Special Thanks to Hermitian

I would take this time to express my gratitude to Hermetian whose never ending questions helped uncover many more supporting data points that the document was created by a Xerox Work Centre 7655. We can even predict that the document was scanned upside down and subsequently reoriented using Preview.

I hope that Hermitian has learned from this experience as much as I have.

It’s time to write up my report announcing that we have found with a high likelihood the true forger of the WH PDF: A Xerox Work Centre 7655 scanner/printer/copier.

Good luck Hermitian and if you have any further questions or suggestions, please let me know. I always love a challenge.

63 thoughts on “Special Thanks to Hermitian

  1. Have you explained how the Xerox Work Centre can munipulate Stanley Ann Dunham’s signature to look the way it is. This is a thing of important concern. Reed Hayes is apparently a handwriting expert. Perhaps in his report, he may report how Stanley Ann Dunham’s signature (Dunham Obama part) appears to be in PERFECT precision, something that a person simply can’t do but a computer can if the signature is imported from another source.

  2. I am also curious. In your experiments, what base document was used to draw the conclusions that the Xerox Work Centre was involved? If I’m not mistaken, didn’t the CCP use an actual Hawaii Long Form Birth Certificate or similiar safty paper construction to baseline their tests?

  3. “Have you explained how the Xerox Work Centre can munipulate Stanley Ann Dunham’s signature to look the way it is.”

    Yes, it has been explained. Zoom in on the signatures. Notice how the parts that stay on the green background are touching one of the horizontal lines that form the border of box 18a. those parts of the signatures and also the letters from the box 118a label were treated as an image, separated to the image layer and compressed as an image, while the rest of the signatures were treated as text and compressed as text.

    “didn’t the CCP use an actual Hawaii Long Form Birth Certificate or similiar safty paper construction to baseline their tests?”

    But the didn’t use the right type of scanner software. That’s the key element they missed. But they have been informed about this new information and since they are interested in the truth, I’m sure the results of their new tests with a Xerox will be forthcoming.

  4. The signature always bothered me. Even on the high resolution JPEG, the “Dunham Obama” part of the signature in perfect precision. I can’t see how person can write signature that is that perfectly aligned.

    It is my understanding that the CCP did use safety paper in their tests and not regular paper to print the PDF file. This is a key element since safety paper has properties that make scanning of it different from that of regular paper. Did NBC use safety paper as well. In addition, the PDF is copy of a copy. Attempting to scan the PDF printout will pick up the anomolies in the PDF. The CCP was faced with this limitation as well but they used safety paper in their test which made a difference.

  5. NBC

    “I would take this time to express my gratitude to Hermetian whose never ending questions helped uncover many more supporting data points that the document was created by a Xerox Work Centre 7655. We can even predict that the document was scanned upside down and subsequently reoriented using Preview”

    I don’t want your thanks !

    In fact I don’t want to have anything to do with your analysis.

    I don’t associate my name with flawed research.

    Therefore I am formally requesting that you retract completely your above post.

    Moreover, I will continue to deny that I assisted you in your misguided pursuit of your Xerox Forger theory in any way.

  6. NBC

    Good luck Hermitian and if you have any further questions or suggestions, please let me know. I always love a challenge

    Yes I have a suggestion.

    Don’t write the report.

    You will just have to retract it.

  7. Hermitian,

    The signature of Stanley Ann Dunham has always bothered me. As I stated before, the “Dunham Obama” part of the signature looks to be in perfect precision. In other words, it looks artificial. Not only that, the print of Ann Dunham’s signature looks very thick and fresh and not worn. In fact if you look that birth certificate as a whole, both in the PDF and high resolution one, Stanley’s signature seems to jump out at you. The signature doesn’t look right. It is the opinion of the CCP that signature was most likely lifted from another source and placed into the document. It would explain the artificial look of the signature on the both PDF and AP image.

  8. It is the opinion of the CCP that signature was most likely lifted from another source and placed into the document. It would explain the artificial look of the signature on the both PDF and AP image.

    One problem though, there is no evidence that the document was somehow forged…

    Other than ‘it does not look right’ but that’s not a scientific argument.

    The CCP has a lot of explaining to do with their claims about the PDF…

  9. Attempting to scan the PDF printout will pick up the anomolies in the PDF.

    Yes, it might but what I showed is that the process creates multiple text layers. This is something the CCP had claimed could not happen.
    The CCP never used the right software/hardware so the use of safety paper was irrelevant.

    The CCP has to explain why they failed to look at the raw PDF data that contains many hints. For instance the JPEG comment and the quantization tables are also found in the Xerox Work Centre used by the White House to scan in the tax returns for POTUS.
    I’d say that the CCP has some work ahead of them to catch up and explain how their ‘experts’ missed all these tell tale signs?

  10. Don’t write the report.

    You will just have to retract it

    Speaking of retracting… Did you not submit an affidavit🙂
    If I have to retract it when you get to find some flaws, I will be delighted to do so.

  11. I don’t associate my name with flawed research.

    And yet you signed your affidavit?… Weird…

    As to my research being flawed, you still have failed to do so.

    While you do not want my thanks, I could not have achieved all this without your continuous support, asking questions, forcing me to think about new ways to test my hypotheses.

    I am looking forward to you eventually showing the flaws in my research, similar to how we managed to show the flaws in your affidavit.
    I love peer review…

  12. I am also curious. In your experiments, what base document was used to draw the conclusions that the Xerox Work Centre was involved? If I’m not mistaken, didn’t the CCP use an actual Hawaii Long Form Birth Certificate or similiar safty paper construction to baseline their tests?
    You can use whatever document, but until you use the Xerox WorkCentre, you will fail to find the multiple layers, the layer alignments, the embedded comment in the jpeg, the quantization matrices, the JBIG2 compression: Even a poor copy of the original PDF shows all these signs, but of course with some differences in what layers are separated.

