Educating the Confused – Preflight

Hermitian appears to be having some troubles parsing the pdf using preflight, so let’s walk him through the process

  1. Open document in Adobe Acrobat
  2. Select preflight
  3. Select Browse Internal PDF Structures
  4. Expand tabs

preflight expand

Note how at the top there is a clipping path set which continues to apply to all subsequent objects. Note that after the clipping path, the graphics state is saved, the image is displayed and the graphics state is restored to the clipping mask setting. This is repeated for each image.

It helps to understand PDF…

graphics state operators manipulate the data structure called the graphics state, the global framework within which the other graphics operators execute. The graphics state includes the current transformation matrix (CTM), which maps user space coordinates used within a PDF content stream into output device coordinates. It also includes the current colour, the current clipping path, and many other parameters that are implicit operands of the painting operators.
Clipping paths are set using the W and W* operators. At the beginning of each page, the clipping mask is reset to the whole page.

There is no way to enlarge the current clipping path or to set a new clipping path without reference to the current one. However, since the clipping path is part of the graphics state, its effect can be localized to specific graphics objects by enclosing the modification of the clip ping path and the painting of those objects between a pair of q and Q operators (see 8.4.2, “Graphics State Stack”). Execution of the Q operator causes the clipping path to revert to the value that was saved by the q operator before the clip ping path was modified.

And side by side to show how beautifully the WH LFBC and the WH 7535 Preview version align…

side by side

71 thoughts on “Educating the Confused – Preflight

  1. NBC

    PDF images are assembled from bottom to top in your screenshot not from top to bottom as you imply.

    The interaction of the clipping path and Im1 or obj [7 0 R] is shown in the middle window detail here:

    See: http://www.scribd.com/doc/162384363/PreFlight-PDF-Tree

    The assembly of the PDF image also flows from bottom to top in this PDF Tree Structure.

    The connection between the Clipping Mask and obj [7 0 R] is exactly as I previously posted.

    The Clipping Mask is applied to IM1 in the branch:

    Path segments to clip
    [12.0 12.0 600.0 780.0]
    “W [set clipping path using nonzero winding number rule] (1 [(5)])
    [12.0 12.0][588.0 768.0] re [Append rectangle to path]

    An identical branch is appended to each XObject in the list.

  2. It appears that the Mac version is ordered differently than the Windows version. Also, the Mac version states the actions using the actual commands in the PDF; the Windows version merely gives a generic description.

  3. The connection between the Clipping Mask and obj [7 0 R] is exactly as I previously posted.

    The Clipping Mask is applied to IM1 in the branch:

    and to all the other images as well..
    So clueless and now backpedalling as well

  4. Well, there is a reason to use a Mac and if using Windows caused Hermitian confusion, he should have looked at the raw PDF.

    Such a waste of money… when you can do all of this for free and far more accurately.

  5. NBC says:

    August 23, 2013 at 05:44

    “The connection between the Clipping Mask and obj [7 0 R] is exactly as I previously posted.”

    “The Clipping Mask is applied to IM1 in the branch:”

    “and to all the other images as well..”
    “So clueless and now backpedalling as well”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Talk about clueless. Look who’s suddenly using all of my tools. Of course only because I mentioned which tools I am using. Anybody remember NBC using Adobe Acrobat Preflight before my post about the Xerox 7535 PDF Tree Structure ?

    First NBC has glommed onto Illustrator and now it’s Acrobat Prefight. His latest ripoff is proof of his growing desperation as he throws away all of his useless freetoys and adopts mine.

    But here’s a little refresher on my recent posts…

    1. I answered the following comment by WKV:

    “W. Kevin Vicklund says:

    August 18, 2013 at 18:21

    “Hermie, the 7535 image was made from printing out the WH LFBC PDF and then scanning it on a 7535 Xerox WorkCentre. Anything blocked by the clipping mask in the WH LFBC PDF does not print out. Therefore, it will not appear on the 7535 image.

    “Also, the clipping mask is not a separate object, as has been explained to you multiple times. It is a command contained in the Obj 6 flate-encoded string.

    “Your entire comment completely fails.”

    And my answer was:

    Hermitian says:

    August 19, 2013 at 00:41

    “W. Kevin Vicklund says:

    “”August 18, 2013 at 18:21″”

    “”Hermie, the 7535 image was made from printing out the WH LFBC PDF and then scanning it on a 7535 Xerox WorkCentre. Anything blocked by the clipping mask in the WH LFBC PDF does not print out. Therefore, it will not appear on the 7535 image.

    “”Also, the clipping mask is not a separate object, as has been explained to you multiple times. It is a command contained in the Obj 6 flate-encoded string.

    “”Your entire comment completely fails.””

    “Per two different PDF code parsers, the background image is Im1 and obj {7 0 R}.
    Per Adobe Acrobat XI Pro Preflight the Clipping mask is not associated in any way with either the DCTDecode filter or the FlateDecode filter. Instead, the clipping mask is associated directly (and only) with obj {7 0 R}.”

    “And to think that you and NBC claim to be the PDF code masters.”

    “But then unbridled arrogance always proceeds a big fail”.

    My reference to “(and only) with obj {7 0 R}” was to make it clear to Vicklung that the Clipping Mask is not “a command contained in the OBj 6 flate-encoded string”. Rather it lies on a separate branch in the Tree. However my comment is also true in the broader context in that the background layer is the only one that is clipped because the boundaries of all the other objects do not extend beyond the Clipping Mask.

    Some backpedal !

  6. NBC

    Here’s a head-to-head comparison of the PDF Trees, mine from Windows and NBC’s from MAC OS. The ordering of objects is different between the Windows and MAC versions of Adobe Acrobat XI Pro Preflight. I prefer the Windows output because Apple has a history of mangling PDF files.

    But since NBC is familiar with the idiosyncrasies of MAC OS PDFs I’ll let him sort out any difference between the MAC and Windows output.

    As far as I know there are no Preflight capabilities that exist in the MAC OS version of Preflight that are not available in the Windows version. The Windows version can display in five different formats which are user selectable by the five “puzzle piece” buttons at the top-right border of the page. The results posted on Scribd are output selected by the leftmost button in the row of five buttons. The tree is only partially expanded in the posted image.

  7. Here are two more head-to-head shots from the Xerox 7535 / Preview and the archive copy of the WH LFCOLB. An obvious difference is seen in the ordering of the numbers in all of the “cm matrices” between the two different files.

  8. Hermie is right, I did make an error. Here is the corrected version:

    Also, the clipping mask is not a separate object, as has been explained to you multiple times. It is a command contained in the Obj 4 flate-encoded string

    So yeah, the rest of my comment still stands.

  9. Here are two more head-to-head shots from the Xerox 7535 / Preview and the archive copy of the WH LFCOLB. An obvious difference is seen in the ordering of the numbers in all of the “cm matrices” between the two different files.

    Indeed, this is because the 7535 file was originally scanned right-side-up, whereas the original WH LFBC was scanned upside-down. The 7655 upside-down fed scans have the same order as the WH LFBC. This is how NBC determined that the WH LFBC was scanned in upside-down.

  10. NBC

    W. Kevin Vicklund says:

    August 23, 2013 at 13:25

    “”Here are two more head-to-head shots from the Xerox 7535 / Preview and the archive copy of the WH LFCOLB. An obvious difference is seen in the ordering of the numbers in all of the “cm matrices” between the two different files.””

