As the previous discussion of the actual NSA document (kindly submitted in the HN thread linked here by brown9-2, and also linked in the article submitted here)

As the previous discussion of the actual NSA document (kindly submitted in the HN thread linked here by brown9-2, and also linked in the article submitted here) http://www.nsa.gov/public_info/_files/Untangling_the_Web.pdf http://www.nsa.gov/public_info/_files/Untangling_the_Web.pdf notes, the NSA document is old, and thus dates from an earlier era of Web research. For all that, applying the Google tips mentioned in the document, specifically domain-restricted searches, would improve the quality of links submitted to Hacker News and thus improve the quality of discussion here. The previous discussion of the actual document link seems to have missed the process by which the document was released. The document was an internal document of the National Security Agency, requested by a Freedom of Information Act request. Some parts of the document had to be redacted, and of course "official use only" had to be struck out in the copy released to the public. The editing process apparently involved mark-up of actual paper printouts of the original internal document, which were then scanned again to form the released document. This makes for ugly formatting, but the content is still interesting.