  13. Reed Hayes is apparently a handwriting expert. Perhaps in his report, he may report how Stanley Ann Dunham’s signature (Dunham Obama part) appears to be in PERFECT precision, something that a person simply can’t do but a computer can if the signature is imported from another source.

    Somewhat begging the question, is it not… I doubt we will ever get to see the report. I hope he looked at the raw PDF to establish his findings. Maybe he can help understand how the tell tale signs were missed?

  14. NBC

    “I received documents captured on a Xerox 7655 WorkCentre which involved the normal
    workflow.

    I thought that the “normal workflow” was that a Xerox Workcenter 7535 was the forger ?

    NBC: That was before people suggested that I take a closer look at the 7655 documents from the Whitehouse again and found that indeed they match the jpeg comment as well as the Quantization matrices.

    Wasn’t the Xerox 7535 our guy yesterday ? Or at least I think it was yesterday ?

    NBC: Anyone familiar with the scientific method understands that the data guides the hypotheses, not the other way around.

    So is the Xerox 7655 today’s forger ? Did Preview drop out as a co-conspirator ?

    [NBC: Preview did not create most of the artifacts seen and of course anyone who looked at the document knew that preview was used as a final step. We now know why.]

    How many different models of the Workcenter does Xerox offer ? And then how come they don’t all use the same MRC package ?

    NBC: They come with different firmware version

    And how can both the Xerox 7535 and the Xerox 7655 be device independent with respect to the MAC OS Preview ?

    NBC: Because they are… It’s an observation…

    But then none of us really knows what NBC and his team of grunts have been doing in the background.

    NBC: Research, hypotheses testing and reporting on the results.

    One of his grunts must be creating all of these 20+ web pages all bearing my name in the headlines. Or is it the page headers ?

    “4, The Quantization Matrices do not match the WH LFBC, and the WH Tax matrices but this can be explained by realizing that the scanner is not using the up-to-date firmware as it scans to PDF 1.4 not PDF 1.5 as found in the WH Tax form PDF.”

    The form of the transformation operators of the Xerox created PDF is fundamentally different from the form of the transformation operators observed in the WH LFCOLB. And this observed difference in form has absolutely nothing to do with the Xerox firmware or the PDF version number. The PDF version of the WH LFCOLB is version 1.3.

    NBC: Well yes, 1.3 is what preview turns it in to. We now know that the firmware version used in the WH Xerox 7655 scanner matches the quantization matrices. So not only have we tracked down the likely forger but we have identified that the WH indeed owns such a machine which matches all the hidden markers.

    However, I can make this statement with confidence only for the Xerox Workcenter 7535 created PDF. I have no PDF files which were created by the Xerox Workcenter 7655.

    NBC: Is that not too bad… Surely a person as inventive as you should be able to get access to the online White House potus tax PDF? Come on Hermitian…

    So what does the Obama tax form PDF which was purportedly produced with version 1.5 have to do with the WH FCOLB PDF which, according to the file METADATA was produced by version 1.3 ?

    NBC: Sigh… They show the same embedded data, and of course, once it is run through preview to correct for the orientation, it was saved in 1.3 format

    xmlns:pdf=”http://ns.adobe.com/pdf/1.3/”
    xmlns:dc=”http://purl.org/dc/elements/1.1/”
    xmlns:pdfx=”http://ns.adobe.com/pdfx/1.3/”>
    2011-04-27T12:09:24Z
    2012-10-02T13:56:18-04:00
    Preview
    Mac OS X 10.6.7 Quartz PDFContext

    The form of the transformation operators observed in the WH LFCOLB PDF file is consistent with the file having been created by Adobe Illustrator.

    Not really, since we all know that the transformation operators have been rewritten by Preview. Of course, we do know that none of the Illustrator markers have been found in the document but we do know that the Xerox WorkCentre markers have been found.

    I hope that Hermitian has learned from this experience as much as I have.

    You can stop hoping Dude !

    I’m way ahead of you NBC !

    And you Dude are on a fool’s errand !

    And please don’t respond in Red letters !

    NBC: ROTFL, I respond in red letters to outline any changes I made. Perhaps I am on a fool’s errand but I still hope that you learned something. I surely did.

  15. Even if the Xerox Workcenter was involved, the birth certificate is still very much of forgery. There still many things that can’t be explained with the Xerox Technology:
    African Race Problem
    Birth Certificate Sequence Problem
    Obama Sr. Age Problem
    Perfect Dunham Obama Signature Problem

  16. Hermitian,

    Mike Zullo has admitted that several Obots have been able to duplicate SOME of the anomolies found in Obama’s Birth Certificate. The problem is that Obots have unable to produce ALL of the anomolies in Obama’s birth certificate and in a manner or behavior consitent with the CCP’s position that BC is forged

  17. Hermitian,
    It is my understanding that in one interview, Mike Zullo did reveal that Adobe Photoshop was involved in producing the BC. Although much of the metadata was gone, Zullo reported that trace metadata evidence still existed that pointed to the used of Adobe Photoshop.

  18. “So what does the Obama tax form PDF which was purportedly produced with version 1.5 have to do with the WH FCOLB PDF which, according to the file METADATA was produced by version 1.3 ?”

    Still not understanding are you?

  19. It is my understanding that in one interview, Mike Zullo did reveal that Adobe Photoshop was involved in producing the BC. Although much of the metadata was gone, Zullo reported that trace metadata evidence still existed that pointed to the used of Adobe Photoshop.