    “Indeed, this is because the 7535 file was originally scanned right-side-up, whereas the original WH LFBC was scanned upside-down. The 7655 upside-down fed scans have the same order as the WH LFBC. This is how NBC determined that the WH LFBC was scanned in upside-down.”

    You mean the paper original had to be scanned upside-down for the image to appear right-side-up in illustrator. So whatever it takes to fit NBC’s workflow and storyline.

    I thought so.

  11. You mean the paper original had to be scanned upside-down for the image to appear right-side-up in illustrator. So whatever it takes to fit NBC’s workflow and storyline.

    No, it has to be saved by Preview for it to appear right-side-up in Illustrator, as you well know, you lying sack of shit. Because Illustrator can’t handle the /Rotate function, as you well know, you lying sack of shit.

  12. WKV

    As I was saying the Obots are getting desparate !

    Sometimes the truth hurts when it just doesn’t fit your storyline.

  13. Sometimes the truth hurts when it just doesn’t fit your storyline.

    You still do no understand the story line now do you and you continue to misrepresent it. That’s desperate my friend. People are getting frustrated by your ignorance and/or inability to understand the simple work flow…

    The truth never hurts, however your interpretations of the ‘truth’ is quite desperate my friend and the frustration of people having to continuously educate you on the use of tools, simple concepts, PDF, JPEG etc becomes a bit tiresome.

    What our friend is struggling with here is that he does not understand how one can determine the orientation of the scan, even though people have explained it to him more than once.

    Hermitian still has no clue here… And yet, it is conceptually so simple…

  14. Here are two more head-to-head shots from the Xerox 7535 / Preview and the archive copy of the WH LFCOLB. An obvious difference is seen in the ordering of the numbers in all of the “cm matrices” between the two different files.

    Yes, the numbers will differ but the overall flow is remarkably similar. You are looking at minor expected differences and failing to see the level of overlap.

    I understand, that’s about all you have left to argue about… You too will come to terms with this… The facts are just too overwhelming in favor of the Xerox workflow and you have done NOTHING to even start to debunk the findings, other than by arguing that it does not look exactly the same.

    You’re doing fine my friend…

  15. W. Kevin Vicklund says:

    “August 23, 2013 at 12:30

    “Hermie is right, I did make an error. Here is the corrected version:”

    “Also, the clipping mask is not a separate object, as has been explained to you multiple times. It is a command contained in the Obj 4 flate-encoded string”

    So yeah, the rest of my comment still stands.

    Your comment stands only if your “Flate decoded string has lots of branches. Like one for each of the 17 objects.

    And again, the PDF image is assembled from the top of the TREE down which is bottom to top in the Preflight TREE Diagram.

  16. Hermitian: You mean the paper original had to be scanned upside-down for the image to appear right-side-up in illustrator. So whatever it takes to fit NBC’s workflow and storyline.

    Nope, the data show that the document was most likely scanned upside down and rotated 180 degrees when the recipient/scanner opened the document in preview and printed it to PDF. You can do experiments, well at least one can do experiments, to see the effect of this on the scanned document. The most plausible explanation that explains all the data is the workflow I outlined.

    My workflow is not based on my imagination but rather informed by the evidence. That’s where we diverge in our approaches.

  17. NBC

    “What our friend is struggling with here is that he does not understand how one can determine the orientation of the scan, even though people have explained it to him more than once.”

    The orientation question is not as simple as re-orienting the original on the glass until the image appears right-side-up in most graphic programs. You took the easy way out and now you have to live with your up-side-down workflow.

    I don’t know any judges that will readily accept that the Xerox operator placed the single page copy of the President’s LFCOLB upside-down on the glass.

    I certainly don’t. And you obviously don’t have a clue.

  18. Your comment stands only if your “Flate decoded string has lots of branches. Like one for each of the 17 objects.

    So you are still ignorant about the Q/q stack, and the graphics state in PDF… Fine, we tried to explain this to you, but with no evidence of understanding on your part.

    It’s quite simple, the “Re W n” instruction creates a state which applies to all subsequent elements. The elements themselves are surround by a push and pop (Q/q) instruction that allows them to temporarily change the cm matrix, but once Q is executed, the state is restored.

    A PDF expert you are not my friend… Looking at Illustrator and drawing conclusions that could have been avoided if you had looked more carefully at the raw PDF data and understood the meaning of the Q/q operators as well as the graphics state, could have avoided yet another embarrassing moment. But there have been plenty already, I guess.

    Sigh…

  19. And again, the PDF image is assembled from the top of the TREE down which is bottom to top in the Preflight TREE Diagram.

    Nope, the top of the tree contains the background, the rest contains image masks, if you were to build it bottom to top, you would not get the same result. Perhaps you want to be more careful in the terminology you try to use here.

    Good luck and do you need a reminder link to the PDF standard, just so that you can educate yourself about the relevant concepts here? I believe I provided the relevant quotes already but I am more than willing to help you understand in much smaller steps. I had assumed that you were at least familiar with the PDF ‘language’ but then again you were also struggling with the order of deflate and DCTDecode, so I should have perhaps taken more proper notice here and should have adapted my explanations accordingly. For that I apologize.

    So let me know what still confuses you.

  20. I don’t know any judges that will readily accept that the Xerox operator placed the single page copy of the President’s LFCOLB upside-down on the glass.

    I certainly don’t. And you obviously don’t have a clue.

    You are now projecting, my confused friend. As to the Judge, I assume that if it ever were to get to that stage, he would understand quite well how my workflow explains all the observations.

    And yes, I believe that I can prove that the document was placed upside down, and if you had taken the time to think about the riddle I gave you a while ago, you would too have understood.

    It’s subtle but I believe that it fully supports my workflow.

    However, just to show how silly your argument is: even if the workflow includes the scan having been done the right side up, and then printed to PDF in preview, my workflow would still explain over a dozen ‘artifacts’ that were once argued to be evidence of forgery and the Judge would throw out anything that argued otherwise. It’s because the data clearly is consistent with my workflow, and there remains no evidence of any forger, nor has there ever existed a case for a forger which was not based on an argument from ignorance.

    The orientation question is simple:

    You scan right side up and the object shows up 90 rotated in one direction, you scan upside down and rotate 180 degrees and the object shows up internally rotated 90 degrees the opposite direction.

    You too could have done the experiments if you had the proper tools.

    Now you show once again that you do not understand, what to most of us, is a simple evidence based component of the workflow.

    Think about it, which edge shows up first in the scan… In one orientation it is the left edge, which is at the top of the rotated image, in the other, it’s the right edge which is now at the top. Compare the orientation of the 7535 with the WH LFBC and you see it is rotated 180 internally, so rather than the cm rotating it clockwise, it is rotated counter clockwise.

    It’s so trivially obvious that I am still flabbergasted that you never really answered my riddle.

    Do you need a step by step explanation? Perhaps with pretty pictures and some illustrator screen shots? Or do you think you can figure it out yourself?

    Let me know, I am a patient person and I am sure I can help you understand this last minor objection on your part to my, by now, well supported workflow.

    If you disagree, then outline to me which of the features remain poorly supported.

    I have created a wonderful list with supporting evidence here just in case you have missed it.

    Good luck my friend, I do not envy your situation.

  21. You took the easy way out and now you have to live with your up-side-down workflow.

    As I said before, my workflow is driven by the data and evidence, not the other way around.

