ABC News PDF and the AP JPEG

As I have shown already, a simple statement converts the AP JPEG into a 200 ppi version, with YCbCr4:2:2 and the exact Quantization matrix.

Of course, our friend insisted that ABC news would not be using such a freetoy, ignorant perhaps of the fact that the code is based on the IJG software (Independent JPEG Group) and the same Quantization Matrix also shows up in Irfanview (Windows), GIMP (Windows/Mac) and some other packages.

What was also interesting that I fed the Xerox JPEG to the same software and it failed to find any matches for the quantization signature… This further strengthens the findings of the embedded comment and the quantization matrix as being unique to Xerox or at least a reliable detector.

And so we continue to chip away at the few issues that our friend tries to raise to support his ‘hypotheses’, and which continue to fail ever so slightly…

For helping strengthen the workflow, we should recognize Hermitian as a wonderful contributor.

The tool I am talking about is jpegsnoop and I have run it against:

AP110427018673.jpg  – Save as 09

  • Photoshop

ap_obama_certificate_dm_110427-000.jpg – Quality 77

  • IJG Library  
  • GIMP
  • IRFANVIEW
  • Others

4db82608b486f-000.jpg- No identification

birth-certificate-long-form-000.jpg – No identification

24 thoughts on “ABC News PDF and the AP JPEG

  1. The METADATA for the Muscatine Journal PDF file “4db82608b486f.pdf.pdf” indicates that it was created with PDF 1.3. The METADATA of the “4db82608b486f.pdf” indicates that it was created with PDF 1.4

    My suspicion is that the differences between the .pdf version and the .pdf.pdf version were possibly due to the differences between the Windows and MAC OS versions of Adobe Photoshop CS2. Does the Windows version of CS2 write Photoshop PDF 1.4 files and the MAC OS version of CS2 writes Photoshop PDF 1.3 files?

  2. The METADATA for the Muscatine Journal PDF file “4db82608b486f.pdf.pdf” indicates that it was created with PDF 1.3. The METADATA of the “4db82608b486f.pdf” indicates that it was created with PDF 1.4

    Nonsense.. Both are PDF 1.3

  3. Oh and Hermitian, you still have to explain your claim that the WH PDF somehow requires two mask releases…

    Were you wrong? If not present your case…

  4. The METADATA of the “4db82608b486f.pdf” indicates that it was created with PDF 1.4

    The METADATA of the “4db82608b486f.pdf” indicates that you overwrote some of the METADATA when you downloaded it, and one of the things that gets overwritten is the PDF version.

  5. W. Kevin Vicklund says:

    September 10, 2013 at 17:20

    The METADATA of the “4db82608b486f.pdf” indicates that it was created with PDF 1.4

    The METADATA of the “4db82608b486f.pdf” indicates that you overwrote some of the METADATA when you downloaded it, and one of the things that gets overwritten is the PDF version.

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    That’s an indefensible claim on your part WKV because I have now downloaded your weird .pdf.pdf. version multiple times with the same computer settings and not once was the file METADATA changed. These copies of .pdf.pdf version were downloaded using the same exact procedure as I had used previously for all the downloads of the .pdf version. The PDF internal dates of the .pdf.pdf (at code line level) are not affected by the file download dates. You can do this experiment yourself. You could also do the same thing with the .pdf version if you had it.

    You don’t have the .pdf version but I do.

    For you to also have it you will first have to admit that there are two different versions.

    In the mean time, I will be posting lots of other differences between the two versions.

  6. For you to also have it you will first have to admit that there are two different versions.

    I have both versions and they are identical… Have you messed up again my friend?

    shasum 4db82608b486f.pdf
    24d1f83ecbac0de8957ccebccfc7932fda2a53d4 4db82608b486f.pdf

    shasum 4db82608b486f.pdf.pdf
    24d1f83ecbac0de8957ccebccfc7932fda2a53d4 4db82608b486f.pdf.pdf

  7. Our poor friend is just struggling with his tools, his memory and refuses to admit to his list of mistakes…

  8. Give it up Hermie. You’ve already admitted that the METADATA from your saved version of the “4db82608b486f.pdf” is the same date as the date you downloaded it. Since we know the file has existed since April 27,2011, and the metadata of your copy gives a date that is (as I recall) over two years later, you must be working from a modified version.

    Until you can admit that, any claim you make regarding metadata is worthless, since we can’t trust that you have figured out what was wrong about your original method.

  9. That’s an indefensible claim on your part WKV because I have now downloaded your weird .pdf.pdf. version multiple times with the same computer settings and not once was the file METADATA changed. These copies of .pdf.pdf version were downloaded using the same exact procedure as I had used previously for all the downloads of the .pdf version. The PDF internal dates of the .pdf.pdf (at code line level) are not affected by the file download dates. You can do this experiment yourself.