    There is NO evidence of any involvement of Adobe Photoshop. Perhaps Zullo could explain why a ‘forger’ would create the document from rotated images?… Trace evidence all points to a Xerox WorkCentre as I have shown.

    How could they have missed all this?…

  20. Even if the Xerox Workcenter was involved, the birth certificate is still very much of forgery. There still many things that can’t be explained with the Xerox Technology:
    African Race Problem
    Birth Certificate Sequence Problem
    Obama Sr. Age Problem
    Perfect Dunham Obama Signature Problem

    True, African race has been explained, sequence has been explained, we know that Obama Sr lied about his age, and there is no ‘perfect signature’ just one signed and (corrected) to include Stanley.

    But I am glad you too have come to reject that the PDF itself shows evidence of a forgery, now we are left with some minor issues, none of which indicate a forgery.

    I love it when birthers backpeddle…

  21. Still not understanding are you?

    I thought that the workflow was quite clear… After all the Preview metadata is written all over the document and I am sure that Hermitian knows that it saves in PDF 1.3 format… Sigh…

    Am I that bad at explaining this? It’s so straightforward:

    1. Document was scanned on a Xerox WorkCentre 7655
    2. Document was emailed
    3. Document was opened in preview and found to be upside down
    4. Document was rotated and saved in preview

    This explains all the important artifacts including the presence of the JPEG comment and the Quantization matrices, it explains how JBIG2 was used and yet it was encoded in FlateDecode, it even explains the margins added by preview. Oh and the monochrome bitmaps also show indication of the scanning process and their alignments again match the expectations and predictions.

  22. NBC: “It’s time to write up my report announcing that we have found with a high likelihood the true forger of the WH PDF: A Xerox Work Centre 7655 scanner/printer/copier.”

    Re: “…we have found with a high likelihood…”

    In other words, “high likelihood” is your theory. Lot of theories floating around on a good number of subjects.

    From the dictionary:
    theory (ˈθiəri) – plural ˈtheories – noun
    1. an idea or explanation which has not yet been proved to be correct.

    One of my favorites that I’ve followed over the years with alot of interest is the theories presented by paleontologists studying the fossilized remains of dinosaurs. The study has been going on now for about 200 years. In the beginning these scientists developed theories from their early collections on what the different kinds of dinosaurs looked like, its habits, and what it ate. Every so many years or decades as more remains were being found these scientists would revise with updates their theories on looks, habits, and diets. But it’s still theory 200 years later. Unless by a quirk of luck the probability of ever discovering a preserved specimen is almost zero.

    With Obama’s records we don’t need to scheme or deal in theories, speculation, and quesswork. The original records are there for examination by qualified document experts. All that’s needed is for Obama and/or Republican leadership to grow a set.

    As always,
    Rambo Ike {riding roughshod over Obot ignorance & deceit]

  23. !! THE FOLLOWING FINDINGS ARE PRELIMINARY !!

    My analysis of the transformation operators is still ongoing. However I have decided to publish some initial findings as an encouragement to those readers who still have an open mind regarding the source of the WH LFCOLB PDF image.

    I have determined that the form of the transformation operators of the Xerox created PDF is fundamentally different from the form of the transformation operators observed in the WH LFCOLB.

    An initial conclusion is that this observed difference in form has absolutely nothing to do with the Xerox firmware or the PDF version number. The PDF version of the WH LFCOLB is version 1.3.

    However, I can make this statement with confidence only for the Xerox Workcenter 7535 created PDF. I have no PDF files which were created by the Xerox Workcenter 7655.

    NBC hasn’t shared the 7655 files with anybody.

    Of course I continue to analyze the original WH LFCOLB which I personally downloaded from the Internet Archive’s WayBack Machine. And anyone can access and download this same file
    from here:

    http://web.archive.org/web/20110427171111/http://www.whitehouse.gov/sites/default/files/rss_viewer/birth-certificate-long-form.pdf

    The transformation matrices for successive simple transformations are concatenated by the appropriate matrix multiplication in the order specified in the PDFReferencXX.pdf.

    From PDFReference14.pdf:


    If several transformations are combined, the order in which they are applied is
    significant. For example, first scaling and then translating the x axis is not the
    same as first translating and then scaling it. In general, to obtain the expected
    results, transformations should be done in the following order:

    1. Translate
    2. Rotate
    3. Scale or skew

    “Figure 4.6 shows the effect of the order in which transformations are applied. The
    figure shows two sequences of transformations applied to a coordinate system.
    After each successive transformation, an outline of the letter n is drawn.

    “FIGURE 4.6 Effect of transformation order

    “The transformation shown in the figure are as follows:

    • A translation of 10 units in the x direction and 20 units in the y direction
    • A rotation of 30 degrees
    • A scaling by a factor of 3 in the x direction

    In the figure, the axes are shown with a dash pattern having a 2-unit dash and a
    2-unit gap. In addition, the original (untransformed) axes are shown in a lighter
    color for reference. Notice that the scale-rotate-translate ordering results in a
    distortion of the coordinate system, leaving the x and y axes no longer perpendic-
    ular, while the recommended translate-rotate-scale ordering does not.”

    The transformation operator for each object within the PDF file is a row vector containing six real numbers. Thus each six-vector assumes the following generic form:

    [a b c d e f ]

    Each image object within the same PDF composite image has an associated transformation six-“vector. The six real numbers that comprise each six-vector are generally different for the different objects.

    The following six-vector was extracted from the archived copy of the file “birth-certificate-long-form.pdf” by means of Adobe Acrobat XI Pro Preflight.