  22. PDF images are assembled from bottom to top in your screenshot not from top to bottom as you imply.

    That depends on your perspective. The commands executed start with the masking layer, then it paints the background and then the foreground. If it were the other way around, the text would have been covered by the background. You may have been confused by how the layers are stacked in Illustrator, which indeed shows them upside down, because the text layers are physically on top of the background layer.

    Think about it for a moment and you too will understand where you went wrong.

  23. NBC – does the data show whether they used the sheet feeder or opened the top and placed it directly on the glass platen?

  24. Gorefan: NBC – does the data show whether they used the sheet feeder or opened the top and placed it directly on the glass platen?

    Based on my experiments so far, they used the sheet feeder. However, to be more conclusive about this, I will have to do some additional experiments and evaluations. Good questions though.

  25. I have been searching for Third-Party Vendors of PDF SDKs that could have supplied Xerox with the MRC software. I have found that a leading contender is Leadtools.

    See: http://www.leadtools.com/sdk/pdf/default.htm

    I have downloaded the Leadtools evaluation package and found that they supplied two example PDFs. One of these PDS exhibits their proprietary MRC technology. Leadtools has a twenty-year history of continuous development of digital imaging tools. They have a long history working with the medical community to meet their imaging needs for medical imaging systems for diagnostic tests. The Leadtools Image SKD supports over 150 file formats.

    This example PDF is a one-page document “PDFSegmentation.pdf”. I have compared this MRC compressed PDF file with the Xerox 7535 scan to PDF file “wh-lfbc-scanned-xerox-7535-wc.pdf in Adobe Illustrator and found that the layer tree structure of the Leadtools PDF is identical to the layer structure of the Xerox 7535 scan to PDF file. Both PDF files display two vertical offsets in the Layer Tree Structure. Both files do not have the extra Clipping Path as is found in the Preview print to PDF.

    The Leadtools one-page file has a total of 11 objects, including three 8-Bit JFIF DCT decoded images, three 1-Bit Flate decoded bitmap images, and five 2-Bit Flate decoded images. The Xerox 7535 scan to PDF file has a total of 17 objects, including one 8-Bit JIFF DCT decoded image and 16 1-Bit JBIG2 decoded bitmap images. Leadtools also supports JBIG2 compression. They also support JPEG “lossy” and “lossless” compression as well as many other popular compression techniques. The following list of file compression options was copied from the Leadtools web site:

    “Maintain quality while maximizing PDF compression with LEADTOOLS advanced image segmentation and compression technologies. The resulting compressed PDF can be loaded and viewed in any PDF viewer that supports standard PDF files. By storing complex mixed raster content (MRC), this process creates PDF files with better compression and quality than a standard raster PDF file.

    -Automatically segment the image with optimization options
    -Manually segment the image to take full control over file size and image quality optimization
    -Compression for different segment types can be automatically or manually selected
    -Multiple compression options including:
    -ZIP
    -LZW
    -CCITT G3 /G4
    -JBIG2
    -JPEG
    -Automatic background detection
    -Compress single or multi-page PDF files
    -Native 32 and 64 bit binaries for compressing PDF files
    -Add PDF compression to single or multi-threaded applications

    All of the images in both files are embedded. The background layer of the Xerox 7535 scan to PDF file is labeled as “image embedded” in the Taskbar. The 16 1-Bit images are labeled as “bitmap embedded” in the Taskbar. All eleven images in the Leadtools PDF are labeled as “image embedded” in the Taskbar. Thus many of these 1 or 2 Bit images may be embedded vector images. This would be expected because many of the images contain text.

    I have also compared the first two objects in each file in Adobe Acrobat Preflight. The PDF Tree structures for both PDFs are posted on Scribd.com here:

    The first page of this two-page Scribd document is the Illustrator CC image of the file “PDFSegmentation.pdf” with all 11 objects selected. The second page is a side-by-side comparison of the Preflight Tree Structures for the two PDF files.

    The ordering of Flate decoded images vs DCT decoded images is different between the two PDFs. Nevertheless, the PDF Tree Structures are very similar. The Leadtools PDF file has the aforementioned “two strings” in each of the DCT decoded JFIF images. The locations of the “two strings” is not in any Huffman Table. None of the three DCT decoded JFIF compatible images has Huffman Tables. Most likely this is because Leadtools has used their proprietary JFIF compatible compression algorithms. The PDF creator and producer are :

    //<>//

    Hence, the Leadtools “CompactPDF” technology was applied to this example PDF image.

    The Leadtools PDF file also includes the JFIF comment “LEAD Technologies Inc. V1.01”. The comment “YcbCr” is not found in the Leadtools file. Neither the “two strings” nor the “YcbCr” comment is found in the Xerox 7535 scan to PDF file. The “two strings” are found in the Preview print to PDF file “wh-lfbc-scanned-xerox-7535-wcpreview.pdf”. The “YcbCr”label is also found in the Preview PDF.

    The following statement from the Leadtools Help files caught my eye.

    “JPEG Compliant
    “LEADTOOLS is fully JPEG-compliant. Any file using compression that is compliant with the JPEG Interchange File Format can be viewed and converted.
    “However, some images that are not fully compliant with the JPEG specification will be displayed upside-down. (The JPEG Interchange File Format specifies that all images must be stored top-down.) You can change the orientation of an image by modifying the ViewPerspective flag in the LEAD bitmap handle.”

    Hmmmmm… Upside-Down Images …Hmmmmmm…

    This sure sounds more probable than an upside-down placement of the Presidents birth certificate on the Xerox WC glass.

  26. Oops !! The comment box ate my data again…

    Creator (LEADTOOLS CompactPDF)
    Producer (LEAD Technologies, Inc.)
    Author (LEADTOOLS PDF Compressor)
    CreationDate (20051107161647)

  27. This sure sounds more probable than an upside-down placement of the Presidents birth certificate on the Xerox WC glass.

    Again you have shown that your forgery explanation continues to fail and how the YCbCr comment is not found in the software….

    Such a joker…

    And you still do not understand the cause of why the JPEG is scanned in landscape and how one can detect upside down.

    Not impressed…

  28. NBC says:

    August 23, 2013 at 19:38

    “”Gorefan: NBC – does the data show whether they used the sheet feeder or opened the top and placed it directly on the glass platen?””

    “Based on my experiments so far, they used the sheet feeder. However, to be more conclusive about this, I will have to do some additional experiments and evaluations. Good questions though.”

    So our drunk paralegal was also so lazy that he used the document feeder for a one-page original ?

    And he was so wasted that he couldn’t figure out the visual aids on the feeder which indicate the correct placement orientation for the original ?

    So you still don’t know whether or not the feeder was used ?

    I thought that by now you would have at least eliminated one of these two degrees of freedom.

    So your simple workflow is not so simple after all ?

  29. HHHH: Oops !! The comment box ate my data again…

    If you want to avoid the stripping of codes in the comments, here is a trick (supposing it works correctly here as it did on a test with another WordPress blog):

    1. Briefly copy the text into MS Word or similar program.
    2. Replace all of the < with “& l t ;” (without the quotes and spaces. I had to use them to show the characters).
    3. Replace all of the > with “& g t ;” (without the quotes and spaces).
    4. Copy and paste into the comment screen.

