Skip to content

Comment on Using Metadata to Find Paul Revereparent

Comments

I just read his recap, which seems to mirror season 1 of "The Wire" fairly closely. What I remember distinctly from that season (which I've seen in total at least three times) is the great degree of difficulty they had in obtaining the warrants, in meeting the burden of proof, and in convincing a judge that the burden was met to a high enough level to allow the kind of carte blanche access to such a broad tap.

Further, they were constantly in fear of running into the expiration date, and the battle was waged on whether they could get data of enough substance before the tap's expiry that they would be able to extend it.

The problem in reconciling the two situations is that in the Baltimore case, assuming the show faithfully retold that portion, it was hard, but in the case of the NSA, their approvals come from the FISC, which numerically is reduced down to a rubber stamp agency. Since their inception in 1979, up to 2012 or so, they have approved over 30,000 requests, and rejected only 11. Further, it seems that they're somewhat unique in a warrant-issuing agency in that they've provided assistance on those 11 such that requests were amended, reissued, and in many cases, granted. So of those 11 denials, I am led to believe that the majority of those proved to be slightly delayed approvals in reality.

Beyond that, I think that Mr. Simon overlooks the capacity of metadata at that scale, and how effective it could be in painting a picture in a way that smaller scale, Baltimore-wide data could never do. If, for example, the British had been able to capture 'only the metadata' of our founding father's meetings in Philadelphia, it wouldn't have taken much analysis at all to have proven what was happening (famous statesmen gathering in Philadelphia court house for overnight meetings, etc.) Just knowing the attendee list alone would have likely been enough to condemn them all for treason or assassination, and the United States might never have existed.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.