Douglas Vogt, president of a company that sells scanners and scanner software, published a letter claiming to have proved that Barack Obama’s long form birth certificate is a forgery, and for that matter that Barack Obama is a criminal. I wrote about it in my article” “Expert” claims: birth certificate fake.
I was mildly surprised when Mr. Vogt appeared and left the following comment here on the blog:
Dear Kevin, You called me a liar without even reading my report and then told your readers that I was “just a salesman.” I have 11 years in the typesetting business and 18 years selling scanners, designing document imaging software and installing such systems in city and county offices here in the Northwest. All the scanner manufacturers and distributors know be very well. I was consulted by the Justice Department regarding the Kodak purchase of Bell & Howell two years ago. Your only experience seem to be working as a bureaucrat in vital records for the State of South Carolina so you should know that what Obama presented to the public was only a PDF of a Certificate of Live Birth and not a birth certificate. There was no paper copy with a seal presented to the US Public therefore none to examine by anyone! My expanded 22-page report is downloadable here: http://www.vectorpub.com/Obamas_Certificate_Forgery.html. I bet you cannot prove me wrong that the Obama COLB is a forgery. You also owe me an apology for unfairly calling me a liar and defaming my character and good name.
Our personal squabble is probably not of general interest, but since Mr. Vogt is the closest thing to an “expert” the birthers have, it’s worthwhile to look at what he says; however, I can prove Mr. Vogt “wrong” in his contention that the long form is a forgery quite simply: The State of Hawaii’s web site says: “On April 27, 2011 President Barack Obama posted a certified copy of his original Certificate of Live Birth.” QED. So there is no need for me to prove that Vogt is wrong, but I will explain why he is wrong.1
First, though, I do need to correct something in his comment above. He said: “There was no paper copy with a seal presented to the US Public therefore none to examine by anyone!” Readers here know that NBC White House Correspondent Savannah Guthrie personally viewed the paper document and even felt the raised seal and snapped a photo. This single fact derails much of what Vogt argues since he is operating under the assumption that the PDF was exported from the Department of Health’s document imaging system (presuming such a system exists for birth certificates).
He also alludes to a difference between a “Certificate of Live Birth” and a “birth certificate.” He is confused. Learn more.
I called him a liar because he said “The birth certificate would have the imprint of the baby’s footprint, weight, length and other information such as the religion.” Do you see any footprints on any of Donald Trump’s three birth certificates? Is religion on yours? Is religion on Mr. Vogt’s birth certificate? Perhaps he could scan it for us. Note that he’s changed his story with the latest version saying instead: “Some even have religion.” The word “liar” may have been a little too strong, and if he will accept it, I will substitute the word “crank.”
I was a graduate assistant with the Division of Information Systems Development at Clemson University developing health information systems and I subsequently worked for the South Carolina Department of Health and Environmental Control for 6 years. My title was “Systems Analyst” but what I did was analysis and software development. I wrote the birth certificate printing and accounting software used by the Greenville County Health Department in South Carolina. After that I spent 30 years in commercial software development, as a software developer and a manager. I was still writing commercial code at the time of my retirement in January, 2011. I wrote parts of and supervised all of the commercial software (including scanning and imaging) developed for state vital records systems in 9 states and 3 large counties.
I hold a Masters Degree in Mathematics from Clemson with an emphasis in computer science. I was a member of the of the Association for Computing Machinery for 40 years. I have presented papers at national conferences on data exchange standards and implementation of medical system interfaces. I was a consultant to the Centers for Disease Control developing a document on best practices for data quality and I was on the CDC-sponsored committee that developed technical standards for the exchange of vital statistics data between state vital records systems and electronic health records systems. I served for two years on the Fraud Prevention Committee of the National Association for Public Health Statistics and Information Systems (the national vital records association).
I am not a PDF expert, although I have coded some with the Adobe Acrobat API and have done imaging coding, including interfaces with scanners and digital cameras (TWAIN).
Mr. Vogt’s Credentials
I spent some time trying to assess Mr. Vogt’s credentials before I wrote my article, with little success. As someone who has hired and fired software developers for most of my career, I have learned how to read a resume, and to spot things that are missing. Reading Vogt’s “credentials” I really couldn’t tell if Mr. Vogt has ever written a line of code in his life. He takes credit for his company’s product TheRepository (a Unix-based document management system), but I do not know what his personal involvement is. As best I could tell from public information, TheRepository does have the ability to index text within a PDF file, which implied to me that someone associated with TheRepository might know how to read OCR text within a PDF. Vogt in his new document says that TheRepository does OCR using a tool set licensed from somebody else. I found nothing to tell me that Vogt has a working knowledge of PDF internals, has any experience with birth certificates, or is qualified on US constitutional law — all things he writes about with the same assertion of certainty.
