I am fond of saying things along the lines of this from March of 2009:
A whole cottage industry has sprouted up finding bogus reasons not to accept a document that for every other Hawaiian citizen is proof of birth in Hawaii, and proof of US Citizenship.
Perhaps in recognition of the need to create jobs in the American economy, Barack Obama has spawned yet another cottage industry with his long-form birth certificate. Lucas Smith sent me an email about the latest such enterprise at Mario Apuzzo’s blog (somewhere I try to avoid if at all possible), which is a repackaging of an article from WorldNetDaily (somewhere else I try to avoid if at all possible), Web expert: Obama certificate falls short in authenticity.
Here is where it gets interesting. The WND argument against the long-form birth certificate (LFBC) deals with an objection related to one discussed here regarding the Lucas Smith certificate (POSFKBC): kerning. Only this time the kerning argument targets the typewritten part of the certificate, not the typeset part of the certificate. WND’s crank expert du jur is named Karl Denninger, whose typography resume includes being a Unix system administrator. (I swear I’ve seen this guy’s picture before.)
Denninger claims that the spacing in the typewritten part is not uniform. One would theoretically expect to see uniform letter spacing in a typewritten document (at least for any common typewriter we’d likely encounter), and we’d expect to see kerning (changes in the space between individual letters based on the particular letter combinations) in a computer word processing document. So here’s the example from the WND article, that purports to show one letter overlapping another:
In this image, one can clearly see that part of the “a” overlaps part of the “p”.
The first order of business is to determine whether this image is “real.” Since we don’t have a bitmap image of the LFBC, I made one by taking a screen shot of the PDF shown at 100% resolution. The first problem is that this image is zoomed in 1000% and at that resolution, much of what is shown is “made up” by the paint program.
In the version I made, you will note that the letters have gray edges where the WND image is solid black. The gray that my program creates makes it a little easier to see where the program is interpolating (making up based on adjacent image elements). When one zooms in this much, much of what one sees is interpolated. One should question, for example, why the letter “K” has extra black at the bottom that you wouldn’t expect to see in real text, typeset or typewritten. The top of the “a” is closed, and you know the original isn’t
I looked further at the White House LFBC, this time using one of the layers in the Adobe Acrobat PDF. (Yes, Virginia, there are layers). These are great because the text is separate from the background. You can look at this layer for yourself. My observation here is that the typewritten section is not very even, to the point of saying that it is “badly spaced.”
Next, we should ask what should a 1961 typewritten document look like. None of the Kapi’olani long forms I have seen are clear enough to be useful. The best I can do is to look at the 1963 Alan certificate, from another facility.
Here we see distinct variability in the horizontal spacing with the “a” in Hawaii touching the “H.” This is caused by the fact that a typewriter types each letter separately, advancing the typewriter carriage after each letter. If the type bar is not perfectly aligned, it will skew left or right or the letters will not line up vertically. So real typewritten text on an old manual typewriter is uneven, just like the LFBC.
The other problem with the WND story is that word processing programs using a typewriter font don’t do kerning, because these are mono-spaced fonts, which by design are evenly spaced.
Here is an example of a real mono-spaced font (Courier New – bold) in Microsoft Word, and you will notice no kerning overlap.
So this is just another WND-grade “expert” with a crank analysis.