    With that, you should be able to paste something like the following (drumroll please):

    <?xpacket begin=”o;?” id=”W5M0MpCehiHzreSzNTczkc9d”?>
    <x:xmpmeta xmlns:x=”adobe:ns:meta/” x:xmptk=”Adobe XMP Core 5.2-c001 63.139439, 2010/09/27-13:37:26 “>
    <rdf:RDF xmlns:rdf=”http://www.w3.org/1999/02/22-rdf-syntax-ns#”>
    <rdf:Description rdf:about=””
    xmlns:xmp=”http://ns.adobe.com/xap/1.0/”>
    <xmp:ModifyDate>2013-03-01T11:56:18-05:00</xmp:ModifyDate>
    <xmp:CreateDate>2013-03-01T11:56:18-05:00</xmp:CreateDate>
    <xmp:MetadataDate>2013-03-01T11:56:18-05:00</xmp:MetadataDate>
    <xmp:CreatorTool>Acrobat PDFMaker 10.1 for Word</xmp:CreatorTool>
    </rdf:Description>
    <rdf:Description rdf:about=””
    xmlns:xmpMM=”http://ns.adobe.com/xap/1.0/mm/”>
    <xmpMM:DocumentID>uuid:4b8b9cd6-2acb-4ed8-b4ef-e51aa7aea163</xmpMM:DocumentID>
    <xmpMM:InstanceID>uuid:f34ba48e-6e10-4ff0-9236-9b36f313941e</xmpMM:InstanceID>
    <xmpMM:subject>
    <rdf:Seq>
    <rdf:li>1</rdf:li>
    </rdf:Seq>
    </xmpMM:subject>
    </rdf:Description>
    <rdf:Description rdf:about=””
    xmlns:dc=”http://purl.org/dc/elements/1.1/”>
    <dc:format>application/pdf</dc:format>
    <dc:creator>
    <rdf:Seq>
    <rdf:li/>
    </rdf:Seq>
    </dc:creator>
    </rdf:Description>
    <rdf:Description rdf:about=””
    xmlns:pdf=”http://ns.adobe.com/pdf/1.3/”>
    <pdf:Producer>Adobe PDF Library 10.0</pdf:Producer>
    </rdf:Description>
    <rdf:Description rdf:about=””
    xmlns:pdfx=”http://ns.adobe.com/pdfx/1.3/”>
    <pdfx:SourceModified>D:20130301165603</pdfx:SourceModified>
    </rdf:Description>
    </rdf:RDF>
    </x:xmpmeta>

  30. So you still don’t know whether or not the feeder was used ?

    All the factors indicate that a feeder was used. But again, it does not matter, the scan showed that it was copied up side down. Your assertion that this would require a drunken paralegal somehow lacks in evidence and relevance.

    I am following where the data lead me, you just speculate rather than engage the known data. The detail of how the image was fed, is quite minor but allows you to ignore the massive amount of indicators that support my workflow.

    I do understand my friend, honestly I do and I am glad that you did not disappoint me.

  31. Hermitian: I have been searching for Third-Party Vendors of PDF SDKs that could have supplied Xerox with the MRC software. I have found that a leading contender is Leadtools.

    You have failed to explain how you have reached this conclusion and what vendors you have taken into consideration. Furthermore, since Xerox has patents in this area, and an incredible research arm called PARC Xerox, I have to express some disbelief with your claims, especially since the SDK fails at many aspects to capture the essence of what is found in Xerox scanners.

    Hermitian may not be familiar with Parc Xerox but a lot of research was done in the compression of documents, for obvious reasons, and Xerox holds many relevant patents in this area.

    So how has our friend eliminated an internal source within Xerox for the MRC software for example?

  32. Talk about clueless. Look who’s suddenly using all of my tools. Of course only because I mentioned which tools I am using. Anybody remember NBC using Adobe Acrobat Preflight before my post about the Xerox 7535 PDF Tree Structure ?

    I am using them since you seem to require them to understand what others can see in the raw data. It’s a favor to you my friend… And I still have to educate you on the interpretation of what these tools show, even though I had reported on the PDF format weeks before Hermitian figured out the basics of PDF and JPEG format encoding.

    However my comment is also true in the broader context in that the background layer is the only one that is clipped because the boundaries of all the other objects do not extend beyond the Clipping Mask.

    Talking about backpedaling now… Hilarious…

  33. NBC: All the factors indicate that a feeder was used. But again, it does not matter, the scan showed that it was copied up side down. Your assertion that this would require a drunken paralegal somehow lacks in evidence and relevance.

    The more data you provide, the more Hermie has to look for some obscure supposition that makes the whole PDF “suspect.” However, even his scenario looks foolish as compared to a simple:

    1. Somebody from the Press office (or the White House Counsel’s office) goes to the Xerox to scan. As an alternative, they pass it off to the Office of Digital Strategy.
    2. They enter their’s or another’s email address after a brief search. I make the assumption that the machine is connected to their Exchange server.
    3. While talking to somebody else, they scan the document, either on the glass or document feeder.
    4. They pull the document off the machine and go on their merry way.
    4. Eventually, somebody from the Office of Digital Strategy (hint.. they are the main users of Macs in the WH) or the Press Office opens the PDF and says, “oops, its upside down”.
    5. NBC’s workflow continues.

    In an office setting, scanning upside down would not be that unusual, especially if it is some little quick project that you want to finish quickly and get back to the more important aspects of your job (yes, Hermie, they’re just not that into you and the birthers). Even I, who can be somewhat anal about details (nothing says wiener like an upside-down slide), has run to the Xerox (yes, I use WorkCentres also), scans something quickly and notice when I get back to my desk that it is upside down. Even Orly enjoys scanning upside down for some of her filings, though the rest of us normally fix it at some point.

  34. Didn’t they invent the GUI and the mouse at PARC way back in the 1970′s? I am not going to cheat and look it up.

    And Steve Jobs payed Xerox a good chunk of money to roam around and thus the Mac was born🙂

  35. NBC

    “”Hermitian: I have been searching for Third-Party Vendors of PDF SDKs that could have supplied Xerox with the MRC software. I have found that a leading contender is Leadtools.””

    “You have failed to explain how you have reached this conclusion and what vendors you have taken into consideration. Furthermore, since Xerox has patents in this area, and an incredible research arm called PARC Xerox, I have to express some disbelief with your claims, especially since the SDK fails at many aspects to capture the essence of what is found in Xerox scanners.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    The capabilities of the LeadTools SDKs are beyond those of the Xerox workcenter. The LeadTools compression technology is not limited to one 8-Bit JFIF color layer and multiple 1-Bit monochrome layers. LeadTools supports over 150 file formats. LeadTools image segmentation supports up to nine different layer types.

    The capabilities of the LEADTOOLS PDF SDK are detailed here:

    http://www.leadtools.com/sdk/pdf/default.htm

    Likewise the capabilities of the LEADTOOLS Mixed Raster Content (MRC) SDK are detailed here:

    http://www.leadtools.com/sdk/mrc/default.htm

    And the LEADTOOLS Image Segmentation SDK capabilities are detailed here:

    http://www.leadtools.com/sdk/image-processing/segmentation.htm

    Finally the details of the LEADTOOLS Image Compression SDK are here:

    http://www.leadtools.com/sdk/compression/default.htm

    Together the SDK tools far surpass the digital imaging capabilities of the Xerox Workcenters.

    For LeadTools to meet any digital imaging software needs of Xerox for their Workcenters would be a walk in the park.

    LeadTools is the leader for providing digital imaging technologies to manufactures of medical equipment.

