Someone under the name e.m. cadwaladr dropped by the blog and left in comments a link to a website that has yet another analysis of the Obama long-form birth certificate PDF. Rather than argue like Mr. Irey that the document is a fake because letters are different, cadwaladr argues that it is a fake because they are identical, something he calculates as astronomically improbable.
I think, though, due to the ubiquity of computers and imaging software, that the number of people making claims about what they found tinkering with the bits of the long form will eventually exceed anyone’s capacity to respond.
Update:
I used Adobe Acrobat 9 to export the bitmap layer cadwaladr discusses and I compared the two occurrences of the name “OBAMA”. The claim that the two letter B’s are identical just didn’t hold up when I zoomed in. First I show cadwaladr’s comparison image (zoomed 200% with Ulead PhotoImpact):
cadwaladr image, zoomed 2X by Ulead PhotoImpace
The letter B, alone among the letters, looks identical. Now here’s what I got, zoomed 400% with Ulead PhotoImpact on the bitmap layer that I exported:
My results, exported from Adobe Acrobat and zoomed 4x
What I see is that in both of my examples, all the letters differ. I cannot replicate cadwaladr’s experiment this way; however if I just use Adobe Acrobat Reader to zoom in, I get results like cadwaladr’s. This is curious. My only observation is that an algorithmic process is more likely to create similarity (through simplification and smoothing) than it is likely to create difference. Unfortunately I don’t know how cadwaladr extracted his images.
I have found a problem with the Obama long form birth certificate PDF that, frankly, is a little more conclusive that Mara Zebest’s. And, no, I’m not a “birther”.
http://cadwaladr.blogspot.com/2011/05/president-obamas-long-form-birth.html
Thanks,
e.m. cadwaladr
Are you kidding – layers. Where have you been in a bunker somewhere. All of this has been resolvedalong time ago.
Go to Hacker Factor Blog
http://www.hackerfactor.com/blog/index.php?/archives/428-After-Birth.html
gorefan
No, NOT layers. My evidence does at least require that you have the attention span to read the article.
Sack of cack remains sack of cack, bugger of “cadwaladr” you’re a birther and not even an interesting one.
Bovril:
Ad hominem isn’t much of a refutation either. Anyone out there with both a brain and an attention span?
You claim the last name is different on the original and the pdf. Yet”
1. The State of Hawaii, whhich actuually has the ooriginal, says they are identical.
2. Photos taken of the paper document at the White House and a high-resolution jpg show the same last name as the pdf.
3. Most egregiously you have no controls; i.e.,an unbiased sample pdfs made from 1960s typed documents . No conntrols = no science. A doctor who treats a single patiient cannot claim a miracle cure.
I could go on for hours, but that should suffice
Why is it sooo gd difficult for birthers to get it through their thick skulls that what is presented on line has absolutely no relevance? The only things of import are the physical pieces of paper that the HI DOH created, signed & sealed (COLB & LFBC). Everything else is hogwash, especially all the analysis and discussion. In addition the HI DOH officials & two governors of two different parties have attested to their veracity. Give it up!!
If you’d like, I could add that the index of Hawaiian births for 1961 shows the name as Barack Hussein Obama. Surely you aren’t suggesting there were 2 kids with that name born in Hawaii in 1961?
Given the provenance and certification of the contents of the original document and its accuracy by the state of Hawaii, your implied accusation of fraud and malfeasance by the state of Hawaii is a serious charge.
“Scientist”…
I took the PDF from the White House web site. I do not claim to know who altered it or why, nor can I speak to the claims of other parties about other documents.
Certain occurences are straightforwardly improbable and do not require controls. If one finds a body on the sidewalk with a bullet through its head, one does not need to examine a few dozen demographically similar people to determine whether or not bullets just naturally occur inside the cranium.
I don’t believe anyone altered it. Since a high resolution jpg is avaiiable, why would you work friom the pdf anyway?
You haven’t convinced me that this is the case with this document. Not at all. I simply do not accept your conclusions. You make a bunch of statements that such and such is true, but you don’t support them. If I got your “paper” to review for a scientific journal, I would reject it in 10 seconds.
Include controls. Any marginally competent scientist would. If you don’t, I can only conclude that you are afraid they will invalidate your conclusion.
Majority Will…
Believe it or not, I have not been not avidly interested in the course of this whole issue. I simply examined the PDF, initially not expecting to find anything. Based on what I know about the technology, I found something I cannot easily explain. I don’t have an agenda. If someone can find a more probable reason for what I found than malfeasance — please do.
Yes, you found nothing. The document is perfectly fine.
Arrch the formatting of the above post is all messed up. I started typing in the middle of the quoted post and didn;t motice before I replied. I hope people can make sense of it
“Scientist”
I don’t know the provenance of the high-res JPEG. I know the provenance of the PDF, since I downloaded it from the Whitehouse web site myself.
I have seen this control argument before. I have neither the resources not the access to to gather control documents — as you could reasonably infer.
On the other hand, if you insist on controls in all circumstances, then you will have to reject a good deal of science. Many things in palentology rest on a statisically inconclusive handful of bones.
I think my evidence would be presentable in a normal court case. However, you may not agree. I give you credit for at least intelligent ad hominem.
You missed the point spectacularly and entirely.
And I wasn’t referring to a PDF, a technology which didn’t exist in 1961.
You know the providence of the savannah Guthrie photo. She took it. Go have a look.
Evidently, you don’t know either. Are you trapped inside the White House?
Do you know what provenance means?
What efforts have you made? (Likely none). When controls exist you have to include them. Otherwise keep quiet and let those who include them speak instead.
It’s one thing if controls don’t exist. You know that controls for your study do exist.. To not do them is simply bad science.
Are you joking? What credentials do you have to establish that you are an expert witness?
By the way, here is a high-res jpg. http://the305.com/2011/04/27/good-lord-i-hope-this-finally-ends-the-whole-obama-birth-certificate-issue/
Who cares what you find in a lousy image when a good one is avaiilable?
Stop with the excuses and do at least a semi-competent job, then you can come back here,.
And if your scenario occurred in a hospital in Oahu in 1961, you wouldn’t search the White House for forensic evidence 50 years later.
Well, it is late here and I have a race to run tomorrow. It has been interesting attempting to argue with three people at once. Not fun, exactly, but interesting.
You will probably all have the satisfaction of knowing that, no matter what evidence is presented, nothing will be done. We departed from any real pretense of the rule of law during the Bush regime, and are unlikely to return to it any time soon.
Don’t fume unnecessarily.
– e.m.c.
I certainly wouldn’t do anything with what you have presented. Really, what would you suugest anyone do with such shoddy work?
Scientist
I did have a quick look at your “hi-res” JPEG. It is not “hi-res” enough to analyse. It’s not even legible.
First let me say that I have not seen, nor verified your notabirther certificate. However, putting that question aside for the moment, I make these observations (which I also posted on your website):
I also do not understand your remarks regarding the “provenance” of the Press Release scan versions. You can download them from the news organizations (e.g. the AP) directly. That’s provenance. It certainly sounds like you are making an excuse to ignore an important piece of evidence, which goes to your credibility as an unbiased observer — as does the unsupported remark you made which goes against what another expert said. If you want to pursue this, then get the file from the AP web site, or use the handy copy here:
http://www.obamaconspiracy.org/wp-content/uploads/2011/06/BirthCertificateHighResolution.jpg
This is the link Scientist was thinking of, I believe
http://www.theobamafile.com/_images/BirthCertificateHighResolution.jpg
If there were anything, anything at all, the Press would have been all over it within 24 hours. Take comfort in that.
When and where did you analyze the original, typewritten and signed Certificate of Live Birth form from 1961 that’s stored in the archives of the Hawaii DOH that’s certified to be accurate by the state of Hawaii?
An argument by analogy isn’t valid unless you can validate the analogy, which you haven’t. At best you have argued that there exists some observation for which controls are unnecessary, not that YOUR observation needs no controls.
Given the incredible improbability of your hypothesis, I think you ought to do a little more than hand waving about the critical assumptions you’re making.
State of Hawaii links to that Obama BC and they say
“On April 27, 2011 President Barack Obama posted a certified copy of his original Certificate of Live Birth.”
Notice the word “certified”
http://hawaii.gov/health/vital-records/obama.html
.
Plus Krawetz, a real expert, states “Before I begin, I need to point out two critical items for this evaluation. First, digital document analysis can detect manipulation, but it cannot determine whether the original subject is authentic. The authenticity can only be determined by the State of Hawaii, and they already said that it is authentic.”.is that clear enough you, or is fiddling with PDF’s your joy?