As a commercial software developer for 30 years (not a bureaucrat as Vogt spuriously claims), I know that specialized functions, specifically graphics processing, OCR and PDF generation, are usually done through off-the-shelf components; it’s not cost effective to develop these in-house. I could write, for example, a program that converts BMP images to JPG images using off-the-shelf imaging components without knowing the first thing about the internal workings of either format. So there is nothing in the public record that I could find that provides any information as to whether Douglas Vogt knows anything about the internal workings of PDF creation software, and particularly that he knows the first thing about the software used by the White House to create the PDF. When I read his original report, I found technical details that I would have expected to see totally missing.
The biggest and most glaring problem in Vogt’s 22-page dissertation is that he fails to identify or address the actual software used by the White House to create the PDF (and in fact he doesn’t even recognize that the White House created the PDF), and without that how could Vogt pretend to draw conclusions about what ought and what ought not to be in a legitimate document? He can’t. To my knowledge, in all the gaggle of web pages and YouTube videos, no birther has yet to try to scan a birth certificate with this software to see what happens. Vogt doesn’t even mention the PDF version; even I know the significance of that.
Step by step analysis
What I expected was to read things that a document imaging expert should know, laced with demonstration of technical savvy on the internal workings of a PDF document. What I found was a fly-speck tour of the pixels in the image with assumptions worthy of Dr. Ron Polarik. In fact, Vogt is mainly just repeating the crank ideas of other birthers. Real experts on PDF’s have pronounced the Obama long form normal (Krawetz, Fox News, National Review) and I don’t need to re-plow that ground. What I’m going to do in the next section is to point out what any technically savvy person should be able to see for themselves.
I have irrefutably proven that the Certificate of Live Birth that President Obama presented to the world on April 27, 2011 is a fraudulently created document put together using the Adobe Photoshop or Illustrator programs.
That’s the first sentence from Vogt, and we have already seen that it’s wrong. The PDF was created by Mac OS X 10.6.7 PDFContext. Not an auspicious way to start. He admits that he doesn’t know which of the two programs created the document. If he can’t tell which one, how does he know it was either — and in fact it was neither. No Adobe program created the long form PDF.
I won’t reproduce this here, but the reader can consult Vogt’s original to see that he is not only commenting on the document, he also is commenting on vital records process, birth certificates, Hawaiian and Constitutional law. He makes statements outside even his claimed expertise. I daresay that Vogt has never been inside a vital records office, except as a customer. I have. Some of what he says about vital records processing is correct and some is not. However he fundamentally does not understand what a birth certificate is. The “hospital certificate” is a souvenir, lacking any official validity. To my knowledge, no one has ever published an example of a Hawaiian hospital certificate if indeed something like it even exists. It is wrong to say that the birth certificate is copied from the hospital certificate. In this article from 1955 on Hawaiian vital statistics process, there is nothing at all about this so-called “hospital certificate” in the birth certificate process.
Since we have an example [Nordyke] of a microfilmed Long Form then it is safe to say that all Certificates in that period were microfilmed and still available for inspection as well as the original paper copy. Also these Certificates were embossed with the County’s department seal (Figure 21) as well as signed.
This may be confusing to the reader. Birth certificates in state archives are not generally sealed. The seal is applied to certify a copy subsequently made from the document in the archive.
Next is a rather odd section on national standards for birth certificates. Vogt says:
The law also makes it clear that the Federal Government wanted the states to computerize their source documents, which also meant scanning them into a document imaging system. In Appendix C I have highlighted in red, the important sections of the law.
I do not know when the Hawaii Health Department acquired and implemented a document imaging system but it had to be within two years of the enactment of this law .
However, there is nothing in Appendix C that says anything about scanning and document imaging systems. It is true that many states have scanned old documents, but others with which I have personal experience have not. I know of no evidence that Hawaii has entered any old birth certificates into a document imaging system. Also, what possible reason would the State of Hawaii have for imaging their old birth certificates in response to a supposed 2004 regulation when they stopped issuing such certificates three years earlier?