    “Medicor Imaging, a division of LEAD Technologies, Inc. was created to provide the medical imaging community with products and professional services that will help facilitate a more rapid transition into the digital era.”

    The LEADTOOLS Partners and Affiliates list is here:

    http://www.leadtools.com/corporate/partners.htm

  36. The capabilities of the LeadTools SDKs are beyond those of the Xerox workcenter.

    Thanks for verifying that Leadtools is not the provider of the Scanning solution for Xerox.

    Geez, you are wasting your time on so many dead ends and have yet to raise any relevant objection to my findings.

    You’re so useful…

  37. Hermie wrote

    The capabilities of the LeadTools SDKs are beyond those of the Xerox workcenter. The LeadTools compression technology is not limited to one 8-Bit JFIF color layer and multiple 1-Bit monochrome layers. LeadTools supports over 150 file formats. LeadTools image segmentation supports up to nine different layer types.

    Thanks for once again helping to prove the Cold Case Posse didn’t know what they were talking about. Well done!

  38. Thanks for once again helping to prove the Cold Case Posse didn’t know what they were talking about. Well done!

    With Hermitian on our team, how could we possibly lose🙂

  39. NBC

    “”The capabilities of the LeadTools SDKs are beyond those of the Xerox workcenter.””

    “Thanks for verifying that Leadtools is not the provider of the Scanning solution for Xerox.”

    Of course most rational people understand that “beyond the capabilities” includes the same capabilities and beyond.

    But then not all people are rational.

  40. Of course most rational people understand that “beyond the capabilities” includes the same capabilities and beyond.

    But then not all people are rational.

    Why would xerox take a subset of the SDK… You are so funny. You have shown how nothing really shows that their solution created the PDF and yet you argue that it could..

    Such despair… Indeed my friend, rationality would not lead one to ignore a dozen markers that have been explained by a simple workflow.

    So far you have been shown wrong in three separate cases: The affidavit, the AP PDF and now you are struggling to not only present some information relevant to your forgery scenario but also unable to rebut any of my findings.

  41. NBC

    I did a search for Mixed Raster Compression on the Xerox Parc website and it returned two hits:
    results for ” Mixed Raster Compression ”

    See: http://www.parc.com/search.html?q=Mixed+Raster+Compression&x=0&y=0

    Sort by Date | Relevance

    “Mockingbird: A Composer’s Amanuensis
    Not shown is an experimental high-resolution, computer driven, raster- ….. Raw piano roll material can be mixed in freely with standard music notation, both on a.

    “Difficult and Urgent Open Problems in Document Image Analysis for …
    ity control and compression are noted in Section 4. Sec- tion 5 summarizes ….. ‘ pure raster’ images, mixed image-and-text formats such as. PDF, HTML, MSWord

    I then did searches in both of these two papers for “MRC” and then “Mixed Raster Compression”.

    All four searches returned zero hits.

    Mind Blowing Technology ! Just Mind Blowing !

  42. NBC

    “Why would xerox take a subset of the SDK… You are so funny. You have shown how nothing really shows that their solution created the PDF and yet you argue that it could..”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    LeadTools just doesn’t give all of their products away for commercial use. Instead they charge a fair price for each capability. So why would any manufacturer load up their MFPs with image capabilities that are seldom (or never) used in that application and are thus not cost effective ?

  43. Mind Blowing Technology ! Just Mind Blowing !

    Indeed… Has it occurred to you to search Google for Xerox Parc and MRC?…

    So inept.

  44. LeadTools just doesn’t give all of their products away for commercial use. Instead they charge a fair price for each capability.

    Huh? Were you under the impression that Xerox would not pay those who provide them with solutions? Of course, you have jumped to conclusions which are somewhat speculative at best.

    You already outlined that none of their technology seems to reproduce the artifacts and now you somehow want us to believe that they only sell old technology to their customers?

    OMG…

    I am soooo glad that we have you play on this blog.

  45. NBC

    “So far you have been shown wrong in three separate cases: The affidavit, the AP PDF and now you are struggling to not only present some information relevant to your forgery scenario but also unable to rebut any of my findings.”

    What about your short-lived excitement when you first realized that the quantization tables for the Xerox 7535 scan to PDF JPEG extracted file were identical to the quantization tables of the WH LFCOLB PDF JEPEG extracted file.

    See: https://nativeborncitizen.wordpress.com/2013/07/25/wh-lfbc-and-the-embedded-jpeg/

    Of course never mind that the Xeroz Workcenter image was upside-down relative to the WH LFCOLB image. And never mind that the luminance tables were identical to the chrominance tables.

    You even blamed me for putting you on the trail of your finding that the quantization tables are identical. However, after that earthshaking finding you evidently found out that the White House has a Xerox 7655 but it does not have a Xerox 7535. So like a “hot potato” you dropped the quantitization tables as another “smoking gun” In fact you changed the subject entirely while you worked the damage control in the background. It was weeks before you finally got back to the Xerox workflow. And then when you finally got back on track you posted that the Xerox 7655 quantization tables do not match the WH LFCOLB quantization tables.

    See: https://nativeborncitizen.wordpress.com/2013/07/26/xerox-7655-additional-experiments/

    Of course you gave your usual lame excuse as to why that was — you blamed it on the out-of-date firmware of the Xerox 7655. I guess because you had no credible way to blame that on me. But then you had no real evidence that the difference in the tables was caused by the firmware. And, you didn’t bother to mention the fact that the Xerox 7535 is an older model than the Xerox 7655.

    And of course you failed to mention the fact that the luminance table was no longer the same as the chrominance table. This difference is not expected for two different machines from the same series and the same manufacturer.

    And since then you launched into a massive program of trials on the Xerox 7655. Of course to carry out such a broad program of trial scans you would have had to have ready access to a 7655. So if you have ready access to a Xerox 7655 why did you not update the firmware?

    Maybe the outcome would not fit your storyline ?

  46. NBC
    Hermitian: Following the Xerox Workflow

    Posted on July 30, 2013 by NBC

    [NBC: Introduction

    “Hermitian’s hypothesis is competing with the following.

    “•The original long form birth certificate was scanned in, upside down, on a Xerox Work Centre, most likely 7655 because it has been shown that the White House owns such a device which produces the same finger prints as found in the WH LFBC PDF”

    “”Have you examined the color fringes in relation to the scan direction and diode sensors to verify that the image was scanned in upside down ? This would prove (or disprove) your upside-down theory.””

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Here’s an earlier question that I posed to NBC that he ducked. I originally asked this question when I noticed that the Xerox 7535 scan to PDF file “wh-lfbc-scanned-xerox-7535-wc.pdf” is completely void of color fringes at the edges of the form lines and text characters.

    Most digitally scanned documents exhibit color fringes. The AP PDF LFCOLB “ap_obama_certificate_dm_110427 copy.pdf” has faint color fringes. The Muscatine Journal LFCOLB PDF “4db82608b486f.pdf” has strong color fringes. Both of these LFCOLB images are believed to have been scanned.

    Only the top-of-the-line scanners can produced scan to PDF documents with no color fringes. I have looked but cannot find any evidence that the Xerox 7535 or the Xerox 7655 MFPs routinely do this.

    I have compared the same two printed characters from the archive copy of the WH LFCOLB PDF and the Xerox 7535 scan to PDF files. The two characters are the word “OF” from the form header “CERTIFICATE OF LIVE BIRTH”.