In other words, e.m., what everyone is trying to tell you is that your analysis adds nothing to the discussion and does little more than waste everyone’s, including your, time. The bottom line is that the state of Hawaii has confirmed that the birth certificate is authentic. As I’ve said before, it’s like accusing the President of counterfeiting money with collusion from the Treasury. You MIGHT be able to prove that there was a conspiracy with Treasury, but I don’t see how you could convince me you could prove the bills printed were “forgeries.”
If the Treasury prints the bills they are genuine.
If the Hawaii DOH prints the COLB it is genuine.
When I tried to replicate cadwaladr’s results, I couldn’t. I will need more info on his methodology. The two letter B’s were quite distinct. See update to main article.
“The claim that the two letter B’s are identical just didn’t hold up when I zoomed in.”
Cadwaladr is not the first to make this claim.
I asked Dr. Krawetz about the claim of identical boxes and lettering back in early June. Which is why I suggested that he visit Dr. Krawetz’s site and discuss it with a real expert.
If he is honest about wanting to get to the truth, we should seeing him posting there very soon.
I told cadwaladr on his blog that he should compare the probability that all the external evidence supporting Obama’s name at birth being “Obama” with the probability that he made a mistake in his analysis.
Cadwaladr’s premise is based on a very common misunderstanding of probabilities.
He apparently looked at all of the different letters on the PDF and found two that appeared to him to be identical. He then assumes that the odds against two letters being absolutely identical are astronomical. (We don’t have any actual evidence of what the odds really are). From that assumption, he declares that the letter must be fake.
But this is a common fallacy, and would lead to the conclusion that no one has ever won a lottery, because the odds against winning are so high. But of course people are always winning the lottery, because the proper question is not “what are the odds of one person winning the lottery?”, but something like, “out of 20,000 lottery entries, what are the odds that one is a winning ticket?”
If I simply look at typewritten characters on the PDF, ignoring punctuation and ignoring pre-printed characters, I get a count of 277 characters. Out of the 277 characters, cadwaladr has found two that he claims are identical. So the probability question is: what are the odds that out of 277 characters, there will be 2 that are identical.
I raised this issue before, it’s similar to the birthday problem. There is a point where probability tips, but it takes a lot more information and a lot more math than I am capable of to figure out the answer.
But the other fallacy is simply to assert that because something is improbable, it didn’t happen. Improbable things happen all the time — it is when separate improbabilities are piled on that you can start to assert that the improbability itself is proof of falsehood. But again, you’ve got the asserted match out of 2 out of 277 printed characters. If you reduce the analysis to only letters that are the same (ignoring for the moment that the fact that they are renditions of the same letter increases the likelihood that they will be rendered identically at lower resolutions) — we have 4 B’s, 12 A’s, 2 R’s, 3 C’s, 6 K’s, etc. (and that’s just counting the capital letters which appear in “BARACK”).
I’d note that I also cannot replicate the supposed “identical” rendering of the B’s on my computer. I’d note that for purposes of his example, cadwaladr has reduced the B’s to 2 colors (black & white), eliminate all of the gray scale that is on the PDF. Even when I do that it doesn’t replicate what cadwaladr claims to see, but I suppose there are lot of different ways you can tweak the algorithm until it produces whatever it is you want it to, since a choice has to be made as to which shades of gray and green will be treated as black, and which will be treated as white, for purposes of the 2 color rendering.
I looked at a photo of Hitler on line and one of Jerome Corsi.
Under magnification, it turned out that the pixels are the same. I ask: what are the odds?
The answer is obvious: Corsi is Hitler.
@ cadwaladr: you are an idiot. The imperfections of, or oddities in, an electronic copy of a document carry absolutely zero value when it comes to asserting the veracity of the document in question. It is the original and its provenance that must be examined. In this case both have passed with flying colors.
I opened the White House PDF (dl’d a fresh copy via OCT’s link just for this exciting test), and opened it in plain old Acrobat Reader, and zooooooooomed way in, all they way to 6400%, until each pixel was about ¼” sq. on my display. Let the bitmap object resolve, and saw the letters e.m. compares, the two B’s and two I’s, in the positions he notes, do match each other pixel for pixel. Took screenshots, pasted them into Photoshop, overlaid one B over the other, one I over the other, just to confirm.
And, with all the other visual oddities being fixated over, the next question is …. so what?
The next step is research and experimentation to search for possible explanations, eliminating possibilities, to establish some likelihoods. Instead, e.m. jumps to assumptions and poses them as conclusions.
Identify the software used to create the image, the workings of processes they were subjected to, and how they are being rendered for display, is essential to estimating the likelihood of two identical renderings of a given glyph. With enough letters for a given process at a given resolution, it will happen. The cruder the process and the lower the resolution, the more likely it becomes.
Or, we could pull up short and say again … so what? Look at the other images available of the copies of the document presented at press conference. The letters in those images show differences. The low-res, bitmap rendering enhancement on the PDF simplifies visual information significantly, ameliorating differences, increasing the likelihood of such coincidences. But again, other images imply this is a curiosity with no significance that bears on the authenticity of the document it was created from.
Three possible assumptions I can think of:
1) Since no other pairs match (that I noticed, certainly not going to compare them all for reasons above), these are genuine coincidences.
2) For some incredibly weird reason, these letters are duplicates, created by a user or by a process. What possible significance would this have on the information in the document? Since the bitmap rendering is an addition to the image of another document, and the apparently duped letters match those on that document, there is no significance. What, the document creator wanted prettier I’s and B’s?
3) The document was built at least partially, letter-by-letter, and the artist forgot to alter an ‘I’ and a ‘B’. Again, since the words/letters reproduced in the bitmap rendering match those in other images of the document … this is just plain silly. Since the other images are clearly not made from prints of this PDF, but rather the document the PDF was made from, any finagling, real or imagined, in this PDF is without meaning.
Lastly, I can’t believe e.m. dismissed the use of controls as “the control argument”!
Sorry for rambling. It’s the PDF madness!
Can make reproduction available of e.m.’s folly if interested.
PS – The reliance on screenshots is potentially problematic, as PDFs are compressed. Appearance of the bitmap object may change depending on software, and will change based on magnification level, which is why I chose max. zoom, and checked both Illustrator and Reader, as those are the likely candidates for e.m.’s use. He did after all, admit to watching at least one of “those” YouTube videos!
OK, I see what you did to replicate the illusion.
I say “illusion” because when you use Adobe Reader to magnify a document on screen, you aren’t seeing what was in the document at time of the scan, you are looking at what the Adobe algorithm does to change (augment) the image as you scale it up. So the fact that Adobe renders the B’s identically under those circumstances is really irrelevant to what is in the document itself. Adobe does whatever it does, but as soon as you are magnifying above 100% you are looking at new stuff that Adobe has added, rather than whatever was there in native format.
No, I’d have to disagree. This is rastered information, the pixels aren’t being split, just magnified, and I have not resampled the image. And the rendering matches across Adobe products, so it isn’t just Reader.
I was exploring how e.m. saw what he saw, and also dismissing it as irrelevant. That is to say, e.m. saw what e.m. saw, and you can too. The significance e.m. assigned is the illusion! Any significance drawn from the PDF is illusory.
Hey, they match in Corel X5, and when viewing the PDF in FireFox’s native PDF viewer. All well and good, but still short on significance. Hope e.m. can sill us in when (if?) he returns. Pulling in Expelliarmus’s excellent piece about probability. I say stuff happens, and again, “So … what?”
I think this is because all Adobe products are written using Adobe code.
Expelliarmus is right- Try using non-Adobe products, and the results won’t look the same.
They match across numerous software platforms, most likely these are all using similar or identical algorithms. It’s late, I know I’m flogging, but a thorough refutation should let the refutee know he was taken seriously, and exactly where he went wrong. e.m. isn’t lying (well, about the B/I he saw anyway), just misinterpreting.
The real offense is then attempting to spreading that misinterpretation. Especially by resorting to the blog equivalent of a “drive-by”.
I just had a look at the LFBC using Foxit PDF Reader and they don’t look the same to me.
For years, the birthers have been screaming that your Birth Certificate has to be more than an image. So, he, again, provided certified copies and showed them to the press and Hawaii vouched for them. To allow others to see it, his staff scanned it. Now, the birthers are spending all their time attempting to debunk the image as if it was the document. Haven’t they been saying an image doesn’t count?
