|
Jam Echelon Day
9/13/99
Mike Hudack
Editor-in-Chief
Jam Echelon Day was announced just about a week ago. The announcement garnered some attention everywhere from ZDNet to Hacker New Network. Hell, even AntiOnline covered it, albeit negatively.
The idea behind Jam Echelon Day is relatively simple... Apparently NSA has a planet-wide eavesdropping network that records every electronic (or other?) message that contains certain keywords. The evidence to support this global network is generally in the form of anecdotal evidence and a patent filed by NSA.
There are several hurdles, both technical and political, to the believability of such a system.
Political Hurdles
First and foremost, NSA isnīt allowed to spy on US Citizens, locally or abroad. Although it may be against the law itīs generally agreed by historians that both CIA and NSA have spied on US citizens -- so there goes that argument. Secondly, Congress (and certainly the public) would never stand for something such as Echelon -- but would they ever know?
Technical Hurdles
The technical hurdles of a system such as Echelon are somewhat more complicated and numerous than the political hurdles.
Itīs generally acknowledged that NSA monitors most foreign traffic, at least that of governments and major corporations. NSAīs function in the intelligence "circle" (as ODCI [Office Director Central Intelligence] likes to refer to it) is communications -- NSA is supposed to monitor communications, break codes and make codes. NSA has listening stations around the world, in locations such as Saudi Arabia, Washington DC and Southern France. And if theyīre listening -- theyīre going to record.
The real question in regards to Echelon is wether it records electronic communication and whether it is a searchable database. At this late date itīs inevitable that it records electronic communication -- the Internet is just too important a medium to be ignored. The more important question is whether the recordings are searchable.
Voice
Voice communication is much harder to place in a searchable database than electronic communication. The spoken word is difficult to recognize and catalog -- and it takes up massive amounts of storage. Given that NSA is widely assumed to have one of the largest repositories of super computers in the world, this may not be too much of a problem for the agency.
Three things are necessary for an effective (and searchable) voice database: excellant voice recognition technology, massive amounts of storage, and huge amounts of raw processing power.
IBM and Dragon voice recognition systems are good enough if you train them. You can hook them up to your PC and talk instead of typing (a definite plus for a certain OSAll editor <g>). The problem with these products is that you have to train them. You canīt just bring in Joe Schmoe off the street, record his voice, and have it recognized.
Given that some NSA technology is up to twenty-five years in front of general public technology (and probably, in some cases, twenty-five years behind), the ability to recognize voices without training may not be too much of a stretch.
With that given, raw processing power and storage isnīt too difficult for NSA to attain.
Data
The ability to search, catalog and do everything else with data simply isnīt a problem for large corporations -- so why should it be for NSA? The only real problem here is the undeniable volume of recordings.
Can It Be Jammed?
The basic premise behind Jam Echelon Day is that the system records every message with certain keywords. But why be limited by that? A subscriber to the Hope 2000 Mailing List (for the H2k conference in New York City next year) pointed out that it would be a trivial matter to add an operation that excludes messages with too many code words.
There seem to be two main goals of Jam Echelon Day -- to raise public awareness of a possible invasion of privacy and to defeat that invasion of privacy. With the public awareness coming into place, now we have to wonder whether the day will actually work. |