One of the reasons I mentioned the Federal law is because when President Obama gave his news conference, presenting his alleged Certificate of Live Birth, he stated that they had to get special permission to get a copy of his Certificate. His statement appears to be blatantly wrong because these are public documents and he could have requested a copy himself at any time as well as his Secretary of Health and Human Services could have audited the County’s records and get a copy any time they wanted.
This is rather silly. Everybody knows that Hawaiian birth records are not “public records.” The exception granted by the Department of Health in Obama’s case is documented in the published correspondence between Obama’s attorney and the State of Hawaii. CNN reported that under normal procedures, only a non-certified copy of the long form is available. The idea that Obama could use a federal auditor to get him a certified copy of his birth certificate is ludicrous.
Keeping in mind that Vogt has not provided any reason to think that Hawaii scanned its birth certificates into an image management system, he says:
When a county or city institutes a document imaging system, it would start scanning their documents into the system. If the forms are all just type and lines, it would most likely be scanned in as binary images (black pixel or white pixel). If there are photos or other halftone graphics on the page, it may be scanned in as a grayscale which consists of 256 levels of gray going from 0 for white to 256 for black. If there is color on the documents, they may scan them in as color images, but color images are large and impractical to store lots of them on the server. The legitimate forms done by the Hawaii Board of Health were scanned in using both grayscale and binary modes. The reason I know this is because I found both methods used on the forms. You cannot have both on an original scanned image from an original.
Of course 0-256 is nonsense; it would be 0-255. He says “You cannot have both on an original scanned image from an original” but this just displays his ignorance of how PDF creation software works. PDF optimization software, such as Adobe Acrobat does exactly this kind of separation.
Obama’s paper long form was either scanned and printed or photocopied onto security paper and the result stamped and certified on April 25, 2011. I have yet to see any reason to think that it was printed from a document imaging system. The fact that the images show the curvature of the bound volume argues strongly against the image coming from a document imaging system, since if whole volumes had been scanned, they would have taken it apart. I’m sure Mr. Vogt knows that in high-volume scanning operations, sheet-fed scanners are used, which requires unbinding the documents. We’re talking about over half a million birth certificates for Hawaii.
I’m going to skip over a lot of what he says at this point because it all hinges on the long form image coming from a document imaging system, and there’s no reason to think that happened.
1. Curved and non-curved type
The next section talks about the curvature of letters. The left side of the long-form birth certificate looks like it was scanned from a book, and was not laid perfectly flat. What Vogt says is that the curvature of the type doesn’t match the curvature of the form, or put another way, some of the words in a vertical column are curved and some are not.
This sounds implausible on the face of it. First if the document were a forgery, why curve it at all? No other published Hawaiian birth certificate image is curved like that. Why make something unusual? Second, for some letters to be curved and others not, each set of letters would have to be individually curved, which is silly. If someone were going to make a fake, they would create the whole document square and then apply the curve to the section to be curved.
What Vogt shows is a highly-magnified section where he has drawn a red line to lead the eye to think the word is not curved. A single straight line through what should be both curved and non-curved text leads the viewer to ignore the small curvature in the part that should be curved. Further, one need spend but a moment looking at the document to see wide variation in the vertical placement of letters. The kind of precision necessary to make a determination of the curvature of a couple of letters is simply not there. If one looks at the “M” in “Male” by itself, it is obviously curved and not aligned with the red line Vogt drew under it.
When Vogt’s straight red line is replaced with a line curved identically to the line below the text, the illusion goes away:
2. There is a white haloing around all the type on the form.
One of the basic errors Vogt made, and I should have commented on it earlier, was to suggest that images in the PDF were in the TIFF format (a compressed format that exactly preserves the image). It isn’t. The color background layer is a highly compressed, and lossy JPEG file. That’s why the long form images have “halos” around the letters and Vogt’s sample images don’t.
3. The Obama Certificate is loaded with both binary and grayscale letters
This is normal. PDF document optimization software isolates the background image (what’s not black) and stores it as a lossy JPEG file. It then makes bitmaps of the black part. In a few cases what we consider black was not quite black enough and was considered part of the background. This is what Adobe calls “adaptive optimization” described as:
Adaptive Divides each page into black-and-white, grayscale, and color regions and chooses a representation that preserves appearance while highly compressing each type of content. The recommended scanning resolutions are 300 dots per inch (dpi) for grayscale and RGB input, or 600 dpi for black-and-white input.