    The print-to-screen image comparing the word “OF” from these two PDFs is found here:

    The bottom window is from the PDF file “birth-certificate-long-form WayBack Machine 04-27-2011_17-11-11.pdf”. The top window is from the PDF file “wh-lfbc-scanned-xerox-7535-wc.pdf”. The lack of color fringes is noticeable because the White halos are more pronounced for the Xerox 7535 scan to PDF image.

    A second anomaly is the absence of “pincushion” distortion of the individual text characters. The pincushion distortion was particularly pronounced in the page 4 LFCOLB image from court documant 35-1.pdf filed in the Southern District of Mississippi Federal court. The AP PDF file “ap_obama_certificate_dm_110427 copy.pdf”, as previously stated, has faint color fringes. Both of these documents are believed to have been produced by digital scanning of a paper original.

    Both the color fringes and the pincushion distortion were discussed in my 2nd Affidavit which is posted on Scribd.com here:.

    See: http://www.scribd.com/doc/151701882/2nd-Affidavit-UL

    Another noticeable difference is that the text characters from the Xerox 7535 scan to PDF image are more bold and “fatter”. Many of the smallest text characters are solid where the holes are filled in with near-Black color. This is particularly true of the letter “A”. The characters are fatter because random pixels were added in a single layer on both the inside and outside edges of the characters.

    So NBC should examine all of the PDF images that he has collected to determine if any of these have color fringes or pincushion distortion. If any have color fringes then these images should be used to verify the scan direction.

  47. Again our friend wants to focus on ‘color fringes’. The Fact that the characters are ‘fatter’ is not surprisingly as they are scans of the WH LFBC.
    Again, our poor friend is focusing on aspects which can be understood with some application of logic and reason.

    PS: There is a much better way to determine scan direction my friend… But somehow you remain somewhat clueless about how to go about this. That’s too bad.

    And the AP file is most likely a photograph of a copied document.

    For goodness sakes, you really have nothing but unfounded speculation that has been addressed in most cases.

    You are doing fine my friend but you now have three instances where your claims about PDF’s or photographs have been debunked…

    Color fringes… ROTFL
    Pin cushion….

    Why can Hermitian not do this himself…

  48. Of course never mind that the Xeroz Workcenter image was upside-down relative to the WH LFCOLB image. And never mind that the luminance tables were identical to the chrominance tables.

    Which is why you have to scan it upside down. And the Quantization matrices were similar to the WH documents and other Xerox WorkCentre documents so again, you are objecting to simple facts.

    What our friend fails to understand is that a good researcher adapts the hypothesis to meet the data, not the other way around, so when I initially found that the quantization tables did not match with my sample, I retracted my findings, only to realize that different firmware versions may have explained the differences and so when I found other 7655 Xerox workcentres with the same quantization, I retracted my conclusions.

    Note that it is certain that it was a Xerox WorkCentre that created this document, the fact that the Whitehouse used a 7655 to scan in the tax forms, a few weeks earlier was used to make an informed guess about the exact product number. Oh and luminance and chrominance are the same for B&W scans…

    Sorry my dear friend but your comments suggest that you really do not understand how one does research, and form hypotheses, define experiments and so onward. Which is why your claims about the AP Documents, and the MS scan as outlined in the affidavit, so easily rebutted. All it took were some careful experiments…

    I suggest you try the same in your efforts to reject my findings and workflow.

  49. NBC

    I guess you didn’t notice that my initial cooments in my last post referred to the Xerox 7535 scan to PDF and the WH LFCOLB. So just to make sure we are on the same wave length I will first clean up your mess.

    In fact my continuing comparison of the PDF files ““wh-lfbc-scanned-xerox-7535-wc.pdf” and “birth-certificate-long-form WayBack Machine 04-27-2011_17-11-11.pdf”has revealed that the two files have both identical quantization tables and identical Huffman tables. You had reported the identical quantization tables but not the identical Huffman tables. These findings would lead any rational investigator who is trying to prove a Xerox forger to put the Xerox 7535 at the top of the list. And considering that you reported that the Xerox 7655 does not have these same properties it’s very strange that you dropped the Xerox 7535 out of contention.

    As you found, I also found that the luminance and chrominance quantization tables are identical. So between the two PDF files there are four quantization tables which are the same.

    Common Quantization Table

    8 10 11 13 11 9 14 13
    12 13 16 15 14 17 22 36
    23 22 20 20 22 44 32 33
    26 36 52 46 55 54 51 46
    50 50 58 65 83 70 58 61
    78 62 50 50 72 98 73 78
    86 88 93 94 93 56 69 102
    109 101 90 108 83 91 93 89

    Common Huffman dht[0]

    0 0 1 5 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11

    Common Huffman dh[1]

    16 0 2 1 3 3 2 4 3 5 5 4 4 0 0 1 125 1 2 3 0 4 17 5 18 33 49 65 6 19 81 97 7 34 113 20 50 129 145 161 8 35 66 177 193 21 82 209 240 36 51 98 114 130 9 10 22 23 24 25 26 37 38 39 40 41 42 52 53 54 55 56 57 58 67 68 69 70 71 72 73 74 83 84 85 86 87 88 89 90 99 100 101 102 103 104 105 106 115 116 117 118 119 120 121 122 131 132 133 134 135 136 137 138 146 147 148 149 150 151 152 153 154 162 163 164 165 166 167 168 169 170 178 179 180 181 182 183 184 185 186 194 195 196 197 198 199 200 201 202 210 211 212 213 214 215 216 217 218 225 226 227 228 229 230 231 232 233 234 241 242 243 244 245 246 247 248 249 250

    Common Huffman dh[2]

    1 0 3 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11

    Common Huffman dh[3]

    17 0 2 1 2 4 4 3 4 7 5 4 4 0 1 2 119 0 1 2 3 17 4 5 33 49 6 18 65 81 7 97 113 19 34 50 129 8 20 66 145 161 177 193 9 35 51 82 240 21 98 114 209 10 22 36 52 225 37 241 23 24 25 26 38 39 40 41 42 53 54 55 56 57 58 67 68 69 70 71 72 73 74 83 84 85 86 87 88 89 90 99 100 101 102 103 104 105 106 115 116 117 118 119 120 121 122 130 131 132 133 134 135 136 137 138 146 147 148 149 150 151 152 153 154 162 163 164 165 166 167 168 169 170 178 179 180 181 182 183 184 185 186 194 195 196 197 198 199 200 201 202 210 211 212 213 214 215 216 217 218 226 227 228 229 230 231 232 233 234 242 243 244 245 246 247 248 249 250

    So where are these same tables for the Xerox 7655 scan to PDF file ?

    Maybe NBC can extract them and post them ASAP !

  50. n fact my continuing comparison of the PDF files ““wh-lfbc-scanned-xerox-7535-wc.pdf” and “birth-certificate-long-form WayBack Machine 04-27-2011_17-11-11.pdf”has revealed that the two files have both identical quantization tables and identical Huffman tables. You had reported the identical quantization tables but not the identical Huffman tables. These findings would lead any rational investigator who is trying to prove a Xerox forger to put the Xerox 7535 at the top of the list. And considering that you reported that the Xerox 7655 does not have these same properties it’s very strange that you dropped the Xerox 7535 out of contention.

    I am glad to hear that you agree that the Xerox WorkCentre has become the main contender. You and I are now quibbling about minor details such as whether it is a 7655 or 7355 WorkCentre.

