If you’re a cutting-edge geek with an interest in investigative journalism, there’s a great job opening at the badly named Reporter’s Lab, a project supported by Duke University’s DeWitt Wallace Center for Media and Democracy. Headed up by former Washington Post editor and reporter Sarah Cohen, the Reporter’s Lab is Duke’s effort to extend what is known as “computational journalism” into the realm of investigative reporting and thereby make investigative reporters more efficient and effective. (I wrote my take on this effort, “Deep Throat Meets Data Mining,” back in 2009; you can find it behind the “columns” tab on the home page of this blog.) The lab, which has an advisory committee that includes many of the top names in American investigative reporting, is looking for a developer, and the description makes it sound like a dream job to me. But then again, I don’t do much in the way of coding (yet). If you do and want to help journalism and advance the public interest in a significant way, you really ought to take a look.
Category Archives: computer algorithms
Developing the future of investigative journalism
Filed under computer algorithms, media, program developers
The rhythm of the algorithm
If you ever wonder where the balance of human and machine intelligence is headed in general, you’ll want to read this piece, “How to Make Money in Microseconds,” in the London Review of Books.” I think most people know, by now, that a lot of market trading is now controlled by computers that place buy and sell orders on the basis of algorithms. This essay goes well beyond that generality to describe exactly what the algorithms do, and in so doing takes a reader down an engaging if increasingly unreal rabbit hole full of statistical arbitrage, volume participation, volume-weighted average price (or VWAP, pronounced vee-wap), and even more arcane and predatory types of algorithms. Such as these:
Far more controversial are algorithms that effectively prey on other algorithms. Some algorithms, for example, can detect the electronic signature of a big VWAP, a process called ‘algo-sniffing’. This can earn its owner substantial sums: if the VWAP is programmed to buy a particular corporation’s shares, the algo-sniffing program will buy those shares faster than the VWAP, then sell them to it at a profit. Algo-sniffing often makes users of VWAPs and other execution algorithms furious: they condemn it as unfair, and there is a growing business in adding ‘anti-gaming’ features to execution algorithms to make it harder to detect and exploit them.
This is an extraordinarily detailed and well-balanced piece that includes no fear-of-computer troglodytism but still argues that computerized trading has become a ” tightly coupled and highly complex” system that probably is, therefore, “inherently dangerous.” Call me fascinated, Hal.
John Mecklin, writer/editor 