    0 -792.96002 612.47998 0 -.24 792.47998 cm

    Thus the form of this six-vector is:

    [0 b c 0 e f]

    And we find that the numbers a and d are both zero. All of the nine objects that comprise the WH LFCOLB have different six-vectors which all have this same form (a = d = 0).

    This form for the six-vector is consistent with a rotation of -90 degrees (clockwise) provided that the skew angles alpha and beta are both zero. A rotation of 90 degrees clockwise for each of the nine objects is consistent with the rotation angles read out from the Link panel data in Adobe Illustrator from the same PDF file. The rotation angle was 90 degrees clockwise for each of the nine image layers.

    Additionally, the non-zero values for the real numbers b and c are consistent with a scaling Sx and Sy of each of the nine image layers. The negative sign for b and the positive sign for c are also consistent with this scaling.

    Finally, the non-zero values for the real numbers e and f are consistent with non-zero translations tx and ty of the origin of the device coordinate frame with respect to the user frame.

    All of these findings are consistent with the PDF file having been created with Adobe Illustrator.

    Adobe Acrobat XI Pro was also applied to the file wh-lfbc-scanned-xerox-7535-wc.pdf. The composite image opened in the Portrait orientation. The same PDF file was also opened in Adobe Illustrator CS6 and Adobe Illustrator CC. However, the image opened in the Landscape orientation in both versions of Illustrator. Additionally, it was not possible to read out the Link data to obtain rotation angles because this file was not created by Illustrator.

    All of the findings reported here were obtained by means of Adobe Acrobat XI Pro Preflight. The extracted transformation six-vector for the Xerox created background image layer is:

    798.71997 0 0 614.40002 -3.36 -1.2 cm

    Contrary to the form of the corresponding six-vector of the background image of the WH LFCOLB PDF image, the form of the Xerox six-vector is:

    [a 0 0 d e f]

    Hence unlike the WH LFCOLB six-vector, where a and d were zero, instead we find in this case that b and c are both zero. All six-vectors of all seventeen objects in the Xerox file have this same form.

    If we assume that the skew angles alpha and beta are both zero then the values of these two constants are given by:

    b = Sy * sin (theta)

    and

    c = – Sx * sin (theta)

    This result is problematic because the scale factors Sx and Sy are positive real numbers and both must be greater than zero. The range for each scale factor 0 < Sx or Sy < or = 1.0. Consequently, the only way for b and c to be equal to zero is if sin(theta) =0. Hence, theta (the rotation angle) must be an integer multiple of + or – 180 degrees. Thus theta = 0, +/-180, +/-360,…

    For example theta = 180 degrees is one solution. However theta = 0 degrees is another.

    The 180 degree rotation would result in an “down-to-up flip” of the image. However, because this rotation has to be recorded within the final PDF code, the flip would require human intervention at some step during the creation of the final PDF file. Consequently this 180 degree rotation of the image would have to be manually applied within a graphics software program such as Illustrator, Acrobat, or Inkscape. Any one of these programs would record the application by a human of a 180 degree rotation of the image and then write the appropriate code for the rotation to the final PDF file.

    A physical rotation of the paper original on the glass of the Xerox could not cause a rotation of the image within the PDF code. Instead it would determine whether the image is initially up or down when the Xerox scan to PDF file is opened by the graphics software program. However the initial orientation would also be the final orientation unless a human rotates the image within the graphics program after the file is opened and before the file is closed.

    Because there are an infinite number of possible integer multiples of 180 degrees all of which would cause b and c each to be zero, then we have to look elsewhere to reach a unique solution. Two possible solutions are theta = 0 degrees and theta = 180 degrees. Again assuming the skew angles are both zero then the values of a and d are as follows:

    a = Sx*cos (theta)

    d = Sy * cos (theta)

    Checking the signs of a and d we find that these numbers are both positive in all seventeen transformation operators. Hence because Sx and Sy are always positive, we see that the only solution is if cos(theta) = +1. Consequently theta must be an integer multiple of 360 degrees. Otherwise a and d would be negative. Thus theta = 0, +/-360, +/-720, …

    One of these possible solutions of course is theta = zero. This value is the case of no rotation. Next considering the other possible solutions, we see that the effective rotation is also zero for each integer multiple of 360 degrees because the final position is the same as the initial position.

    Finally, we see that the only solution that yields zero values for b and c and positive values for a and d is effectively zero degrees.

    Thus we must conclude that the transformation operators (i.e. six-vectors) of the Xerox created PDF demand a zero rotation angle for all seventeen image layers.

    This finding is contrary to the findings for the WH LFCOLB PDF file as determined by means of Adobe Illustrator and Adobe Acrobat.

    Another significant finding is that the Xerox PDF does not contain any code lines which explicitly change the default unit of distance from pts (i.e. 1/72 in.) to 1/150 in. and then to 1/300 in. The PDFReferenceXX.pdf requires that the default distance unit be one point (pt) unless the default unit is explicitly changed within the code. I have searched all of the text of the Xerox PDF file and find no such lines of code. Consequently, the only place that the pixel sizes are found is within the embedded bitmap images of each image layer.

    This turns out to be true also for the WH LFCOLB however, this is a non problem for the human forger scenario. That is because the human controls the placement of each image layer within the WH LFCOLB PDF file and thus the human specifies both the reduction scale factors and the rotation angles which produce the final composite image. These parameters are recorded as the human applies each one. After all of the image layers have been placed in their final resting positions, the forger then embeds all nine images. He then saves the PDF to create the WH LFCOLB PDF file. After saving the Illustrator created PDF, the forger then opened the PDF in MAC OS Preview and re-saved the file to PDF to erase all previous METADATA.