Of course we aren’t dealing with Adobe software with the President’s certificate. Other PDF software creates color and grayscale layers in different ways.
4. The Sequential Number is a fraud.
Enough has been written here about that. See for example: Obama’s birth certificate number.
5. Two different colors and font sizes in Form box 22 and 20
This is just the separation of almost black in the background to bitmap for the foreground text. Also note that since these are zoomed in images beyond the resolution of the document, a lot of the pixels we’re seeing are made up interpolations. You can see the left-over bits on the background (and tell me with a straight face that a human would create something like this):
6. The official seal is not part of the Certificate of Live Birth.
I’m not sure what that title means. Vogt writes:
Figure 19 shows Obama’s seal on the COLB presented on April 27, 2011 is visible only because a color filter was used to see it, otherwise it completely disappears in the design of the security paper (Figure 20).
Good job on your figure 19, which brings out the seal in the PDF image.
It’s hard to see here, but the top of the seal is running through the top of blocks 20-21. The main problem with Vogt’s Figure 20 which is supposed to show the lack of distortion in the text due to the seal, is that it crops the main spot where text intercepts seal at the top of Blocks 20-21. I looked at my own state-sealed birth certificate under a magnifying glass. I didn’t see distortion, but I did see that the ink had flaked off in a couple of spots there the seal was impressed. When I look at the long form, I can see a couple of places where the ink might have flaked off too — but such things are somewhat subjective.
Vogt talks about “TXE” on the rubber stamp and thinks the stamp was applied too straight (like the Department of Health wouldn’t be careful on a document to go around the world). However, this is another of Vogt’s red line illusions. When you look at the his image with small grid lines, it’s obvious that the stamp is slightly raised on the right, not “perfect” as Vogt claims. For example compare the top line letter “R” in “CERTIFY” with the same letter in “OR.”
8. Multiple layers in the PDF file from the White House.
This is another one fully discussed before. I just refer folks to Krawetz’s page. However, Vogt does say a couple of interesting things:
The only rebuttal to the nine layers discovered in the PDF file released by the White House was a statement from a Canadian graphic artists from Quebec by the name of Jean-Claude Tremblay on April 29.
First of all the Obama PDF certificate was supposed to have come directly from the Health Departments office.
But this is silly. All the reports say that the Department of Health sent two paper copies with stamp and seal, not a PDF. One need only look at the PDF itself to see that the creation date was April 27, the day of the White House press conference (the PDF is time stamped after the press conference, by the way). That was two days after the Department of Health printed and certified it.
Vogt then says:
My qualifications on OCR programs are considerable. … The text file and matrix files would never be seen as separate layers and there is certainly no nine layers. The three files would be in a PDF “wrapper” and that’s all. All OCR programs work on the same principle.
Does Vogt not even know that there is no OCR data in the Obama long form PDF?
The short of it is that PDF optimization (such as that done by Adobe Acrobat) separates the foreground from the background saving the background in a lossy JPG format, and then it generates bitmaps for different foreground regions of the form in varying resolutions.
So now you know, courtesy of the Hawaii Department of Health that Douglas Vogt is wrong, and I hope I have shown you why.
For the benefit of the reader who hasn’t read Mr. Vogt original paper, I will add some comments about other things he says that are not related to the long form birth certificate.
Vogt makes a big deal about the affidavit of Tim Adams, an elections clerk in Hawaii. Adams says that some unnamed person at the Honolulu Elections Division told him that Obama had no long form birth certificate. Whether this conversation is true or not, the Honolulu Elections Division does not have access to birth records, and so Adams’ comment contains no valid information, and is in direct contradiction to what the Hawaii Department of Health has said in public since 2008.
Vogt then says that the 1961 newspaper announcements of Obama’s birth are not proof because the State of Hawaii registers out of state births. However, Hawaii only started that practice (law passed) in 1982, so Vogt is wrong. Vogt claims that his previous experience with tax law makes him qualified to comment on Hawaii statutes, and in particular that Barack Obama is guilty of forgery. So if he is telling the truth about his qualifications, then he must be lying about his comment on Hawaiian law. (Actually it’s his qualifications that are really impeached here.)
1This reminds me of an experience in high school. The algebra teacher presented a mathematical “proof” on the blackboard that 1 = 2. Of course, we didn’t try to prove that the answer was wrong, but why his reasoning was defective.