I am a little surprised that they have not responded to comments here with, “well, you guys tried to debunk Lucas Smith’s image.” However, there are some key differences:
1, The LSBC (Lucas Smith BC), had no providence. No government agency vouched for it.
2. The chain of custody is missing has Lucas never once provided evidence he was every anywhere near Kenya (or Africa for that matter).
3. I do not recall anybody seeing his “original” document except maybe those in the back of the truck in the Dominican Republic (or somewhere similar).
4. Most debunking attempts here consisted of issues on the actual document, not how it was scanned (date format, misspelled names, odd sized child, lack of documents for comparison, etc.).
With Obama’s long form, all the birthers have are scanning issues and some made up items that, when compared with BCs from the same period, show it to be consistent.
While I think we both think this is unlikely, it is certainly more likely than the speculative conclusion that cadwaladr reached, that Obama’s birth name wasn’t Obama, Even cadwaladr can’t explain how that is possible given the external evidence (Birth Index, newspaper announcements, divorce decree, COLB).
When confronted with these wild speculations, one really needs to exhaust normal explanations before jumping to implausible ones, and just as with the initial observation of “layers” it was found that this is completely normal based on the normal algorithmic process of PDF creation.
What I find so ‘amusing’ is that birfers have no problems in claiming that the scanned image of the POS LSBC is a valid BC even though they have never held it in their hands. There’s so obviously wrong with the way their brains work.
Can you export the layer to a bitmap in Illustrator?
You can always tell at a glance which comment threads involve birthers, because those always have many more comments than the ones that don’t.
I’m just going to go ahead and say the LFBC is fake because of what appear to be a pencilled in “X – X” above the “Twin” and “Triplet” boxes in field 3. Everyone knows that real birth certificates don’t have that.
My qualifications as a document expert:
— I have set up scanners for users at corporations.
— I have installed Adobe Reader and Adobe Acrobat on computers.
— I have used scanners and Adobe.
— I know what the acronym PDF stands for.
— Most importantly, I am willing to say whatever you want me to say for money and / or attention.
e.m. cadwaladr is not a birther. He says so himself in his first post. 😉
Birthers, please ask yourself: if you were already President and no actions were active in the senate or house to impeach you or even question your validity as president, would you then produce and release an easily debunked fraudulent BC?
And if it IS so easily spotted as a forgery, why have no high level republicans yet called for any investigations?
Wouldn’t a GOP rep or senator have a lot to gain from proving this BC is fake, especially with an election coming up?
Why call attention to yourself if you’re a fake, but already have the job and no one in power is questioning you over it?
That would be like robbing a bank, getting away, then calling the police to tell them you didn’t do it and please don’t come over to look in a shoebox in your closet.
>>>Scientist
1. The State of Hawaii, whhich actuually has the ooriginal, says they are identical.<<<
Sir,
You made the above statement. Can you direct me to the website that has an Department of Health official that makes the specific claim that the released’ document matches the internet document. I need to be specific here, not hearsay from a reporters interpretation, but an actual unequivocal statement that verifies the document provided on the internet is an match to that released. Thank you for any assistance you can provide with this chain of evidence search.
Pete
The state of Hawaii has repeatedly stated that they have Obama’s original birth certificate. Now some what to claim that the copy of Obama’s birth certificate is a forgery. Talk about a lack of attentions span.
For those paying attention to my search for chain of custody and verification of the long form birth certificate as authentic. See the below website.
http://www.wnd.com/index.php?fa=PAGE.view&pageId=297985
Please note that Hawaiian DoH official REFUSED to confirm long form Birth Certificate as authentic.
Be that as it may, I am going under the assumption that perhaps they didn’t want to give confirmation to WND, because of political reasons. However, thus far I am only able to demonstrate that Hawaiian DoH official HAS REFUSED to authenticate. I am still looking for a DoH official statement that the copies made are identical to those posted by the White House.
AnotherBird, your statement above is acknowledged, and Hawaii has said they have his ‘Original Birth Information’, I can confirm this via DoH representatives. They do not say that the White House long form Birth Certificate, released in April matches all information identically and is a true copy. However, if you can provide a reference to such a statement, it would be greatly appreciated.
Pete
Pete, you are not being accurate. They REFUSED to talk to WorldNetDaily at all. The DoH has made it very clear that they stand by the statements they have released. Since WND has promoted the idea that the DoH is hiding something or even lying I am sure anyone from WND is on the list to be given a big FU. A reasonable approach to take, yes?
Have you ever read Mr Onaka’s stamp? Quit being silly.
http://hawaii.gov/health/vital-records/obama.html
“On April 27, 2011 President Barack Obama posted a certified copy of his original Certificate of Live Birth.”
What more do you want, hula girls and dancing Hyena’s?
AND THERES MORE!
http://hawaii.gov/health/vital-records/News_Release_Birth_Certificate_042711.pdf
“HONOLULU – The Hawai„i State Health Department recently complied with a request by President Barack Obama for certified copies of his original Certificate of Live Birth,
which is sometimes referred to in the media as a “long form” birth certificate.
“We hope that issuing certified copies of the original Certificate of Live Birth to
President Obama will end the numerous inquiries related to his birth in Hawai„i,” Hawai„i Health Director Loretta Fuddy said. “I have seen the original records filed at the Department of Health and attest to the authenticity of the certified copies the
department provided to the President that further prove the fact that he was born in
Hawai„i.” ”
That should be enough for anyone. Read the links and stop believing WND.
I seem to have created quite a stir. I have read through the comments. Some raise good points, including a few I had not considered. Others are simply not worth acknowledging. I will attempt to answer at least a few of the questions that have come up.
First, I am not an idiot and did not base my letter comparisons on screenshots. I exported the layers out Adobe Illustrator as individual uncompressed TIFF’s. For comparison purposes, I superimposed the letters in Adobe Photoshop. I did not resample at any point in the process.
Second, when I made my analysis I did focus solely on the PDF document. If other electronic documents were available at that time I was not aware of them (with the exception of the short form). My assumption was that other documents that might be floating around the web could not be verified as reliable. They might have been tampered with by anyone, including, obviously, Obama’s enemies. The PDF was posted on the Whitehouse site. Strictly speaking, I will acknowledge that even this does not prove intent on the part of the president. It is not inconceivable that staffers can be bribed.
At the request of several commenters, I have now reviewed the notorious “High-res” JPEG. If this had been posted on the Whitehouse site, I would probably not have questioned it’s authenticity. This does not mean that it’s authentic (frankly, I do not believe there is any such thing as an unforgible electronic document) but nothing about it is particularly suspicious. It deviates from the PDF in quite a number of respects, but this is to be expected with two documents that are at least two generations apart.
Looking at all the various confirmations by this official entity and that, the presumed sequence of events, etc — I have to admit that an sort of conspiracy is improbable. I don’t “like” conspiracy theories, whatever my detractors may think. I am, however, still left with the question of which is more improbable — tampering by someone, for reasons unknown (I stated in my article the illegitimacy theory was just that — a theory) or what still appear to be some very improbable irregularities on the PDF.
It may be that the PDF irregularities I have focused on are artifacts of processing, but it would probably take an expert on Adobe’s code to prove that. Experiments could be performed, I suppose — but I do not have access to the equipment used.
It may be that my calculations are wrong (I admit they are very general, and I myself can think of one or two other factors that might enter into the matter) — but I have not seen anyone offer more accurate and specific calculations. They are better, I think, than the usual run of “this does not look write to me” objections.
The problem with the relentless chant of “controls, controls” is that neither I nor any of you have access all the equipment, the software, and statistically relevant samples of Hawaiian birth certificates of the same vintage. To demand controls, in this case, is simply an attempt to end debate. It is as much as to say, “we can know nothing, we must trust the experts who have access to the truth.” And who would that be? The networks –who are sympathetic to Obama? Fox — whose sole purpose is to destroy him? The State of Hawaii? The FBI perhaps? All such disinterested parties! Such steadfast, unflinching defenders of fact!
If you can clear up the letter matching problem in some detailed, specific way — please do! Or, take some vacation time, fly to Hawaii, collect several dozen 1961 birth certificates at random, and solve the problem experimentally.
Not without altering it by resampling in some way. As you and I and others were getting at (in various wordings) it is a compressed object, and e.m. has been distracted by one rendering of it, by relying on screenshots. And, as always, since the wording in this image (PDF) matches that of the other images, and the PDF clearly has additional processing implying other available images were not made from the PDF, then it doesn’t matter anyway. More obsession over ephemera.
e.m.,
As I’ve said before, and as I posted on your blog, all your analysis is irrelevant. Again:
It’s like accusing the President of counterfeiting money with collusion from the Treasury. You MIGHT be able to prove that there was a conspiracy with Treasury, but I don’t see how you could convince me you could prove the bills printed were “forgeries.”