    The final embedded nine images are consistent with a nominal page size of 8.5 in. X 11.0 in. (i.e. Letter size). The forger had previously created all nine images on the screen of his MAC OS computer within his computer graphics program. All nine images were created at a screen resolution of 72 PPI X 72 PPI which is the actual pixel resolution of the original Macintosh computer. The size of each image, as created, was larger than the final size of the image within the WH LFCOLB image. I reported the two sizes for the background image and for the mostly text image in a recent post.

    All of these findings are consistent with a forger using Adobe Illustrator to create the WH LFCOLB.

  24. In other words, “high likelihood” is your theory. Lot of theories floating around on a good number of subjects.

    It’s as much a theory as the claims by the Cold Case Posse.
    Since I do not have access to the original documents, I have laid to rest claims by the CCP that the PDF is a forgery.

    Of course, any theory is open to revisions, that makes it so useful. Even qualified experts will present a theory.

    Nothing too surprising here.

    But unitl the unlikely event that this ever happens, we have to rely on imperfect data. However, the data has allowed me to lay to rest certain speculations by those who claim that the PDF shows evidence of forgery.

    Anyone should welcome an update to our knowledge.

  25. It’s a good attempt but poor Hermitian does not understand what happens when the Xerox PDF is run through preview and thus his conclusions fail to properly take these changes into consideration.

    Once Preview has done its job, the transformation matrices all look the same as the one found in the WH PDF.

    Pfew…

    I have no idea why he is comparing apples and oranges here… But only he may know.

    A fundamental flaw that destroys most of his findings as irrelevant but I will comment regardless and may even upgrade his comment to a blog posting for educational value why Garbage In Garbage Out is an accurate description.

    I am happy to hear though that Hermitian has been familiarizing himself more with the PDF standard. It’s still a bit rough at the edges but it’s a good start.

    Of course, Hermitian has failed to explain why a forger would make it look exactly as if it had been created by a Xerox work centre…

    Oops.. I am sure our friend is familiar with “occam razor”, especially since he has failed to explain all the tell tale signs that point to Xerox as the culprit.

    Sigh… So far his findings merely reproduce what I have already shown.

    A bit late to the party… Still, progress I guess.

  26. John | July 26, 2013 at 15:31

    “Hermitian,


    The signature of Stanley Ann Dunham has always bothered me. As I stated before, the “Dunham Obama” part of the signature looks to be in perfect precision. In other words, it looks artificial. Not only that, the print of Ann Dunham’s signature looks very thick and fresh and not worn. In fact if you look that birth certificate as a whole, both in the PDF and high resolution one, Stanley’s signature seems to jump out at you. The signature doesn’t look right. It is the opinion of the CCP that signature was most likely lifted from another source and placed into the document. It would explain the artificial look of the signature on the both PDF and AP image.”

    You are exactly right John ! The whole deal is a digital cut and paste job.

  27. You are exactly right John ! The whole deal is a digital cut and paste job.

    And no evidence has been presented to support this. Is that the best you have to offer my friend? Just because it looks ‘strange’…

    Hilarious…

  28. “NBC | July 26, 2013 at 15:51


    It is the opinion of the CCP that signature was most likely lifted from another source and placed into the document. It would explain the artificial look of the signature on the both PDF and AP image.

    One problem though, there is no evidence that the document was somehow forged…

    Other than ‘it does not look right’ but that’s not a scientific argument.

    The CCP has a lot of explaining to do with their claims about the PDF…

    Does anyone else see anything wrong with this comment ?

    I thought that the whole field of forensic analysis of signatures was for the expert to reach his conclusion that “it does not look right”.

  29. Does anyone else see anything wrong with this comment ?

    I thought that the whole field of forensic analysis of signatures was for the expert to reach his conclusion that “it does not look right”.

    Nope, it doesn’t look right is a subjective interpretation. Wow… You need to be able to point to specifics, not generalized claims.

  30. NBC | July 26, 2013 at 15:55


    Attempting to scan the PDF printout will pick up the anomolies in the PDF.

    Yes, it might but what I showed is that the process creates multiple text layers. This is something the CCP had claimed could not happen.
    The CCP never used the right software/hardware so the use of safety paper was irrelevant.

    The CCP has to explain why they failed to look at the raw PDF data that contains many hints. For instance the JPEG comment and the quantization tables are also found in the Xerox Work Centre used by the White House to scan in the tax returns for POTUS.
    I’d say that the CCP has some work ahead of them to catch up and explain how their ‘experts’ missed all these tell tale signs?

    Actually, NBC doesn’t have a clue about what the Posse has done or hasn’t done.

    And then there’s the fact that there’s tons of software out their that carves up a scanned image to reduce the file size. I’m sure the CCP did trials on dozens of them.

    We are all waiting for NBC to get over his fetish and love affair with MRC compression.

    And then there’s the problem that there is absolutely no evidence that JBIG2Decode filter ever touched the WH LFCOLB PDF.

    Then there’s the problem that the YCbCr color space is the mandatory standard for all JFIF JPEGS.

    And it would have been sooooo… easy for the forger to edit the PDF to plant that tidbit — just like a squirrel burying his acorn to dig up later.

  31. A physical rotation of the paper original on the glass of the Xerox could not cause a rotation of the image within the PDF code.

    It causes a rotation in the captured images which is why they need to be rotated 90 degrees clockwise or counter clockwise, depending on how they were scanned.

    This is basic stuff but I may have to take smaller steps.

