Storm Clouds Over the Badlands
When I first began watching the TV series “Person of Interest”, I blogged about it here. I can now proudly proclaim that I have successfully binged all 103 of its episodes. This may not seem like all that much of an accomplishment, but for me completing a TV series is rare. I don’t think that I’ll ever do the same with “Game of Thrones”. Kudos to creator Jonathan Nolan for holding my attention.
Over the show’s five seasons it morphed from a buddy act to a battle for the future of humanity. In the beginning, two guys with the help of an all-seeing artificial intelligence try to do good and save people whose number has come up. Over time an ensemble coalesces into a resistance to a rival AI that is taking over the world. One of the series’ high points was its prediction of Edward Snowden and his data breach that outed the NSA’s spying on America. Homage was paid to Snowden in the show’s final episode when the wi-fi modem that he purportedly used to first breach the NSA network is filched from an evidence locker and is again used to breach the agency’s firewall.
“Person of Interest” is fiction, but in this week’s New Yorker is an article that goes down many of the same rabbit holes that it had. Author Dexter Filkins’ “Enigma Machines” as the article (Paywall) is entitled in the magazine’s print edition, dissects a particularly arcane aspect of the Russian investigation. It involves the 2016 computer communications between the Trump organization and the Russian Alfa bank that could have been the mechanism for collusion.
The Domain Name System (DNS), a worldwide network that acts as the Internet’s phone book, is at the heart of this investigative piece. The DNS is ubiquitous on the Internet. You used it to find this post. The gist of the article is that much like the NSA use of phone metadata, who called who, when and where, a similar hack of the DNS existed in 2016. With this hack, as the article lays out, a meticulously detailed communications chronology is described.
Filkins has written an interesting article, but as the print edition’s title alludes to, it is ultimately unsatisfying and the reader is left with an enigma. This is the fundamental problem with metadata. It can tell you who and when, but never what. You know when two parties communicated, but you don’t know what they were saying. In the case of the Trump-Alfa logs, it could be collusion or it could just as well be marketing spam.
For the NSA, just knowing who a person of interest is communicating with is relevant. Piecing together such leads is how they eventually track and takedown terrorist networks. Filkins’ article does offer some tantalizing clues using the timing and frequency of the Trump-Alfa communications, but there is no smoking gun here and in the end it is all circumstantial. The NSA uses metadata as a filter to whittle down their leads to a manageable number that can then be prosecuted using more traditional means. Filkins concludes that any resolution to the enigma of the Trump-Alfa logs will require an analogous approach.
In The Atlantic, Franklin Foer, who first broke the Alfa Bank story in Slate, a week before the 2016 election, has revisited his story in light of Filkins’ New Yorker article. It provides some journalistic back story to this investigation.