If the Treasury prints the bills they are genuine.
If the Hawaii DOH prints the COLB it is genuine.
Do you despise our foundation and system of laws too?
Dr. C.
If you have time to look at the math, I would genuinely be interested.
I did not get around to reviewing the PDF of your math creditials, but I do not seriously question them. If you would like a PDF of your doctorate in mathematics, let me know. Specify the university of course…
@ e.m. cadwaladr July 2, 2011 at 3:19 pm
“First, I am not an idiot and did not base my letter comparisons on screenshots. I exported the layers out Adobe Illustrator as individual uncompressed TIFF’s. For comparison purposes, I superimposed the letters in Adobe Photoshop. I did not resample at any point in the process.”
This is better, you have intimated you process, but you still miss the flaw. The rastered layer in the PDF is already compressed. When you export a TIFF from Illustrator, Illustrator asks you to select a resolution, and a color palette. Based on your selections, Illustrator is rendering to a file as it would render for the screen, in a screenshot. The image is being resampled. If you select a high enough resolution (I just repeated your process at 600dpi), you get a result that appears to match that from screenshots.
“Second, when I made my analysis I did focus solely on the PDF document.”
Your error.
“I have now reviewed the notorious “High-res” JPEG … frankly, I do not believe there is any such thing as an unforgible electronic document”
True, but there is reasonable and unreasonable. Can you not see that, the JPG and the PDF almost certainly had a common ancestor, a photocopy or laser print onto security paper?And that the JPG, which is almost certainly a simple scan of a photocopy of that common ancestor, is free of post-processing enhancements … that the PDF could be made from the JPG but not vice-versa? That the JPG is closer to the original visual information? And that the lettering and wording in all the images match? It becomes a question of likelihoods. Since these other images were not made from the PDF, then they must all be separate forgeries, or the PDF was created from a forged JPG or their common ancestor was forged. Is that what you’re saying?
“To demand controls, in this case, is simply an attempt to end debate.”
Not at all. It is saying here is a curiosity; to find its explanations and define any meaning, that’s the next step.
“It is as much as to say, ‘we can know nothing, we must trust the experts who have access to the truth.'”
Not at all. It is saying trust to reason. Judge the probabilities on what you can know / prove.
“And who would that be? The networks –who are sympathetic to Obama? Fox — whose sole purpose is to destroy him? The State of Hawaii? The FBI perhaps? All such disinterested parties! Such steadfast, unflinching defenders of fact!”
Some biases here. Don’t let these affect your research.
“If you can clear up the letter matching problem in some detailed, specific way — please do! Or, take some vacation time, fly to Hawaii, collect several dozen 1961 birth certificates at random, and solve the problem experimentally.”
You presented this hypothesis, you are convinced it has value, and now demand others do the experimentation for you?
DC claims a Master’s, and you offer him a forged PhD? While purportedly trying to present an honest analysis?
Stunning. You try to treat someone respectfully ….
Like most state government department heads nationwide, the Director of the Hawaii Department of Health holds news conferences. There is nothing stopping any news organization from attending the next Hawaii Department of Health news conference and asking Director Loretta Fuddy if there is any difference between the Certificate of Live Birth that she issued to President Obama and the photocopied scan of a Certificate of Live Birth for the President that appears on the White House web site.
Except what reporter wants to look like an idiot for asking a question so dumb. The whole issue of custody was addressed about as clear as it can be with a national press conference first by the President, showing the actual document obtained to the press and the world and discussing the issue, which included not only publishing the birth document, but also a letter establishing full chain of custody from request.
The next day, the HI DOH provides their own leter confirming chain of custody from their side and CONFIRMING that very same document is valid. They were very clear about it and their own DOH website and FAQ explains all this and provides DIRECT links to what the White House site.
That is about as clear as you can get as attesting for authenticity. There is NO legitimate dispute. (see Suranis’ 3:09pm post for the links…I’m sick of having to restate and post the same obvious stuff over and over again to the terminally dense).
Therefore, no serious reporter would waste his question time to come across so foolish on something that has already been resolved well within any reasonable sense. He might as well ask if Alvin Onaka works at the DOH or ask if they’d be willing to release the name of who the Director of their own department is. That is the same level of resoundingly obvious *DUH* questions as to what you are suggesting.
Have you posted your theories to Dr. Krawetz’s website? If not why not?
There’s no “math” in your analysis, only faulty assumptions. The faulty assumption is
As I pointed out above, the question is NOT “how probable is it that any two letters will match” (as you have framed it), but “how probable is it that among hundreds of characters, after discarding all color and grayscale information, that two binary renderings will match?”
We don’t know what the probability is. It’s the birthday problem – or the lottery problem — and the probability of finding a single match within the document is probably fairly high, especially when you are looking at the SAME letters. The letter B, by definition, consists of a vertical line on the left, three horizontal lines at the top, center, and bottom, and two rounded loops connecting the horizontal lines on the right. So two B’s, by definition, share a large number of black pixels when rendered in binary format, probably the vast majority. So the real question, is how many different possible renderings are possible, given that if the document is in fact genuine, the two letters on the original would be in the same type font, same size. There is even less variation possible in a low resolution binary rendering of the letter I, which consists of one center vertical line, and two short horizontal lines, top and bottom. The simpler the letter form, the greater the likelihood of similarity.
That’s what the point of a “control” would be. It would potentially give you some evidence to back up your faulty assumption as to probability.
The problem with your method is that to get the B’s to be the same, you had to first run through a computer process that DISCARDED all the differences in the B’s (the grayscale that shows up when the PDF is viewed as a whole, without separation into layers).
Then you “assume” that something which might very well be an expected and common occurrence (bitmap renderings of the same letter) is “highly improbable”.
Then you go one step further and cite the evidence from the page that the other letters do not look alike as if it were evidence that makes two lookalikes more improbable, when in fact it is the other way. Going back to the birthday problem analogy, it is like first saying how surprising it is that in a room with 30 people, two share the same birthday; and then citing the fact that the other 28 people all have different birthdays as evidence the one of the two birthday sharer’s is lying.
Whatever the probabilities are of a letter match, it has to be weighed within the universe of all the letters on the page. You did not just pick two letters at random for comparison — you compared all the letters until you found a match. (And, as noted, could do so only after going through a process designed to reduce the number of pixels and color shade that were in play).
We’re not talking about snowflakes. These are letters that you have chosen to view in altered form.
And that brings us back to your other fallacy, that “highly improbable” (your assumption, not math) is the equivalent of “impossible.” Probability isn’t a measure of what can’t happen, it is a measure of the likely frequency of something that can happen.
Yes, it is here:
http://hawaii.gov/health/vital-records/obama.html
Hey Doc,
Seems like we’ve got our own “expert” — his name is John Woodman (not an outing, he states it at the outset of his first video, “Hi, my name is….”
He’s produced a series of videos debunking the all the birther claims based on the PDF.
First one here:
http://youtu.be/Hc3b8xVJStU
Here’s the link to his youtube channel:
http://www.youtube.com/user/springfieldcompguy
From your blog:
Anonymous said…
“I would certainly prefer that public officials and their staffs refrained from lying. It does, however, concern me more that the Consumer Price Index and other key economic measures are beinging systematically “cooked” than it does the president or members of his staff may have made minor alterations to his personal history. Obviously, the problems with the birth certificate do bother me enough for me to raise the issue.” -e.m.c.
Strange post.
Unit April 27th 2011 I suspect the answer may well have been “Lester Kinsolving”
I’m not even bothering reading birther posts for a while. Is he, in fact, just a concerned citizen with a few questions to a serious issue?
Then again, he was there as a “reporter” for WND. That outs him as an idiot right there.
Birther Lite, perhaps. No taste and less filling.
Yeah, so he claims. As have many who try to hide from the stink of birtherism that is all over them. When will these folks learn that if they buy into birther nonsense, the title goes with it…
Yes you are if you continue to believe that an analysis of an electronic copy of a document, any document, or a work of art, say, a painting, is going to prove or disprove its authenticity.
I have had some dealing with experts in charge of authenticating artwork in the past, and they would laugh at you and your post.
In this case the provenance is ironclad and certified by the issuing entity, so there really is no room for debate.
In other words if Obama’s BC was a Gauguin, having Gauguin say “yes i painted this and I sold it to Joe Blow in Tahiti on such and such date” would be enough for all the courts and auction houses and museums in the world. Against this, we have your clownish theories.