    1. Observe that the document was scanned in landscape orientation… Once you understand this then you have 90% of the solution. Think about the scanline of the scanner… How does it move… And how is the image captured…
    I also deduce from your comments that you have still not resolved my little puzzle… THink about the orientation of the document and its effect on how the image is scanned…

  32. NBC

    “”Don’t write the report.

    “”You will just have to retract it””

    “Speaking of retracting… Did you not submit an affidavit
    If I have to retract it when you get to find some flaws, I will be delighted to do so”.

    I sure did submit an affidavit and I posted it so anyone could read it. I believe in standing behind my work.

    Where did you submit your affidavit ?

    I’ll remind you that you said you would be delighted “to do so”.

  33. Hermitian: Actually, NBC doesn’t have a clue about what the Posse has done or hasn’t done.

    Oh but I do know what they have done as they reported on it. If they had only looked a bit deeper and they could have reported on all these tell tale signatures..

    Hermitian: And then there’s the fact that there’s tons of software out their that carves up a scanned image to reduce the file size. I’m sure the CCP did trials on dozens of them.

    Yes, and they missed the one that was used…

    Hermitian: We are all waiting for NBC to get over his fetish and love affair with MRC compression.

    It explains everything… No wonder I love how our predictions were found to match the data…

    Hermitian: And then there’s the problem that there is absolutely no evidence that JBIG2Decode filter ever touched the WH LFCOLB PDF.

    There is, through inference. Again, a simple work flow would create the duplicate letters found in the WH PDF.

    Hermitian: Then there’s the problem that the YCbCr color space is the mandatory standard for all JFIF JPEGS.

    That is not a problem. What your problem is is that these JPEGs, unlike others, include a COMMENT which is totally useless and yet is found consistently in Xerox generated PDFs

    Hermitian: And it would have been sooooo… easy for the forger to edit the PDF to plant that tidbit — just like a squirrel burying his acorn to dig up later.

    Sure, a forger can explain anything which is why you have to compare it to another hypothesis. Ad hoc explanations are just not very impressive. We know that the embedded string shows up in a Xerox workflow and that the Xerox workflow is consistent with all the other ‘artifacts’.

    Which makes it the best explanation.

    Basic science.

  34. NBC

    “While you do not want my thanks, I could not have achieved all this without your continuous support, asking questions, forcing me to think about new ways to test my hypotheses.”

    Right Dude ! You didn’t need my help to create your Xerox Turkey.

    [NBC: True but you helped me formulate it in simpler steps which allowed me to consider additional ways of testing.]

    Better get back to testing …

    [NBC: A good scientist never stops…]

    That will avoid your having to scrap hundreds of pages of Bull.

    [NBC: ROTFL… you are so much fun… and entertaining. You are my inspiration, my muse so to speak…]

  35. I sure did submit an affidavit and I posted it so anyone could read it. I believe in standing behind my work.

    Where did you submit your affidavit ?

    I’ll remind you that you said you would be delighted “to do so”.

    So do I, which is why I have shown all my work as well. Of course, I have done more. I have shown your interpretations to be flawed and how simple work flow explains much of it.

    I have no reason to submit an affidavit? Why would I? Like you I am not an expert, and like you, my affidavit will never be used in a court case. Sure, it may be filed as ‘evidence’ but it will fail to be taken into consideration by the courts.

    This is not about courts, this is just about finding out where the evidence leads us.

    And it leads to a simple conclusion: The forger was a Xerox work center.

    And since you are upset that I do not give enough credit to Preview, I will add that Preview added its own telltale signs when it rotated the image 180 degrees and saved in in PDF 1.3 format.

  36. THe content free responses indicates that you know I am right…

    And you are still unable to generate a Preview document… After all this time…

    In the mean time, I managed to find the hidden comment and matrix quantization tables and showed how they align with the PDF created by the Whitehouse for Obama’s tax returns.

    Again, everything lines up so nicely…

    Unlike the ad hoc: A forger could have done this as well…

    Funny…

  37. Oh, and I showed why your work on the affine transforms was a bit of a waste of time, other than of course providing you with an opportunity to work through them yourself.

    I am impressed… You managed to do what you claimed before was impossible because of some ‘conspiracy’ involving Adobe…

  38. NBC | July 26, 2013 at 16:08


    I am also curious. In your experiments, what base document was used to draw the conclusions that the Xerox Work Centre was involved? If I’m not mistaken, didn’t the CCP use an actual Hawaii Long Form Birth Certificate or similiar safty paper construction to baseline their tests?
    You can use whatever document, but until you use the Xerox WorkCentre, you will fail to find the multiple layers, the layer alignments, the embedded comment in the jpeg, the quantization matrices, the JBIG2 compression: Even a poor copy of the original PDF shows all these signs, but of course with some differences in what layers are separated.

    Yup ! The forger did all this with his old MAC OS running some old version of Adobe Ilustrator to cut and paste all of his images together at a common screen resolution of 72 PPI x 72 PPI.

    [NBC: THat’s it? No supporting evidence, just some random comment? How come the background has half the resolution of the foreground?]

    He set his grid to eight pixels for the major grid line and one pixel for the minor line spacing. He then simply used his mouse and cursor with the rectangle selection tool and snap to grid and/or snap to point to set the rectangle boundaries coincident with either a major or a minor grid line.

    It really isn’t complicated…

    But it does seem to be hard for NBC to grasp this reality.

    Too bad his numbers don’t work on that Xerox thingy.

    [NBC: Too bad that the Xerox workflow explains everything that was argued to be evidence of a forgery of the PDF.]