    Quite some progress…

    The Huffman tables are identical in many examples and not exclusive to Xerox, I thought you would have realized that by now, after your embarrassment of not recognizing that they were huffman tables.

    The 7535 is hardly out of contention, however given that the White House released the President and Vice President tax forms a few weeks earlier, created on a 7655 makes it a likely candidate.

    I have shown how the 7655 used by the Whitehouse matches the chrominance luminance quantization tables, did I not?

  51. From the Tax forms

    0x08 0x0A 0x0B 0x0D 0x0B 0x09 0x0E 0x0D 0x0C 0x0D 0x10 0x0F 0x0E 0x11 0x16 0x24 
    0x17 0x16 0x14 0x14 0x16 0x2C 0x20 0x21 0x1A 0x24 0x34 0x2E 0x37 0x36 0x33 0x2E 
    0x32 0x32 0x3A 0x41 0x53 0x46 0x3A 0x3D 0x4E 0x3E 0x32 0x32 0x48 0x62 0x49 0x4E 
    0x56 0x58 0x5D 0x5E 0x5D 0x38 0x45 0x66 0x6D 0x65 0x5A 0x6C 0x53 0x5B 0x5D 0x59

    which I have translated for Hermitian’s benefits

     8 10 11 13 11  9 14  13  12  13  16  15 14 17 22 36 
    23 22 20 20 22 44 32  33  26  36  52  46 55 54 51 46 
    50 50 58 65 83 70 58  61  78  62  50  50 72 98 73 78 
    86 88 93 94 93 56 69 102 109 101  90 108 83 91 93 89

    Compared to Hermitian

     8 10 11 13 11  9 14  13  12  13 16  15 14 17 22 36 
    23 22 20 20 22 44 32  33  26  36 52  46 55 54 51 46 
    50 50 58 65 83 70 58  61  78  62 50  50 72 98 73 78 
    86 88 93 94 93 56 69 102 109 101 90 108 83 91 93 89

    Next:

    00 00 01 05 01 01 01 01 01 01 00 00 00 00 00 00 
    00 00 01 02 03 04 05 06 07 08 09 0A 0B

    To help Hermitian

    0 0 1 5 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11

    Compared to Hermitian

    0 0 1 5 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11
    0x10 0x00 0x02 0x01 0x03 0x03 0x02 0x04 0x03 0x05 0x05 0x04 0x04 0x00 0x00 0x01
    0x7D 0x01 0x02 0x03 0x00 0x04 0x11 0x05 0x12 0x21 0x31 0x41 0x06 0x13 0x51 0x61 
    0x07 0x22 0x71 0x14 0x32 0x81 0x91 0xA1 0x08 0x23 0x42 0xB1 0xC1 0x15 0x52 0xD1 
    0xF0 0x24 0x33 0x62 0x72 0x82 0x09 0x0A 0x16 0x17 0x18 0x19 0x1A 0x25 0x26 0x27 
    0x28 0x29 0x2A 0x34 0x35 0x36 0x37 0x38 0x39 0x3A 0x43 0x44 0x45 0x46 0x47 0x48 
    0x49 0x4A 0x53 0x54 0x55 0x56 0x57 0x58 0x59 0x5A 0x63 0x64 0x65 0x66 0x67 0x68 
    0x69 0x6A 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7A 0x83 0x84 0x85 0x86 0x87 0x88 
    0x89 0x8A 0x92 0x93 0x94 0x95 0x96 0x97 0x98 0x99 0x9A 0xA2 0xA3 0xA4 0xA5 0xA6 
    0xA7 0xA8 0xA9 0xAA 0xB2 0xB3 0xB4 0xB5 0xB6 0xB7 0xB8 0xB9 0xBA 0xC2 0xC3 0xC4 
    0xC5 0xC6 0xC7 0xC8 0xC9 0xCA 0xD2 0xD3 0xD4 0xD5 0xD6 0xD7 0xD8 0xD9 0xDA 0xE1 
    0xE2 0xE3 0xE4 0xE5 0xE6 0xE7 0xE8 0xE9 0xEA 0xF1 0xF2 0xF3 0xF4 0xF5 0xF6 0xF7 
    0xF8 0xF9 0xFA

    Or in decimal

    16 0 2 1 3 3 2 4 3 5 5 4 4 0 0 1 125 1 2 3 0 4 17 5 18 33 49 65 6 19 81 97 7 34 
    113 20 50 129 145 161 8 35 66 177 193 21 82 209 240 36 51 98 114 130 9 10 22 23 
    24 25 26 37 38 39 40 41 42 52 53 54 55 56 57 58 67 68 69 70 71 72 73 74 83 84 85 
    86 87 88 89 90 99 100 101 102 103 104 105 106 115 116 117 118 119 120 121 122 131 
    132 133 134 135 136 137 138 146 147 148 149 150 151 152 153 154 162 163 164 165 
    166 167 168 169 170 178 179 180 181 182 183 184 185 186 194 195 196 197 198 199 
    200 201 202 210 211 212 213 214 215 216 217 218 225 226 227 228 229 230 231 232 
    233 234 241 242 243 244 245 246 247 248 249 250

    Compared to his

    16 0 2 1 3 3 2 4 3 5 5 4 4 0 0 1 125 1 2 3 0 4 17 5 18 33 49 65 6 19 81 97 7 34 
    113 20 50 129 145 161 8 35 66 177 193 21 82 209 240 36 51 98 114 130 9 10 22 23 
    24 25 26 37 38 39 40 41 42 52 53 54 55 56 57 58 67 68 69 70 71 72 73 74 83 84 85 
    86 87 88 89 90 99 100 101 102 103 104 105 106 115 116 117 118 119 120 121 122 131 
    132 133 134 135 136 137 138 146 147 148 149 150 151 152 153 154 162 163 164 165 
    166 167 168 169 170 178 179 180 181 182 183 184 185 186 194 195 196 197 198 199 
    200 201 202 210 211 212 213 214 215 216 217 218 225 226 227 228 229 230 231 232 
    233 234 241 242 243 244 245 246 247 248 249 250

    So again the Tax Forms align…

    But there are some additional tables are there not🙂

  52. The Leadtools PDF file also includes the JFIF comment “LEAD Technologies Inc. V1.01″. The comment “YcbCr” is not found in the Leadtools file. Neither the “two strings” nor the “YcbCr” comment is found in the Xerox 7535 scan to PDF file. The “two strings” are found in the Preview print to PDF file “wh-lfbc-scanned-xerox-7535-wcpreview.pdf”. The “YcbCr”label is also found in the Preview PDF.

    The first sentence fits in well with what I discovered last week. Brother scanners (I tested several different models) have the JFIF comment “LEAD Technologies Inc. V1.01″, and no “YCbCr”, but does have the “two strings” in the Huffman tables. Note that Hermie is incorrect about the “two strings” or the “YCbCr” comment not being found in the Xerox scan to PDF file – it is there, but must be deflated.

    The ordering of Flate decoded images vs DCT decoded images is different between the two PDFs. Nevertheless, the PDF Tree Structures are very similar. The Leadtools PDF file has the aforementioned “two strings” in each of the DCT decoded JFIF images. The locations of the “two strings” is not in any Huffman Table. None of the three DCT decoded JFIF compatible images has Huffman Tables. Most likely this is because Leadtools has used their proprietary JFIF compatible compression algorithms.