I can’t put it into words, but it keeps striking me as hysterical that e.m. “Not a Birther I Just Think Like One” Cadwaladr appeared so quickly after the PUMA discussion on the Zebest thread. Almost like a summoning.
I think (and it scares me) I can see the “logic” …. that if tempering with the PDF can be proven, then the entire chain of events concerning Obama’s birth certificate back to 2008(?) is automatically discredited. It all collapses like a house of cards or falls over like dominoes. Except … it doesn’t. Particularly in light of the other available images and evidence … and law and reason … the PDF is an outlier. They don’t understand it, their victims don’t understand, so they’ve clung onto it, death grip-on-a-security-blanket style.
We have all learned a valuable lesson: When prepping documents for digital release to biased parties, be sure to uncheck all the “Text Enhancement” features in your “Scan to PDF” dialogue. They’re really handy, but even arbitrary, automated processes constitute intentional fraud. After all, technically altering the visual information.
*relax tongue, withdraw from cheek*
I’ll look at the problem when I get back from vacation. The two issues I see off hand are the one Expelliarmus mentioned, not taking into account the number of comparisons you made of other letters before you found two that matched (an argument that leads to no one winning the lottery) and if I recall right your assuming that each event (whether a pixel was black or white) was an independent. The second is analogous to saying that one in a thousand people are named John and one in two people are male and then computing the odds that a particular person selected at random is a male named John at one in two thousand. The problem with that is that once you have John the conditional probability that John is male given that his name is John is close to one, not one half.
In fact, because the bits in the image are the result of two algorithmic simplifications, and we don’t know what those algorithms do, the only way one could come up with an estimate of the conditional probabilities would be through sampling, and that gets to the whole business of controls.
I don’t think many would fail to find the similarity in letters that you found striking, but sometimes normal things seem odd when one doesn’t know all the details.
By the way, you got a rough reception here. I would explain that by the fact that nearly 100% of people who have said they were not birthers here in the past, but had any objection whatever to any aspect of the official version of Obama’s history or documentation shortly revealed that they were lying about not being birthers. In fact, you may be the only exception in two and a half years. We deal with so many obnoxious cranks that it’s hard not to jump to conclusions. Once you build a reputation, you shouldn’t have any problems.
To fellow posters,
I have read your posts. I have not been able to get verification that the documents submitted by the Hawaiian DoH are identical to those posted by the White House. I notice that quickly people on this blog mentioned that it ‘should be good enough’ that the Gov. of Hawaii posted a link to the white house document, or that it seem intuitive. However, without verification the documents and claims are suspect of any released PDF. Everyone can believe what they want, or even go with the odds, but the facts of proof remain elusive. Sadly, the media, outside of WND didn’t make the effort to got to Hawaii and force them to confirm that the PDF was originally created by them, or confirm the copies that have been provided by the public.
No tongue in cheek here, seriously. The problem encountered is one of the releasing body. For instance, I don’t go and get a transcript from my school and then release a PDF document for proof of my perfect grades, it’s released from the source unaltered. Any disagreements here with this notion can quickly evaporate with individuals attempting to apply for say graduate school with a PDF transcript they bring in. The question really isn’t how the document was scanned, but why it wasn’t released to a group of reporters, including WND, by the Hawaiian DoH representative with Perkins Cole representatives present. Outside of that event, DoH officials should have immediately confirmed to all political bodies and news media the authenticity of the online PDF. I have yet to find a Perkins Cole representative claiming that the document PDF is a digital copy of what they received.
I do not claim, based upon my search for verification, that the PDF is false or forgery. Legally, and I am not a lawyer, it is claimed that Hawaii MUST release documents when they are public released, i.e. the privacy right is gone when someone makes a public release of these documents. They have now refused to verify the White House Documents, as acknowledged by Reality Check, and have refused to release copies (confirmed).
I made an attempt to prove that the online PDF was what it was claimed to be by the White House by verification and chain of evidence. I was unable to do so. Finally, I have witnessed outright fraud on the part of the White House before sometimes with disastrous results. LBJ and the Gulf of Tonkin being one burned into my memory with the subsequent misery that was Vietnam. Only to find later it was a staged event with the leaders of Congress, in both political parties, to afraid to tell the American people the truth about the administrations deceit. Theodore Roosevelt said that the President should be the person we as Americans question the most, not just give a pass.
To those whom make light of this attempt, because they are entrenched in their convictions, so be it. Dr. Conspiracy, thanks for tolerating my attempt.
I said no such thing and please do not misquote me.
See folks. A birther would never say something like this.
Yes, Pete, you’re right. It’s certainly possible for the Hawaii DOH to link to the birth certificate on the White House website and have it not be an image of Obama’s genuine birth certificate. Because the Hawaii DOH did not specifically say the exact words you wish them to, that throws the whole thing into question. I encourage you and your conservative friends to make this an issue during the 2012 campaign, especially with whichever Republican becomes the nominee.
Ah, that is very helpful. When I export the images out of Acrobat Standard, it does not ask me for a resolution, but appears to use the internal resolution of the original. When I exported, I got three images and each at a different resolution. If you follow my link in the main article to my exported image, you can see the actual resolution. A next step for someone with a copy of Illustrator would be to export the object at exactly the same resolution and compare.
I think that we now know why cadwaladr got the results he did, and why his letters match, and mine don’t.
I was about to write that I was just lucky to do it a better way, but I did it in the way that I thought involved the least chance of software intervention, and it looks like that was right.
you don´t do bright, do you?
let´s try again….
“HONOLULU – The Hawai„i State Health Department recently complied with a request by President Barack Obama for certified copies of his original Certificate of Live Birth,
which is sometimes referred to in the media as a “long form” birth certificate.
“We hope that issuing certified copies of the original Certificate of Live Birth to
President Obama will end the numerous inquiries related to his birth in Hawai„i,” Hawai„i Health Director Loretta Fuddy said. “I have seen the original records filed at the Department of Health and attest to the authenticity of the certified copies the
department provided to the President that further prove the fact that he was born in
Hawai„i.”
in other words, they´ve PUBLICLY verified the authenticity of the LFBC.
do you think they have the time or inclination to reply to every birther mongtard that gets in touch with them since saying `i want you to verify to me personally that it´s legit!´?
no, they don´t, and under the circumstances i´d believe they´d be more than justified in transferring calls of this type directly to a psychiatrist
You know, this was a pretty chaotic exchange, and not the most civil, but it looks like we’re reaching a resolution. Perhaps if e. m. cadwaladr can export the object in Illustrator at the proper internal resolution and view them in Photoshop finds that the pixels don’t match, as we see in my article, then perhaps he will write about his experience on his web site. He seems a thoughtful fellow and might offer some insight on how one approaches claims such as the ones in the birther video he saw. Perhaps I will think about this and write something too when I get back home.
Puzzling just a bit over the wording. I’m tempted to think “certified” in the above is a misplaced modifier. Surely the DOH doesn’t mean it has certified the PDF on the White House server?
It’s “An image of the certified copy of his original certificate oflive birth was posted on line”.
Having said that the DoH of Hawai’i has stated that the certified copy was released and explicitly link to the WH posted image so…….. 8–)
Yes, but if you were a recent grad and decided to post your transcript online along with your resume, in hopes that your excellent grades would help you get a job, you would have to post a PDF or some other image. Your school won’t post a PDF online, nor grant access to your transcripts to anyone who asks — so the only way to create an online version that anyone can see is for you to create an image, either PDF or JPEG form. PDF is a little better choice for something in print, where you want good contrast & sharp readable letters.
We know that Hawaii ALSO provided Obama with TWO certified copies of the birth certificate. How do we know? They SAID so — and that’s what you claim you are looking for, their statement:
Source: http://hawaii.gov/health/vital-records/News_Release_Birth_Certificate_042711.pdf
Obama had a press conference and brought at least one of those paper, certified copies to the conference, allowing at least one member of the press to inspect it and handle it. We know that because she said so — see http://www.obamaconspiracy.org/2011/05/be-sealed/
So either the Director of the Hawaii Department of Health and NBC White House correspondent Savannah Guthrie are both liars, or else the paper, certified copy does in fact exist and has been provided. (And that’s what your school would provide as well, a photocopy with some sort of certification on it attesting that it was genuine — in the case of a transcript there is no single original document. Your actual grades are stored in digital format in a database).
So you tell me: how would you solve the problem of posting your transcript online?