  39. Yup ! The forger did all this with his old MAC OS running some old version of Adobe Ilustrator to cut and paste all of his images together at a common screen resolution of 72 PPI x 72 PPI.

    THat’s it? No supporting evidence, just some random comment? How come the background has half the resolution of the foreground?
    Wow… and you call that forensic examination? Or what exactly?

  40. NBC


    This is basic stuff but I may have to take smaller steps.

    1. Observe that the document was scanned in landscape orientation… Once you understand this then you have 90% of the solution. Think about the scanline of the scanner… How does it move… And how is the image captured…

    Yes I suggest that you start over and do it right the second time. But then I have made this same suggestion several times in the past and….

    I don’t know about your Xerox family but I have a Canon MFP that will run circles around your Xerox.

    You probably don’t know but Canon made all of the print engines for HP for many years.

    The diode bar on my Canon scans from the top of the page to the bottom at the speed of light.

  41. NBC

    “I also deduce from your comments that you have still not resolved my little puzzle… THink about the orientation of the document and its effect on how the image is scanned…”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    Actually, NBC I could care less about how that paper copy is spinning around in your mind !

  42. I don’t know about your Xerox family but I have a Canon MFP that will run circles around your Xerox.

    But you should be looking at the Xerox which is after all the candidate identified. Apples and oranges again

    You probably don’t know but Canon made all of the print engines for HP for many years.

    Nothing relevant here…

    The diode bar on my Canon scans from the top of the page to the bottom at the speed of light.

    More details please… But also imagine how the paper is fed into the Xerox work center and how the scan is created. Hint: It starts on the left hand side and works is way towards the right hand side. Now you may understand why the images are stored rotated? Or do you need even more visual an explanation?

  43. Actually, NBC I could care less about how that paper copy is spinning around in your mind !

    Fair enough, I cannot expect you to spend the time to figure out novel ways to test my hypothesis.

    I will provide a write up on this one, it’s quite a nice one…

    But first let’s help you figure out why the images are stored in a rotated fashion…

    Various people have already provided you with a clue…

  44. NBC

    “”Hermitian: Actually, NBC doesn’t have a clue about what the Posse has done or hasn’t done.””

    “Oh but I do know what they have done as they reported on it. If they had only looked a bit deeper and they could have reported on all these tell tale signatures..”

    Sure the Posse publicly released all of their hard evidence of criminal behavior.

  45. Sure the Posse publicly released all of their hard evidence of criminal behavior.

    I doubt that they have more… They already know that this will never result in criminal charges especially now that the forger has been identified to be a Xerox work center.

    If they had looked deeper why would they present a report that is probably totally off base?

    Weird…
    While I cannot prove a negative until they come forward, if ever, I doubt that they have anything else that adds up to anything… Otherwise we would have seen criminal filings by now and not the courting of some ‘congress people’ in the hope they will pick it up.

    I can only hope that some foolish person ‘runs with it’… But unlikely, the GoP is not that dumb

  46. NBC

    “”Then there is the problem that the YCbCr color space is the mandatory standard for all JFIF JPEGS.””

    “That is not a problem. What your problem is is that these JPEGs, unlike others, include a COMMENT which is totally useless and yet is found consistently in Xerox generated PDFs”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    Yup ! It figures that Xerox would plant a useless comment in all their JPEGS.

    I wonder why Preview would leave it in there ?

    Now let me see ! MAC OS Preview was the last machine to touch the PDF.

    So wouldn’t MAC OS be the forger rather than Xerox ?

    We may need IBM Blue to figure out all of this machine stuff.

  47. NBC

    “So do I, which is why I have shown all my work as well. Of course, I have done more. I have shown your interpretations to be flawed and how simple work flow explains much of it.”

    Just where did you show the original paper copy that was scanned with your Xerox twins?

    And then you haven’t posted a link to any of the Preview created PDFs.

  48. NBC

    “And you are still unable to generate a Preview document… After all this time…”

    And just why would I want to do that ?

  49. Yup ! It figures that Xerox would plant a useless comment in all their JPEGS.

    I wonder why Preview would leave it in there ?

    All these have simple answers

    1. You can take the comment out and it still renders. Of course, since it is a comment it has no real relevance. At best it may have had some relevance internally.
    2. Preview does not touch the JPEG data streams, again for obvious reasons…

    Now that was not too hard but i am sure you could have done the experiments yourself…

    Sigh…

  50. So wouldn’t MAC OS be the forger rather than Xerox ?

    In that case, even if the forger were a human, your suggestion is that we should consider MAX OS to be the forger.

    I can’t wait for you to properly inform the court of your new finding🙂

  51. And just why would I want to do that ?

    I could think of quite a few:

    1. Intellectual curiosity
    2. Supporting evidence to debunk my hypotheses
    3. No wasted time on affine transforms for the wrong document…

  52. Just where did you show the original paper copy that was scanned with your Xerox twins?

    I have not… How do I show a paper copy other than by scanning it? A photo perhaps?

    And then you haven’t posted a link to any of the Preview created PDFs.

    And it is driving you crazy… I am feeding you the relevant information and if you believe I am somehow hiding evidence that disproves my thesis, I suggest that you do some legwork…

    It cannot all be one sided… You managed to work your way through the affine transforms, surely you can find someone with a Mac?…

    You have been given all the necessary information to do your own experiments. In fact, you could even download the Xerox 7655 tax return…

    Let’s consider it as a test… Yes…

  53. NBC

    “Oh, and I showed why your work on the affine transforms was a bit of a waste of time, other than of course providing you with an opportunity to work through them yourself.”

    If its OK with you I prefer to do my own math. That way I get the correct solutions.

