CBS knew at the time. Not sure where the evidence or the logic for that comes from.

Permalink to Comment

32. Matt Brubeck on September 30, 2004 08:20 PM writes...

Yes, I agree that most of Hailey's analysis (especially the final conclusion) lacks support. However, I think the information about the "Typewriter" font family is useful. It's relevant to Newcomer's analysis because Newcomer's conclusions rely heavily on proving that the memos could be produced only using modern typefaces and typesetting.

If a 1970s-vintage typewriter font does match the memos as well as modern Times New Roman, then it invalidates much of Newcomer's analysis. To my eyes, the "Typewriter Condensed" font used in Hailey's examples is a better match than Times New Roman. (However, Hailey doesn't give sufficient evidence regarding which fonts were actually available in Vietnam-era typewriters.)

Hailey may well prove to be wrong, but I think this new information is worth considering.

Permalink to Comment

33. Ernest Miller on September 30, 2004 10:50 PM writes...

It is worth considering, however, there is much to be desired in a report that took a couple of weeks to put together. An awful lot of explanation of exactly what he did and how is missing.

Then there are the questions raised here:
http://wizbangblog.com/archives/003851.php

Permalink to Comment

34. Matt Brubeck on October 1, 2004 03:53 AM writes...

Thanks. The discussion at Wizbang is still ongoing, but the evidence emerging seems to indicate that Hailey did not in fact find a true matching typeface for the memos. He apparently manipulated character shapes (using characters from more than one font family) and spacing (using Photoshop to reposition individual letters) to produce his "exact" matches.

Permalink to Comment


EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
Kitchen Academy - Course II - Day 23
Kitchen Academy - Course II - Day 22
Kitchen Academy - Course II - Day 21
Kitchen Academy - The Hollywood Cookbook and Guest Chef Michael Montilla - March 18th
Kitchen Academy - Course II - Day 20
Kitchen Academy - Course II - Day 19
Kitchen Academy - Course II - Day 18
Salsa Verde