The death of Aaron Swartz earlier this month marks an important moment in the history of open data activism.
In the last 15 years, in response to theorizing and activism by people like Lawrence Lessig and Richard Stallman in the late 20th century, a number of movements with varying degrees of legitimacy have sprung up – from Creative Commons and the Open Source Initiative to WikiLeaks and The Pirate Bay. Now that these movements have begun to gain traction, some of them have started to see strong opposition from governments and corporations. Aaron Swartz was part of this broad range of movements and, according to his family, he was hit with “an exceptionally harsh array of charges [ . . . ] to punish an alleged crime that had no victims.” They believe his prosecution drove him to suicide.
In light of all this, it seems important to take a look at what exactly Aaron Swartz (allegedly) did, why he was prosecuted so ardently, and why his allies believe his actions were right or at least excusable.
What did Swartz do?
According to the indictment, he entered a restricted networking closet at MIT, connected a laptop to the network, and hid it under a cardboard box for several months while it ran a program to automatically download articles from JSTOR, a subscription service that offers access to academic journals. The laptop was discovered and, shortly after Swartz returned to retrieve the data, police apprehended him.
U.S. Attorney Carmen Ortiz, whose office was in charge of prosecuting the case, said in 2011, “stealing is stealing whether you use a computer command or a crowbar, and whether you take documents, data or dollars.” However, even if Swartz did break the law, it’s not quite that simple – it appears from the information that’s been made public that Swartz’s access to JSTOR may have been legitimate.
Visitors to MIT can register a guest account on their network. Guest accounts can access JSTOR, and one of the things you can do with JSTOR access is download PDFs of academic articles.
However, Swartz did register his guest account under a false name, which might be a violation of MIT’s terms of service. Also, JSTOR’s rules specifically disallow the use of automated programs to download from their servers. Swartz was being prosecuted under the U.S. Computer Fraud and Abuse Act of 1984, which was designed to allow for the government to prosecute people who steal information from computers or disrupt computer functionality. Their argument is that violating a site’s terms of service runs afoul of the act, which is a stretch but not as much as you might think.
When technicians at both ends tried to block Swartz’s computer, he evaded security by spoofing his MAC address—a number that is supposed to uniquely identify a computer’s network card—and reassigning his IP address. The constant downloading strained JSTOR’s servers, and users at MIT were unable to access the service for days at a time.
In other words, Swartz may have had the right to download the articles from JSTOR – and if he didn’t, it hardly matters because academics are usually happy to liberate JSTOR articles for people without access. What’s more problematic is the way in which he downloaded them, which was underhanded and disruptive, to say nothing of his alleged intention to upload them to file sharing services. Philosophically, the problem isn’t that Swartz was “checking too many books out of the library,” as activist David Segal put it, but that he drove a dump truck through the library’s wall to carry them.
Many of Swartz’s allies, such as law professor Lawrence Lessig and open data activist Carl Malamud, have acknowledged that Swartz’s methods were problematic and possibly illegal, but have questioned the way the case was prosecuted. It does seem strange that for something that is so minor in the grand scheme of things, Ortiz’s office hit Swartz with charges that carried up to 35 years in prison and US $1 million in fines. This is an especially salient point of consideration given that JSTOR was not interested in pursuing charges.
In the end, the main takeaway from this situation is that issues of copyright, property, and theft become much more slippery when we introduce modern technology. We’re dealing not with solid objects, but abstract patterns of data that can be endlessly reproduced, and we have to think in ways that are still, after all this time, new to us. It is imperative that we approach cases like this in a spirit of philosophical calm, not polarized hysteria.