    I found this statement to be rather odd. So I, too, downloaded the LEADtools evaluation. And upon inspection of the PDF, I found that Hermie, not surprisingly, was mistaken about the “two strings” not being located in a Huffman table. In each case, the “two strings” were indeed located inside a Huffman table. However, this highlights a major difference between the Xerox/Preview/WHLFBC PDFs and the LEADtools/Brother PDFs. In the first group, the Huffman tables are defined in separate DHT markers. However, in the latter, the Huffman tables are defined in the same DHT marker. It is possible, but not required, to put all of the Huffman tables in a single marker – the choice to do so can act as a fingerprint, much as inserting a comment marker can.

    I’ll also note that the PDF structure between the two groups is almost completely different, but the structure of the Brother scans are very similar to the example LEADtools pdf.

    So this is good evidence that Xerox WorkCentres do not use LEADtools as third-party vendors for their pdf compression.

  53. Note that Hermie is incorrect about the “two strings” or the “YCbCr” comment not being found in the Xerox scan to PDF file – it is there, but must be deflated.

    He refuses to acknowledge this embarrassing oversight on his part.

    So this is good evidence that Xerox WorkCentres do not use LEADtools as third-party vendors for their pdf compression.

    Yeah… But Hermitian rarely lets evidence and experiments change his opinion… At least that’s how he has been so far…

    Which makes it so much fun debunking his claims.

  54. NBC says:

    August 23, 2013 at 18:54

    “”PDF images are assembled from bottom to top in your screenshot not from top to bottom as you imply.””

    “That depends on your perspective. The commands executed start with the masking layer, then it paints the background and then the foreground. If it were the other way around, the text would have been covered by the background.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Nope ! Not perspectives but facts. The PDF file is always interpreted from the end of file back to the beginning of file. The interpretation always starts with the cross-reference table which is at the end. This allows the particular application to process the objects in any order without affecting the final image.

  55. NBC says:

    August 26, 2013 at 01:22

    “Note that Hermie is incorrect about the “two strings” or the “YCbCr” comment not being found in the Xerox scan to PDF file – it is there, but must be deflated.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Nope! one just has to extract the JPEG. I thought you knew that. But then one can never know for sure with NBC. He’s one slippery dude.
    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    “He refuses to acknowledge this embarrassing oversight on his part.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Not an oversight at all. One just needs to know the steps to extract it from the PDF.

    But my initial comment stands. The strings do not appear in the Xerox scan to PDF file. However, they do appear in the extracted JPEG. But then we already knew that.

    However, the two strings do appear in the LeadTools example PDF. That’s an important fact because it proves that LeadTool’s compact PDF technology is totally compatible with JFIF. Therefore any application such as Adobe Reader which can render a JFIF file can render the LeadTools compact PDF which completely simulates JFIF.

    This is proof that LeadTools has on-the-shelf capabilities that are beyond those of Xerox in all aspects of digital image technology.

    Obviously, if LeadTools can simulate JFIF then they can also do JFIF.

    The layer structure in Illustrator of the LeadTools sample PDF is identical to that of the WH LFCOLB.

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    “So this is good evidence that Xerox WorkCentres do not use
    as third-party vendors for their pdf compression.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHH
    But you haven’t a clue where Xerox got it’s software from. One thing is obvious. Xerox doesn’t develop its on software at PARC.

    I have actually produced more evidence that LeadTools provided the Xerox software than you have produced that Xerox produced it.

    You have produced zero evidence that Xerox has done anything with MRC except for a few old papers and several patents.

    Ask yourself ! — How come Xerox doesn’t have exclusivity on MRC.

    How come LeadTools has on-the-shelf MRC capabilities far beyond Xerox’s ?

    HHHHHHHHHHHHHHHHHHHHHHHHHHHH

  56. The PDF file is always interpreted from the end of file back to the beginning of file.

    I see you have read the specification. Yes, you are correct, and for faster loading one can further optimize the layout. However, the order in which operations happen still matter when they are all captured in one Object.

    The rendering of the images is controlled by one object which first paints the background and then the bitmasks, any other way and the background would obscure the bitmask. That should be trivially self evident.

    Note that Hermitian does not disagree with my claims but rather resorts to a non sequitur about rendering from the back… I doubt that our friend has ever properly followed the flow taken by a PDF renderer…

  57. ROFLMFAO! Hermie just doesn’t get PDF, does he?

    He tries but so far, with very limited success.

  58. But my initial comment stands. The strings do not appear in the Xerox scan to PDF file. However, they do appear in the extracted JPEG. But then we already knew that.

    In other words, our friend does not know how to properly extract the JPEG, which is trivially simple. Just reverse the FlateDecode on the object and save the remaining object as JPG. That will do it.
    I have shown him the results, I have provided him with the tools and yet he remains incapable of doing this simple test.

    You have produced zero evidence that Xerox has done anything with MRC except for a few old papers and several patents.

    Other than the patents and your failed search for Xerox papers. Man you are not very well versed in doing a proper internet search now are you as there are some very relevant documents.

    As to Xerox, PARC and software, he appears to be quite unfamiliar of Xerox as a document processing company. That I find not too surprising.

  59. Herms resorts to trolling, conceding the debate. About time!

    He is now sharing with us his inability to do a proper search for Xerox and their work on MRC compression. How embarrassing… This is getting just too depressing, even for me, to have to continuously correct him.

    He still has not been able to properly extract the jpeg in the Xerox raw PDF… A trivial process, but of course prone to errors when using high level tools which tend to rewrite the JPEGs.

    I have shown him how to do it, but so far… No success…

    Those researchers nowadays…

  60. But my initial comment stands. The strings do not appear in the Xerox scan to PDF file. However, they do appear in the extracted JPEG. But then we already knew that.

    In other words, they’re there, you just have to deflate it first. At least we finally got you to admit it, after you spent two months declaring that they weren’t in the extracted JPEG.

  61. In other words, they’re there, you just have to deflate it first. At least we finally got you to admit it, after you spent two months declaring that they weren’t in the extracted JPEG.

    Could Hermitian really admit to being wrong🙂

  62. Hermie said

    But my initial comment stands. The strings do not appear in the Xerox scan to PDF file. However, they do appear in the extracted JPEG. But then we already knew that.

    How is that for parsing words? I can paraphrase for him: “My initial comment that they are not there stands even though they are there.”

  63. Hermitian has trouble admitting to the fact that I was correct about the embedded JPEG. Took him quite some time though….

    When will he ever admit that he is wrong? Many times now has he been corrected on issues, but nothing suggests that he even acknowledges this.

    He is still touting his affidavit which talks about issues that have long since been shown to be created by a simple workflow.

    Fascinating… He also has accused various people of being the forger or collaborating with the forger, even though he has no evidence to support his foolish notions.

    He makes for a very interesting person to observe in action though.

  64. NBC: …he appears to be quite unfamiliar of Xerox as a document processing company.

    Shocking… Do you mean that, went I to the Xerox room to Xerox a document, Xerox may have been one of the document processing companies available?

  65. Hermitian may have been confused about his failed attempt to search for Xerox and MRC publications or patents.

    Funny… Why would Xerox even care about scanning software🙂

    Impeccable Hermitian ‘logic’…

  66. “Hermitian may have been confused about his failed attempt to search for Xerox and MRC publications or patents.”

    Whoawhoawhoa … holdup. Is he claiming to be unable to find such things? After I and many others provided them to him countless times?

    Set preferred reality to … “selective” …

Comments are closed.