    As recently as last Friday (or possibly Saturday, the CreateDate is somehow after the ModifyDate), your method was changing the metadata, as shown by you posting the metadata of your download of the “Obama_Birth.jpg” file.

    https://nativeborncitizen.wordpress.com/2013/09/08/ap-jpeg-and-the-abc-news-pdf-resolved/#comment-68146

    Some of the metadata that changed:

    [ xmp:CreateDate ]2013-09-07T20:33:17-04:00[ |xmp:CreateDate ]
    [ xmp:ModifyDate ]2013-09-06T13:19:24-04:00[ |xmp:ModifyDate ]
    [ xmp:MetadataDate ]2013-09-06T13:19:24-04:00[ |xmp:MetadataDate ]

    Not all of the metadata changes, btw. You need to check the XMP formatted metadata. Tip: If you’re getting a lot of xmp instead of xap for files from 2011, you’re probably changing the metadata. Also look at the XMP Core – should be 4.0 for that time period, not 5.x.

  10. NBC says:
    September 10, 2013 at 17:17
    “Oh and Hermitian, you still have to explain your claim that the WH PDF somehow requires two mask releases…

    Were you wrong? If not present your case…”

    You’ll just have to talk to the forger about that. And while you are at it you might get his opinion as to why all your Xerox scan to PDF files open in Adobe Illustrator in landscape mode but his forged Obama LFCOLB opens in portrait mode.

  11. Hermitian: You’ll just have to talk to the forger about that. And while you are at it you might get his opinion as to why all your Xerox scan to PDF files open in Adobe Illustrator in landscape mode but his forged Obama LFCOLB opens in portrait mode.

    How slow can one be?… The WH LFBC does not require two mask releases… Geez… And the reason the PDF opens in portrait is because it was opened and saved in Preview.

    This is basic stuff… Do you still not understand this?…. Wow… Or do you refuse to admit to you being wrong?….

  12. I am waiting on Hermitian to file a corrected affidavit in Taitz v MDEC.

    It would be even better if he gets called as an expert🙂

  13. “How slow can one be?… ”
    Notice how he keeps repeating things that have already been explained to him.
    Is he showing signs of the onset of Alzheimer?

  14. He does seem to forget things that have been explained to him in quite some detail. I see it more as a problem having to reconcile facts with his unsupported beliefs, where the beliefs are given much more credibility than facts and data.

    I have seen it before. Not a very pretty sight when one is interested in scientific hypothesis formation yet quite understandable.

    He has failed on so many fronts now that it will take some time to come to terms with all of this and denial is one way of ‘dealing’ with it.

  15. You’ll just have to talk to the forger about that. And while you are at it you might get his opinion as to why all your Xerox scan to PDF files open in Adobe Illustrator in landscape mode but his forged Obama LFCOLB opens in portrait mode.

    Since we have explained this to you well over 30 times and you still can’t grasp this concept, I think it is time to invoke Clarke’s Third Law:

    Any sufficiently advanced technology is indistinguishable from magic.

    It’s magic, Hermie.

  16. Since we have explained this to you well over 30 times and you still can’t grasp this concept, I think it is time to invoke Clarke’s Third Law:

    ROTFL… Hermitian is still stuck on issues that have long since shown to be his own ignorance. The WH LFBC PDF does NOT require two mask releases, contrary to his claim, preview DOES change the way Illustrator interprets the pdf and the Muscatine pdf and pdf.pdf are one and the same file.

    Love it…

  17. Hermitian remains unwilling to admit that

    1. He was wrong about the WH LFBC requiring two mask release operations. Several people have observed it takes the same as the 7535 created version
    2. He fails to comprehend how preview files are rendered correctly in illustrator, explaining why the image shows up in portrait.
    3. He still insists without any evidence that there are two muscatine pdf’s, one with a single extension .pdf the other with double .pdf.pdf… He is correct that these exist and are the same file, he somehow believes that they are not…

  18. Hermie: You’ll just have to talk to the forger about that

    I tried to discuss it with a relative of the forger, but they don’t hold up their end of the conversation and people look at you funny when you’re talking to the copier.

  19. Hermitian has decided to not respond to the questions I asked and instead, in his usual ignorance, has accused me of cherry picking and ignoring him.

    He is right, he no longer will be contributing on this blog. Let’s see how long it takes for him to figure this out…

    As to the pdf.pdf and .pdf file, they are identical files. I bet you that the server software was updated which added a double pdf.

    Hermitian’s understanding of technology is someone limited.

  20. Time to turn to the halo/x-ray issue. I think that these are two sides of the same coin. When the background is a mostly solid color, it fills in behind the lifted text with the background color. However, when it is a light and mottled color, it ends up creating a halo, because there is no clear color choice. This then gets exacerbated when the image gets downsampled later in the process.

    I’m not certain how it will handle a dark and mottled background color.

  21. You are probably correct, the two are related. I see you have done the same analysis I have done🙂

    Hermitian could learn something about doing proper research.

Comments are closed.