In this case, Obama chose to use a PDF. The Hawaii Dept. of Health not only confirmed that they provided the document, they linked to it — and their press release specifically affirmed that the document released by Obama in 2008 was his COLB:
(Same link as quote above)
So now you either have the top officials of the State of Hawaii lying — or you must hypothesize a situation in which Obama receives a genuine document but posts an altered version on the web site — but altered in what respect? The only issue relevant to his place of birth has been specifically confirmed by the state, in writing — that Obama was born in Honolulu, Hawaii, at 7:24pm, August 4, 1961, to mother Stanley Ann Dunham and father Barack Hussein Obama. Other info on the certificate is irrelevant to the birthplace issue, so what would be the point of altering it? And why take the risk of an alteration that could easily be spotted by an official who had seen the original, assuming there was any benefit to making such an alteration?
…so here’s a thumb in Orly’s eye 🙂 I’d love to see Taitz march a copy of the PDF into court, only to be slapped in the face with one of the certified BCs Obama’s lawyer’s been carrying around in his briefcase for the past couple of months.
I posted a direct quote from the Hawaii web site at http://hawaii.gov/health/vital-records/obama.html — I agree that it would be more accurate for them to have said, “On April 27, 2011 President Barack Obama posted a digital copy of a certified copy of his original Certificate of Live Birth” — but that would have been rather clumsy construction.
I interpret their statement together with link to simply mean that the thing posted on the WhiteHouse.gov web site is indeed a digital copy made from a scan or photocopy of the paper certified copy they released.
*sigh* Cripes on toast! Your problem is that you are so obsessively focused on a narrow and irrelevant detail of examining an image copy (yes, irrelevant, because it is the actual document and its origin that matter, not some electronic scan made of it later), that as the whole saying goes, “you can’t see the forest for the trees”. And thus, you end up down some crazy conspiracy rabbit whole of paranoid madness that might sound “plausible” to you at that isolated detail level, but is completely incongruent in the big picture view.
Your whole “Gulf of Tonkin” analogy is making the tired and lame connect-the-dots play between unrelated events that the typical lazy cynic pulls to cast doubt without thinking, giving them something to flap their gums about, without having to actually use their brain and come up with a viable explanation. It’s just the paranoid cynic’s version of “deux ex machina”, except that you substitute X government/person/group that you don’t like for your carelessly wagging finger of doubt.
You have obviously already invested yourself too much into your little conspiracy adventure that you’re trapped under your own microscope and are having difficulty seeing the bigger world anymore. The one in which an official issuance office openly vouching for a document and linking to the White House is confirmation not “conspiracy”. The one in which the LFBC just so happens to corroborate what every other common sense piece of standard information has stated along this whole strange and stupid trip of discovery: COLB, Obama’s books, divorce records, passport records, birth notices in newspapers, etc. etc., etc.
Yet instead of using logic and reason, you choose to ignore Occam’s razor and look for anomalies BECAUSE…
…because…
…because WHAT??? …and WHY???
…AND that’s the REAL dilemma for you.
As the old truism goes, Extraordinary claims require Extraordinary evidence . For your whole little dabble into PDF land and passing vapors of whiffs of conspiracy…then there MUST be some actual conspiracy story, with a meaningful reason and motivation behind it.
…which for your troubles must mean that somehow “everyone’s in on it”…all officials, all elements of government…and some goofy “grand conspiracy plot” going back decades…
So really. Take a deep breath and step back and try to look at the bigger picture clearly for once. Just where does your “analysis” take you in a bigger context and how in any way does it make sense and explain away everything contrary to it? If you are not so biased as to actually be able to take that step back and reflect, you will see that the whole rabbit hole you are trying to pursue goes nowhere and makes no sense. You’d have better luck trying to find convincing evidence that Mole People secretly rule us from the Center of the Earth…
A really entertaining thread. Don’t know about Cadwaladr being a “birther” or not, but I was amused by his analogy vis-a-vis “controls”.
He wrote:
“Certain occurences are straightforwardly improbable and do not require controls. If one finds a body on the sidewalk with a bullet through its head, one does not need to examine a few dozen demographically similar people to determine whether or not bullets just naturally occur inside the cranium.”
In fact, in the situation where one finds “a body on the sidewalk with a bullet through its head” one is, implicitly, using “controls”. If the “body on the sidewalk” were non-anomalous, one wouldn’t notice it. On the other hand, you know that the body on the sidewalk is anomalous because your experience with bodies (your rather large control group) suggests it is.
In the situation involving the PDF of Obama’s LFBC, we need an explicit control document because most of us have no familiarity with what makes for a “non-anomalous” PDF file of a copy of an original document.
The assumption that a “control document” would have to be either identical to the one the WH used or as similar as possible is … not quite right. I mean, I understand that someone could object that every PDF file of an original document is sui generis, and so demand that the test be made on the SAME document. But I’d be happy to know if Cadwaladr has actually copied ANY document, checked the right boxes in Adobe and saved it as a PDF file. And, if he’s done so, if his document (his own birth certificate, say) behaved in a way markedly different from the WH file on offer.
I apologize if someone already suggested all this and I missed it.
As do I, hence my presumption of the misplaced modifier: “certified” was certainly intended to apply to “original Certificate” rather than “copy”.
Or simply, “an image of a certified copy of his original Certificate of Live Birth”. Still a bit awkward, but more accurate.
In any case, it’s a grammatical point only a lawyer or an English teacher would be likely to find interesting, and IANAL. I’ve already given the point more attention than it probably deserves.
Good point, because cadwaladr is assuming that two instances of the same letter would not look the same, which is analogous not to finding a dead body, but to meeting an person who is an identical twin and being surprised that he looks exactly like his brother.
Why would he assume it to be “improbable” that one letter B would look the same as another, when both were presumably created by a single typist working with a single typewriter when entering information into a document? The fact that the other B’s in the document look different would mean nothing – that would be the analog of the twin’s other siblings looking different.
What he really needs to do is obtain a manual typewriter, and type out a page with lines that look something like this:
B B B B B B B
He could start by typing 100 B’s and 100 I’s, as those were the letters that he found identical in the birth certificate.
He could then scan the documents, convert to PDF, optimize for web, and then export the letters to bitmap form and conduct a comparative analysis. That would give him a single sample on which he could survey to determine the incidence of identical letters. If he found none at all, that would indicate a likely low incidence, and he would have to repeat the experiment with more letters.. If, on the other hand, he found more instances of duplicates (let’s say, hypothetically, that out of 100 typed B’s, he found 3 pairs that were identical) — then he would see that his hypothesis about probabilities had been disproven, and that the likelihood of a duplicate was probably higher than he had thought. That wouldn’t give him an accurate way to predict the numbers — he’d be better off if he had 1,000 B’s to look at rather than 100, and if the trial were repeated with different typewriters — but at least he would be making progress.
I have yet to find a Perkins Cole representative claiming that the document PDF is a digital copy of what they received.
That is one of the silliest comments I have seen from a birther in some time.
Legally, and I am not a lawyer, it is claimed that Hawaii MUST release documents when they are public released, i.e. the privacy right is gone when someone makes a public release of these documents.
Claiming it doesn’t make it a fact. Per the State of Hawaii: “State law prohibits the DOH for disclosing any information about a Hawaii vital record unless the requestor has a direct and tangible interest in the record. This includes verification of vital records and all the information contained in a record.” Nothing that Obama has done changes Hawaii law. He could set up a cable channel which displays his birth certificate 24 hours a day and Hawaii DOH still would be prohibited by law from commenting on it.
Legally, and I AM a lawyer, that claim would be wrong.
* I love the use of the passive, “it is claimed” language. Who makes such a claim? “It”. Is “it” a lawyer? Probably not. Where was such “claim” made? When?
I find the use of such weasel language by someone who takes issue with the wording used by Hawaii officials to be quite ironic.
Doesn’t the fact that the jpeg has in your own words, “nothing suspicious” destroy your argument? Both the pdf and the jpeg are designed to represent in electronic form the paper document. Anything that appears in one but not the other would most likely be an artifact of the digitization.
Furthermore, no one is saying that forgery of electronic documents is impossible. But here you have a paper document produced in front of several state officials (who have not been previously accused of being conspirators in forgery). The paper documents have been shown to reporters and photographed by them.
The “anomalies” you say you see don’t conviince me. They appear and disappear depending on the type of file and the magnification. That strongly suggests they are electronic artifacts.
