I think the trend with technology is that it will always reveal information and not hide it. DeCSS is an example, a few small lines of c code which DVD Jon stored on his mobile got around the stupid protection on dvds. Not one DRM system has been found to be uncrackable. So it shouldn’t be too surprising that a PDF file containing a detailed U.S. military report on the killing of an Italian agent by U.S. troops was cracked.
Many lines in the report were blacked out to keep sensitive information censored. The guy who made the PDF was obviously clueless about making PDFs and an Italian student pretty much clicked a few buttons and the black was removed. This reminds me of a script a guy wrote recently which goes over a document with words blackened out. Using some clever algorithm he can still see partial letters and figure out what the words are and so he removes the censorship.
In my own experience with FOI documents in Ireland they are a little better at taking out data. They cover swathes of text with paper and then photocopy them. So instead of black lines you have giant white spaces. Much harder to decode. However they don’t do that with all documents and I have a significant amount with blacklines running through them. I should pass them through that declassifier script one of these days …