    And did I mention that I am way ahead of you?

    “I am impressed… You managed to do what you claimed before was impossible because of some ‘conspiracy’ involving Adobe…”

    Ahhhhhhh ! but you don’t know the end of that story. You didn’t take my advice so as always you remain clueless.

    And take my word for it. The bottom line of this little omission on your part is all bad for your storyline.

    I fully expect one of your readers, who is inclined to work fascinating mathematical problems, to pick up my challenge and beat you to the punch.

  54. Ahhhhhhh ! but you don’t know the end of that story. You didn’t take my advice so as always you remain clueless.

    And take my word for it. The bottom line of this little omission on your part is all bad for your storyline.

    I fully expect one of your readers, who is inclined to work fascinating mathematical problems, to pick up my challenge and beat you to the punch.

    Again expecting others to do your work? Fascinating. And yes, I am all in favor of you working your way through an example, even if the findings did not help you.

    Sure, you have been saying forever that you are ‘ahead of me’… So why do we see you repeating my own findings… Days later, if not more…

    As I said, I would love for you to prove me wrong… That’s the wonderful world of science…

  55. NBC

    “And since you are upset that I do not give enough credit to Preview, I will add that Preview added its own telltale signs when it rotated the image 180 degrees and saved in in PDF 1.3 format.”

    Did you and your grunts stand there and watch the image flip 180 on the screen ?

    Gee, I wonder how Preview programmed itself to do the flip ?

    [NBC: That’s just silly… YOu now call someone flipping a PDF a forger?]

  56. NBC

    “Again, everything lines up so nicely…”

    If you mean that nothing in your PDF supports any of your claims then I agree with you.

  57. NBC

    Again expecting others to do your work? Fascinating. And yes, I am all in favor of you working your way through an example, even if the findings did not help you.

    Too bad for you but it’s all done and checked.

    [NBC: I can’t wait..]

    And it destroys you little theory of the machine forger.

    [NBC: The ever promissory note…]

  58. NBC

    “I have no reason to submit an affidavit? Why would I? Like you I am not an expert, and like you, my affidavit will never be used in a court case. Sure, it may be filed as ‘evidence’ but it will fail to be taken into consideration by the courts.”

    Well that’s where you and I depart Dude. You see Engineers faithfully document and report their work in detailed technical reports, standards, test procedures, risk assessments, failure analyses, and various publications and technical journals.

    And then they stand on their work until it is universally accepted.

    And until it is they defend it totally.

    Engineers don’t get as pass Man.

  59. Well that’s where you and I depart Dude. You see Engineers faithfully document and report their work in detailed technical reports, standards, test procedures, risk assessments, failure analyses, and various publications and technical journals.

    And then they stand on their work until it is universally accepted.

    And until it is they defend it totally.

    Engineers don’t get as pass Man.

    Well, your report has been found wanting and I have documented this… Can we expect something similar from you?

  60. If you mean that nothing in your PDF supports any of your claims then I agree with you.

    And yet, it all does… So far you have been totally unable to refute anything….

  61. NBC,

    “So far so good, yes, as I have been explaining for some time now, the instructions rotate the images 90 degrees clockwise as they are stored in “landscape”.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    Too bad that your Xerox six-vectors prove otherwise.

    I guess you missed the fact that at this point in my post that I was discussing my findings for the WH LFCOLB and not your Xerox PDF. I later proved how the rotation angle for your Xerox PDF must be zero.

    Maybe it’s time for you to post that Preview PDF file Dude. Then in mere minutes we could print out all of those “fixes” that you claim that Preview provides for the mistakes made by your Xerox forger.

  62. Too bad that your Xerox six-vectors prove otherwise.

    I guess you missed the fact that at this point in my post that I was discussing my findings for the WH LFCOLB and not your Xerox PDF. I later proved how the rotation angle for your Xerox PDF must be zero.

    Maybe it’s time for you to post that Preview PDF file Dude. Then in mere minutes we could print out all of those “fixes” that you claim that Preview provides for the mistakes made by your Xerox forger.

    I have shown that the PDF format matches what you expect it to be. You relied on comparing apples and oranges. Yes, I had already shown that the xerox vectors are different from the Preview vectors. Which is why I was surprised why you would insist on comparing them…

    You do know that preview rewrites a lot of the PDF when saving? It has to do this for a variety of reasons, most importantly PDF 1.3 compatibility.

    These are simple experiments one can do with Preview. And yes, I have done them.

    Yes, the Xerox vectors explain why Illustrator renders it sideways, however the Preview vectors show why the file when printed by Preview, correctly displays.

    Did you miss my postings on this? I documented all these steps. Including the 0 and 90 degree rotations.

    You would have a point if the WH PDF had not been saved by Preview. But in order to compare you need to convert the Xerox document through preview.

    That’s so self evident that I am shocked I have to point this out.

    You promised that you were going to post a rebuttal of my work flow. But you do not even have the Preview PDF…

    Step 1. Convert Xerox PDF to Preview PDF…
    Step 2. Redo the analysis.
    Step 3. Check your findings with my reported findings.

    Step 4. Explain the embedded comment
    Step 5 Explain the rotation in the images
    Step 6 Explain the matrix quantization
    Step 7 Explain the alignment of the monochrome bitmaps
    Step 8. Explain the 8×8 bit alignment

    I can do all 8 steps… Now I am waiting for you to do the same…

    Then we can compare notes by researcher friend. Ideally you would try to disprove my hypotheses but for now, it is good that you repeat my work to double check. I appreciate your efforts in this area. Having someone independent confirm my findings is important too…
    It’s called repeatability.

Comments are closed.