Not at all. i never said the controls had to be Hawaiian birth certificates. Your contention is that the odds of 2 instances of a given typed letter will map identically in a pdf are some very large number. But that is simply an assertiion, not backed by any empirical data. To validate this you should at least look at a sample of typed documents from the 1960s-they don’t have to be Hawaiian birth certificates. You could find old letters, original birth, death, marriage, etc. certificates from family members, etc. I have several such documents in my possession and I bet you could find a few. Then you digiitze them and see whether any letters look the same as you claim the B’s in Obama’s b.c. do. A few documents should easily give you several hundred letter pairs to analyze, Suppose you found none that matched- that still wouldn’t prove the odds were 1 in 1 billion as you claim, but it would at least assure you that the odds are < 1 in several hundred, which would give some support to your position. But suppose you found a few identical letter pairs in the course of your analysis. That would say that the phenomenon is far from rare and is in fact quite expected.
The fact that you steadfastly refuse to do such simple controls leads me to believe that you are afraid the answer will be that such "anomalies" are quite normal and expected.
I’m neither a mathematician nor a scientist, so I can’t analyze those aspects of EMC’s argument. On an intuitive level, however, as I read the story, one of my first thoughts was, assuming all else being equal — same typewriter, same typist, same scanner, same software — the more interesting question is “Why aren’t all the Bs the same?”, and then work toward explaining the divergences.
There is no need to assume either way. The question is easily tested. Many things were assumed by somebody or other -the Earth is flat, the Sun orbits the Earth, thoughts occur in the heart rather than the brain. Those who assumed those things constructed quite elaborate world views based on those assumptions. All those world views collapsed in the face of empirical evidence.
Such use of weasel words is often a sign of a Concern Troll. I strongly suspect that is what we’re dealing with here.
By this point in time, the odds of this just being a highly myopic and misguided rational person as opposed to a concern-trolling, die-hard birther, hopelessly brainwashed by their own ideologically driven biases, are slim.
If it were a word processed document of recent vintage, then I think it would be reasonable to expect that the letters would be the same. But as it is a typewritten document, the quality of the impression of each letter varies with the typist’s speed and the degree of wear on various spots of the ribbon. Also, with an older document, you would expect to see some fading of ink with age.
Anyway, my guess is that the incidence of similarities in typed characters at the resolution that cadwaladr chose to probably is more like 1:20 or 1:50 than the one in a billion he asserts. I don’t even know how he would get to that number because there can’t be more than a few hundred pixels where there could be variance and still have it come out looking like the same letter. I mean, look what happened to the H in the the word THE in the certification stamp with only a handful of pixels missing –it somehow was transformed into an X. So too many points of variance, and the B would turn into an R or an H or an E.
Since it is clear that cadwaladr has NOT looked at any other documents, the only “evidence” he has is the fact that he found similarities within the Obama birth certificate. So his sample of 1 document disproves his hypothesis that it is a 1 in a billion occurrence. Rather than recognizing that it his unsupported assumption that is faulty he claims fraud — but that’s backwards analysis.
i would probably guess close to what you guessed, but there is no need to guess, since the actual numbers can be determined by empirical observation. All it requires is a scanner and a few 1960s era documents, something which most people 45 or older should have and which younger people should be able to get from their parents or other older relatives.
As far as I can tell, not a single one of the so-called “experts” screaming fraud has subjected any other 1960s document to the “analyses’ he has done on the Obama b.c. to demonstrate that any of the claimed “anomalies” are unique to the Obama document. To call this quackery would be an insult to quacks.
Why do you keep rambling on about a PDF? the irrelevant PDF? The utterly, completely, stupendously irrelevant PDF? Shouldn’t you be focusing on the actual BC? The one with the HDOH raised seal, and Onaka’s signature, and mutliple official corroborating statements from the HDOH?
Just for the sake of argument, let’s stipulate that everything birthers says about the PDF is true — it’s a fake, a fraud, a forgery, a lie, a deliberate falsification. You have now proved exactly nothing. The legal proof of Obama’s Hawaiian birth has nothing AT ALL to do with the PDF.
Have you found any that deny it? Then what’s your point?
Do tell: precisely what steps did you take in making this “attempt”? What, exactly, do you think the White House claims the PDF is? And tells us A) exactly what level of proof you require in order to be convinced, and B) who you are that anyone should go to that much trouble on your behalf?
Arguing with birthers is like a giant game of dodgeball. No matter how many facts you lob at them, they simply jump out of the way, yell, “Nyah – you missed me!” and continue to pretend you haven’t answered their questions. Any elusivity of facts you imagine, Pete, is due entirely to your nimbleness in dodging them.
The HDOH has made multiple official statements corroborating that Obama’s original BC is in their vaults. They have verified that the short form on Obama’s website is an accurate facsimile of the facts it represents. They issued certified copies — complete with raised seal and Onaka’s signature — of Obama’s original COLB, which the White House subsequently made available at a press conference for reporters to examine. Then, for the convenience of the 300-odd million Americans who weren’t able to squeeze into that conference room back in April, they made digital facsimiles available on the Internet. Nobody, anywhere, has claimed either the PDF or the JPG are legal evidence of anything; they’re simply there for convenience.
Here’s more on the flaws of cadwaladr’s “math”:
He wrote:
Fair enough — I assume that there are probably 40 possible points of variation.
He is making the mistake of assuming a random variation, but the variation is tied to real-world factors, like the quality of the impression made by a typewriter. So if there is pixel that will usually render as black whenever there is a firm, clear impression, the odds of that particular pixel being black are not 1:2 — it may be more like 9:10.
This is exactly my point. I don’t mean to suggest that one in reality should reasonably expect them to be the same, but simply that it would be more reasonable to begin by positing similarities and then seeking to explain the differences, rather than vice versa.
I was struck by the thought that EMC’s analysis seems to be taking a backward approach. In other words, the proper resonse to EMC’s “Why are they the same?” is “Why shouldn’t they be?”
As to the mathematics in EMC’s analysis, I’ll leave that to someone competent in math.
Thanks. What I meant was assumed as an initial hypothesis only, not as a truth, as one begins with a presumption of g=9.81m/s^2, then seeks to explain terminal velocity, rather than vice versa.
Apologies. I’m tripping all over my layman’s tongue trying to discuss scientific principles ðŸ™
It sounds to me suspiciously like the subpoena Taitz issued to the Hawaii DOH as part of the discovery process that the judge in Taitz v. Astrue never ordered.
Seems I wasn’t quite right. It’s in the letter Taitz wrote to the DOH when they originally told her what to go do with herself:
“Your assertion, that you are refusing to allow access to the original long form birth certificate for Mr. Obama due to consideration of privacy is of no merit. If the document produced by Mr. Obama, is a true and correct copy of the original type written birth certificate, than there is no consideration of privacy, as the information in the document on file will be identical to what Mr. Obama released.”
http://www.liveleak.com/view?i=395_1307643602
By backward, I mean that he interpreted evidence that refutes his hypotheses as evidence of fraud. As I’ve also pointed out, his hypothesis is based on a faulty mathematical assumption that pixel variation is random and occurs at a 50/50 rate, as opposed to the more likely assumption that there would be a high degree of concordance among pixels in an image made from the same typewritten letter, produced by the same typewriter at or about the same time.
You don’t really need to know any math at all understand this. All you need to know is “garbage in, garbage out”.
All that EMC has learned from looking at the Obama certificate is that both his assumptions and his math is wrong. This is clear because (a) we have all been informed of the provenance of the document, and (b) if there was a forgery, the forger would not be using a letter-by-letter cut & paste approach. If his hypothesis is that two matching letters are indicia of fraud, he has failed to explain why none of the other letters match.
Hi friends! Check out my new blog:
http://www.wasobamaborninkenya.com/blog
I don’t find the answer interesting at all, actually. No two letters are the same because of differences in the paper fiber from location to location, the imperfect spread of ink in the ribbon, the differences in the weave of the ribbon, the different wear of the ribbon that has gone through several rewinds, the difference in pressure of the typists fingers on the keys, differences in the ‘give’ of the platen surface due to manufacturing imperfections, and do on and so on. So many variables that it would be very surprising if any of them were the same.
I meant question: I don’t find the question interesting…
Keith, Nathanael: Your discussion points out once again how empiricism is more powerful than deductive reasoning. In the case of terminal velocity, if one considers only the effects of gravity, terminal velocity should not exist, since the accelerating force of gravity continues to act throughout the fall. Of course, empirical observation shows that objects falling towards the Earth do indeed stop accelerating once a certain speed is reached. The answer, as we all know, lies in the fact that consideration of the single variable of gravity ignores the force of air resistance which also acts on the falling body. No restatement of Newton’s laws are needed; merely a complete accounting of all forces that act on the falling body. In a vacuum, terminal velociity does not exist and objects will continue to accelerate indefinitely-at least until they encounter the relativistic effects of increased mass as they approach the speed of light, as Einstein noted.
