[Kleon]

I've noticed that these cloned pixels seem to show up in scans that went through some OCR (optical character recognition) process.

For example, these come from a blog that provides OCR-processed documents for searching. I opened the first link and found numerous examples of cloned letters

This is proof that these cloned letters are not evidence of tampering, but in fact a product of the OCR scan. It makes sense that you would want letters as uniform as possible.

[conservativegramma]

Your OCR theory has been debunked.

Here

and Here

[moehoward]

Your Google link is indeed the same pixel pattern. We don't know the history of that file though. I tried to find another pdf of the same book for comparison, but no go. I did find this which is a much better/cleaner scan. No dups on the title page though. I don't see any obvious dups in the few Korean files I perused either. I would be surprised if there weren't some though. More characters=higher chance of dups, relatively speaking.

"This is proof that these cloned letters are not evidence of tampering, but in fact a product of the OCR scan".

Depends on what you consider OCR. There's been a pile of confusion on this subject. And there would be no reason to use such a program on this COLB

Bottom line is this doc has so many things wrong with it, it defies logic.

[Mr. K]

You dumbass- I just looked at your post

You are trying to pass off COMPUTER GENERATED ASCII TEXT as ‘pixel by pixel’ duplicates

OF COURSE THEY WOULD BE YOU MORON- If you were to look at each leter here: tttt <~~~ those would all be bizel by pixel duplicates

Shut the hell up with this moronic “OCR” argument- there is NO OCR in this document and there is not reason for there to be.

You have no clue what you are talking about so shut up, stupid