In retrospect, there have been various people hinting as to the nature of the ‘forger’ and pointed out how a Xerox Workcentre was a likely candidate. I have collected a variety of references that in retrospect should have resolved the question earlier. As I recounted in my Reality Check Radio blogtalk, this is the end result of the hard work of various people, starting with John Woodman’s excellent book, which was sadly enough misunderstood by the MCSO Cold Case Posse, even though it contained enough hints to have guided their research in a more constructive manner. The most influential poster to me personally has been gsgs, whose diligent and hard work looking at the finer details of the process, allowed us an insight into the finer details of the process, all pointing towards an algorithmic process. It was his/her work that finally encouraged me to look beyond the superficial data in the PDF and look at the raw data more closely, looking for hints that could help us identify the ‘forger’.
What I have learned is that relying on programs like Illustrator, hides the real and relevant details used by forensic examiners to identify the ‘creator’ of a particular document. Programs like pdf-parser have been instrumental in allowing me to look at the inner workings of the PDF, and the use of hex editors and the JPEG standard, as well as jpeg analysis programs have also been helpful in uncovering the tiny hint left behind.
Posted May 9, 2011, 12:23 PM by olePigeon
Oh for ****’s sake. That’s what happens when you OCR something. The differences in color are the characters the OCR engine either did or did not pick up. Most OCR (particularly Xerox and Adobe) software creates multiple layers when you save as either a TIFF or a PDF; 1 with the source image, 2 with the OCR glyphs, 3 selectable text. The boxes in the middle of nowhere or around the signature are image containers created by the OCR program when it thinks it found a picture.
That’s it. There’s nothing nefarious about this scan.
Here’re three documents I scanned in last year using a Xerox:
Index of /68k/FastPath
ALL of them exhibit identical behavior to the birth certificate PDF. You’ll notice that the Xerox software attempts to erase the original text on the source image, then replace it with its own glyphs on top. That is exactly what happened with the birth certificate. My hunch is that it was simply scanned into a Xerox WorkCentre machine and auto-OCRed.
Bright Horizon Press (August 13, 2011)
Is Barack Obama’s Birth Certificate a Fraud? (Amazon)
Posted March 1, 2012, at 6:28 p.m. By Whet Moser
justlw March 20, 2012 at 5:31 pm
I finally got a chance to play around a little with my Mac and a scanner, and have come to a shocking conclusion:
I have a really old, crummy scanner.
But based on my poking around, I would put money on the White House having a Xerox WorkCentre color copier/scanner with a “scan to email” feature, rather than anything attached directly to someone’s Macintosh.
This would fit the available data, including the MRC artifacts. There is no intrinsic MRC functionality that I can see in Mac OS Quartz, so it’s much more likely it came from the scanner with MRC compression already in place, and WorkCentre copier/scanners do come with MRC built in.
Obama Birth Book
Dr. Conspiracy says: August 5, 2012 at 2:27 am
So the Cold Case Posse said you tried about every PDF generator out there for Mac and PC. Did you actually try a Xerox Document Centre with MRC compression? Did you try OmniPage 16? PaperPort Professional 11?
Extensive analysis which was one of my inspirations. GSGS found many interesting explanations for the artifacts.
JPotter July 19, 2012 at 1:14 pm
June 5, 2013 at 1:18 pm