Here's my analysis of the PDF hosted on whitehouse.gov.This is truly an interesting mystery! I am hosting the images in this post on ImageShack, so apologies if they reach a bandwidth limit.
The PDF contains several layers, as Karl has pointed out. Using Adobe Illustrator, I hid all layers except for one that contains what appears to be a monochrome document scan.
http://img96.imageshack.us/img96/9420/monochromescan.png
Using Adobe Photoshop, I attempted to convert the image to indexed color. The indexed color converter tells me how many unique colors exist in the document. It reports 5 unique colors. This monochrome layer is not monochrome, but close.
A much more interesting feature about the layer is that some characters are completely missing. I've highlighted some of them with red below (there are several more that I did not highlight).
http://img31.imageshack.us/img31/4331/monochromescanhighlight.png
This scan raises very interesting questions:
- What was the original medium used to generate this image? Was it a paper photocopy (unlikely for reasons Karl mentioned; there would be far more colors and artifacting present in the layer)? Was it some pre-existing digital image, simply pasted into the PDF?
- Why are so many characters completely missing from the layer? Note that the missing characters are completely missing, not partially. The '1' in 10641, The 'R' in BARACK, the 'S' in STANLEY, etc.
This image appears to me to be a pre-existing digital scan of the original paper document, rather than a new, high-quality scan of the original paper document. The missing characters could be missing simply because the original digital scan did not detect enough blackness to consider a particular pixel as being black. Seeing as how the original paper document was probably composed with a typewriter, it's possible that the typist simply didn't strike the keys hard enough to cause a black-enough glyph to appear on the paper.
Imagine you are the person in charge of presenting Obama's long-form certificate of live birth to the public. You are presented with what is a seemingly incomplete scan of the original document. Several key pieces of information are missing, including very important dates and signatures. You could choose to release just this image, knowing it's authentic, or you could decide to spruce it up a bit by re-scanning the paper document using newer scanning technology and further adding in missing content by hand in an image editor. Enter another layer in the PDF.
http://img15.imageshack.us/img15/5913/scanz.png
This layer is very interesting indeed! Most of the text present in the monochrome layer has been partially erased. What's interesting to me is that the letters weren't completely erased. There may be some technique I'm not aware of for partially erasing ink from paper that would leave these markings. The big question is, how or why would this be done? This image appears to be a real scan of something. Why would the scan have so much content apparently erased? Theorizing, perhaps what happened was the ink somehow evaporated from the paper, leaving the white strike mark from the typewriter and ink traces behind. Why certain information remains clearly legible, like the signatures, scattered characters and some check boxes, remains a mystery however.
Now, let's lay the monochrome layer on top of the apparently scanned layer.
http://img829.imageshack.us/img829/6385/monochromeoverlay.png
The interesting thing to note here is that several pieces of information are missing from both the monochrome layer and the scanned layer. Note the missing 'Non' of None and the missing dates. Why this information is missing is a mystery.
The other layers contain the information necessary to fill in the remaining missing information.
http://img138.imageshack.us/img138/6997/missinginformation.png
This information was probably drawn by-hand in an image editor such as Photoshop, using the partially erased whitespace on the scanned layer as a guide. The tool used was probably the Photoshop pencil tool, which draws aliased pixels (non-blurry, hard-edged) by default. This would explain why this information looks different from the scanned layer.
Speculating: perhaps the reason Obama did not want these images released is because they were in such bad condition. If the paper document has degraded over time, leaving almost no legible information, it could raise questions about why the document looked this way. Perhaps the reason was as others have speculated: that Obama's mother was underaged at the time of Obama's inception -- perhaps statutory ****. I'll leave it to others to get closer to the truth on the matter of "why."
What troubles me is that Obama's crew would attempt to pass this off as the "real" certificate of live birth, when in fact it is apparently three things: an almost-monochrome scan, a scan with most information mostly missing and hand-edited content. I guess they thought they could get away with it. After analyzing this information, I believe the real certificate of live birth is the layer that contains the strange almost-erased ink.