Just to check if I'm getting this correctly: There's an immense amount of sigint data that's (being) leaked into public infrastructure - and wilfred@vt.edu is telling us about it?

Can we access this data **? How is Wilfred knowing of this, and allowed to speak of it? Why not speak of the onloader's identity? 

Tracking cash currency is certainly interesting from many standpoints... actually doing it seems outrageous. Leaking this data intentionally is extremely outrageous - no matter the target's value the laundering can not warrant the backlash. As a politically minded shake-up-leak, this one is the most daring so far, and would most likely be the most effective at dismantling the espionage engine. It almost seems too good (and in a way terrifying*) to be true...

Is dear Wilfred pulling our legs? How would we know at this point?

Assuming truth.. please validate known NSA locations and other US-secret areas. If the dataset is manipulated at all - and one would assume that it is - it should exclude sensitive persons first. Sensitive persons should hang out near sensitive person areas - if the sensitive person area's are less full than they should be.... Think Obama sitting in the white house and going home, think the first family, think area 51, think coins not being attached to people properly. Note that notable persons may be falsified, so ideally one would find an atypically understaffed military base or something of the like. Perhaps an agent/secret-base that was exposed? All those that entered an Internet-tap-room in a datacenter? Military ships' crewmen? If such data-gaps are found/indicated, compare it to other nations and you'll know which who's data you're receiving (although everyone understands it'd probably be the USGOV/NSA).

Wilfred, are you publishing this to prevent the data just disappearing?

* I'd actually really like to know where I've been in the past, and I know they know but won't tell me. And the amount of exceedingly valuable scientific (census) data one could parse from such a database.... Still, we'd move rather suddently from panopticon to omniopticon (a term I thought of to describe "everyone watches everything" instead of "they watch everything". I know it's not a flawless name but it works).

** I realize there's no way we're going to store or transfer this much data - but there should be something that can be done to preserve this dataset!