As applied to a pdf made from a typed document, different people appear to have different expectations as to whether all examples of a given letter should be identical or different. Keith and cadwaladr seem to “expect” that no 2 should be identical, while Nathanael seems to expect they should. Who is right? Only an empirical study of a set of pdfs made from 1960s typed documents can answer the question. In fact, such studies may have already been done. cadwaladr should have at least made an attempt to find such studies before issuing his pronouncements as though he had descended from Mount Sinai.
In the end, failure to meet someone untested expectations no more indicates a fraudulent document than the existence of terminal velocity in air means Newton was wrong.
I see your point, and maybe even Nathanael’s point.
If the PDF conversion sees the document as text and does a competent OCR conversion, then the digitally stored characters will then be as identical that the applicable variables (e.g. when printed: paper weave and diffenet electrostatic properties from place to place on the drum and the paper; when displayed, ability of screen to match the resolution of the scanned document, dead pixels, etc).
But if it processes the image to conserve the original look of the original document, then it will duplicate the imperfections in the original. Except that even then the process can introduce imperfections of its own, from imperfections in the glass platen of the scanner to dead pixels in the photo-receptors to memory errors inserting random bits.
So, if I understand Nathanael’s point properly, I don’t actually have a different expectations depending on context.
By the same token, your call for good old fashioned experimentation, while admirable, is probably not required in this case. If we start from the assumption that any given typed letter is subject to the variables I listed, then chaos theory informs us that it is virtually impossible for any two examples of a typewritten letter to be identical.
On the other hand, if we look at the OCR text scenario, then we have a new variable added to the litany of variables described above: the ability of the OCR software to recognize a glyph that has been printed imperfectly, so again the imperfections combine to put us into the same argument in the above paragraph. Of course if we start from the software treating it as an image, then we carry over all the same imperfections as in the original.
Experiment and observation don’t discourage flat-earthers or young earth creationists. Why should experiments make any difference to a birther?
The Bs are the same because the optimization severely reduces the resolution, removing the differences, and the viewer interpolated the display version when blowing it back up, making it appear that there is more resolution than is really there.
I think you are right about that, though not because we should stand on untested assumptions, but because it is virtually certain that the experiments have already been done. Obama’s b.c. is hardly the first typed document to be digitized. I suspect that historians have digitized many old documents in the course of their`research and such files are probably easily found by anyone who would care to look. The job of doing so rests with those like cadwaladr and others who wish to propound all kinds of outlandish theories. If they refuse to validate their assumptions then they should consider what the first 3 letters in “assumption” spells.
As far as who believes and doesn’t believe experiments, i don’t really care. It’s evident that birtherism is dead and its impact is nil. Nevertheless i expect people who wish to throw around science and probablility theories to be familiar with and follow the scientific method.
But doesn’t the software used to create the pdf (Mac v. Windows, Adobe v. other software makers) make a difference? Will they all optimize in the same way?
Thanks for the reply, Keith. I guess I’m not doing a very good job of explaining my reasoning. Perhaps I shouldn’t have used the word “expect”. Of course I wouldn’t expect in a real-world sense the letters to be identical for the reasons you list (and others). But the explanation for the variations lies in factors outisde the initial hypothesis.
Exactly. But when doing one’s experiment, would one not begin with the hypothesis of the gravitational acceleration constant, note that empirically observed data deviate from the initial hypothesis, then seek for the explanation in external forces — in this case air resistance?
That’s what I’m trying to get at. Our actual empirical data (in this case, the real-world shape of the letters) deviate from (what, to my mind, should be) the initial hypothesis. But the deviations should not be a part of an initial hypothesis, in the same way that one doesn’t begin with an expectation of terminal velocity then seek to explain why objects in a vacuum deviate from it. The acceleration constant is the “norm”, terminal velocity is the “deviation”.
My initial reaction was that cadwaladr was beginning with an expectation of terminal velocity. But as I’m the only one who finds my reasoning helpful, I’ll stop now 🙂
Of course, as you previously suggested, the jpg clearly demonstrates cadwaladr’s identical pairs aren’t so identical, so he might have saved himself a lot of trouble.
It’s a valid question and i don’t know the answer. I look at it in this way: an honest investigator tries to falsify his own work. A reasonable first step would be to look at a sample of the documents that are available. If they show examples of letter pairs that look the same, like cadwaladr claims for the “B”s in Obama’s b.c., then the hypothesis (that two identical or simillar letters suggests forgery) is falsified (since it would show that image processing can produce identical letter pairs) and one need go no further. If such examples are not found, that certainly doesn’t prove that the b.c. is a forgery, but it at least says one needs to try to replicate the conditions used more closely.
What is not acceptable is to look only at the b.c. and make unsupported assertions that something is anomolous (when it may be quite normal) and conclude forgery. An honest person has to first show that the feature is unusual (by actual data, not supposition) and then rule out innocent explanations.
i think cadwaladr’s goal is to make trouble, not save it.
On another topic, here is a great examination of the curvature in the letters on the LFBC:
http://rcradioshow.blogspot.com/p/paul-irey-and-doug-vogts-incredible.html
Doc,
If still interested … the resolution of the image you linked to is 240dpi. Extracting an uncompressed TIFF from Illustrator, as EMC did, at this resolution yields no match. It’s a train wreck at 150dpi, close at 240, a match at 300, and any multiple of 300, so long as anti-aliasing is disabled.
Looking further, the markup in the PDF says all the objects encoded at 1-bit (the bitmaps) were filtered with Flate. Flate is a complex “algorithm of algorithms”, that breaks an image into zones and selects an approach (one of several compression algorithms) to each based on Flate’s own criteria. Flate may also decide to leave a zone, or by extension an entire object, uncompressed.
Soooo .. it may be possible that it’s uncompressed, and the letters do indeed match, if the original object was 300dpi.
Once again, in light of other images with matching word/lettering, that weren’t created from the PDF, this remains a curiosity and nothing more.
As I work with the conversion of vector art to raster images everyday, the idea of a formula to determine the likelihood of identical renderings at a given resolution and color depth is interesting, but the practical value? The list of variables involved would be very lengthy. However, at a color depth of 1-bit, and a relatively low resolution, it’s going to happen, even on typewritten copy, with frequency. It’s interesting, but not significant.
So we may still have some loose ends. I don’t understand, though, why the letters could be the same, but my Adobe Acrobat export show them as different.
In any case, whatever letters that appear the “same” where not the same originally as evidenced by the AP scans. The simple reduction of gray scale to 1-bit loses a lot.
What steps did you follow to export from Acrobat?
Yep, and vastly increases the likelihood of identical renderings. Anyone started on that formula, yet? LOL
I think, from the get-go, it was at least intuitively clear this had no significance beyond curiosity value. At least, for once, it could be demonstrated the point wasn’t an outright lie, just a misinterpretation. I don’t think EMC was intentionally deceptive, he thought he was on to something. I know it’s frustrating to think you have something, and be rejected, and even harder to admit it was nothing. This thread has been like the OCT version of Letterman’s lame skit, “Is This Anything?”
Since PDF mania has been such a massive issue the past two months, and wasted much bandwidth, are there any online clearinghouses / catalogs of the various object fixations regarding the PDF? The charlatans have generated quite a list.
I personally don’t think emc was honest about his methods — I think he did the Adobe reader onscreen magnification thing that others used to replicate his results, and then just claimed to have done an illustrator export later on when challenged.
However, if I understand Potter correctly, he’s saying that before you do the export, you can set the DPI, and at 300 or any multiple thereof (600, 900, etc.) with the 1 bit image, he’s seeing a match. That makes sense for the onscreen stuff as well, as we start seeing the match at high end multiples. The common link is that any time the resolution is increased beyond native format, the program has to add in data, so you are looking at Adobe’s best guess of filling in new spots between pixels. So the bottom line is you are looking at the way that a particular piece of software makes changes to an image, not the image itself.
Obviously the much better point of comparison is the high resolution jpg. It would be nice to have a photographic image in a non lossy format, such as a TIFF — but we do not have that. Exporting a PDF to TIFF is not the same thing, because of the conversion process that took place with the PDF & PDF compression algorithms.
type 1000 Bs on a document and scan it by a similar process
how many of them are identical
if someone does it and produces a .tif or .bmp (rectangular)
I’ll do the analysis