We were alerted today to the fact the Dr. David Hailey had produced what he claims to be the final version [PDF] of his analysis of the Bush memos (Archived here). This is of course old news, as the veracity of his work has been thoroughly debunked both here and elsewhere already.
Here’s a sample of the relevant previous pieces reviewing Hailey’s work:
[Ed Note: Much of this information was originally compiled by Paul. We left in unpublished. I have heavily rewritten the content to address the current state of the Hailey report, and created the font mappings, so it’s sort of a co-written post.]
Hailey is newsworthy, if for nothing else, his timing. The report from the two-man panel of Dick Thornburgh and Louis Boccardi, Jr. on the CBS Memo scandal is expect Friday, and there’s speculation as to whether the Hailey report is cited. For their sake we hope they don’t. In addition to the posts linked above, here’s a little more information we’d been holding back.
Since as early as 8 days after the original 60 Minutes II story (Sept. 16th) Dr. Hailey’s work was available for anyone to read. Eight days after the airing of the story (September 8, 2004) Hailey is seen promoting his research paper (that contain the same conclusions as he is making in his “final report”) at Take Back The Media (scroll down to see the article). This thoroughly discredits his assertion in the “final” version of the report that, “It took IMRL five weeks of careful examination to determine that the documents are typed.” Perhaps what he meant to say was, “it took a few days to reach our conclusion, and several weeks of putting lipstick on it until it looked presentable.”
It was not until Wizbang took a critical look at the work that confusion over what was in (and not in) the report set in.
One of the first points of confusion is that Hailey has changed the title and scope of the work since we first examined it. Originally the report claimed to be about the physical source of the memos CBS used in their story.
Now that title has been changed. Hailey now claims the report is “Toward identifying the font families used in the Bush memos” (emphasis ours).
Other points of confusion cropped up as Hailey changed the report (without version control) in response to the glaring holes we identified. When Paul first found the problems with the report, he called the university and the head of his department asked him to wrap up a case for misconduct and present it to them the following Monday (4 days away). Within minutes of that call Dr. Hailey changed the study to address the exact concerns he had mentioned.
When confronted with what on it’s surface appeared to be a possible case of academic dishonesty or misconduct, Dr. Hailey and the University claimed that a vicious mob of bloggers was attacking his academic freedom, rather than address the substantive issues with the raised about the work.
No one at Wizbang ever questioned Hailey’s right to pursue his research, nor to our knowledge did any other blogger. Hailey and the University are were building a convenient strawman to divert attention from the content of Hailey’s report. We did and do question the methods he has used to try to prove his hypothesis. It’s beyond comprehension that a university would not recognize that when you try to pawn off a cut and paste job as visual proof that a typewriter produced the Bush memos, someone is bound to call you on it. In this case it was a large number of bloggers and Dr. Newcomer, whose critique of the Hailey research is devastating, who performed the unwelcome peer review.
The funny thing about the final version of the report is that the item that originally sparked our critique – the pasted superscript “th” – does not appear in the final version of the report. The following (from our archives – not the final report) is what appears as Figure 26 on Page 26 of the final report.
Hailey claims, even without the critical superscript “th” that he “proves” the typewritter font family produced the documents. Even for Hailey this is an amazing leap of logic, since the primary assertion that the documents were NOT typed was that no one could reproduce the superscript “th” as the appeared in the Bush memos with an actual typewritter ball in use by the Texas Air National Guard at the time, or anywhere else for that matter.
About The Examples – A New Problem
There is another glaring problem with the report which we had not addressed until now. The main thrust of the study is that the font is not Times New Roman but belongs to a “genus” called “typewriter.” He claims expert knowledge of this typeface.
This graphic is from the bottom of page 11 of the original pdf and it continues on page 12. In the final version of the report (comically dated Dec. 3, 2005) the same information (reordered) appears at the end of page 13 and continues on page 14.
Click image for larger version
Where did the professor get these samples about which he makes the following claims?
“The above [sic] provides examples of variations in Typewriter, digitized. In some cases, the examples are taken directly from antique typewriters and digitized for use on computers.“
This search at Myfonts.com
This picture [view image] is a screenshot of the professor’s screen that he saved as he surfed myfonts.com in an apparent effort to find fonts which met his criteria. This picture is saved from one of the Photoshop files (4.3MB) that Hailey had left exposed on his web server. If you look at the Photoshop file and and examine the various layers in the document, you will see that he captured all 62 fonts.
Remember the professor’s description, where he said, “On the other hand, the cross strokes on the “t’s” change little. Typically, the right half of the cross stroke will be somewhat longer than the other.” It turns out that’s not exactly true. He cherry picked his 8 examples from 62 different fonts his search returned.
You won’t need to go to the trouble of looking through the Photoshop file to identify the fonts because we’ve done that for you below.
Careful visual inspection of the the Hailey example (above) and the myfonts.com result set indicate that the screen captures of the following fonts at myfonts.com were used in the report (the number in the second column is the position of the font in the result set):
1 – 43 – Writing Machine
2 – 57 – ITC American Typewriter
3 – 56 – Minya Nouvell
4 – 30 – Keystoned
5 – 60 – Prestige Elite
6 – 58 – FF Elementa
7 – 1 – Passport
8 – 34 – KfontZ
So what’s the big deal? None of these fonts are from typewriter balls and none were available for 1970’s era typewriters (with the possible exception of Prestige Elite). Hailey claims that ITC American Typewriter is the font, yet that font was not designed until 1974 by Joel Kaden and Tony Stan, and had no lineage to IBM (IBM is not one of the listed license holders for the font).
We have now identified yet another of Hailey’s exhibits that is not exactly what it purports to be. Since the report is final, we await an explaination or revision in the really, really final version.
Of course the best part, is that by identifying ITC American Typewriter as the font and failing to produce a superscript “th”, Hailey unintentionally proves what he set out to disprove – the memos were created on modern computer equipment.