Page:Code Swaraj - Carl Malamud - Sam Pitroda.djvu/153

Note on Code Swaraj The Courts Call the FBI on Us

Video was fun, but it wasn’t my main focus. That was the law. The law is what led me to a deep study of civil resistance. I began with case law, working with Harvard professor Larry Lessig to purchase all the U.S. Court of Appeals backfile from a vendor and posting them on the Internet. The archive cost us $600,000, but it was the first time these opinions were available for access without charge on the Internet. It was worth it.

After we did the U.S. Court of Appeals, I turned my attention to the U.S. District Courts, which ran a system called “PACER” (Public Access to Court Electronic Records) which provided access to briefs, opinions, dockets, and other materials, but at a charge of eight cents per page (now up to 10 cents per page). This seemed really stupid to me, so I put together a system for recycling PACER docs, with an extensive set of “Frequently Asked Questions About PACER” which went through the economic and technical flaws in this awful system.

This was in 2008, and soon my phone rang. On the line were a student from MIT named Steve Schultze and his buddy Aaron Swartz, at the time a freelance force of nature. I had known Aaron since he was 12 and was a protege of Larry Lessig and a frequent attendee at industry get-togethers. Aaron and I had worked together on various issues such as the IRS and he worked closely with my ex-wife, Rebecca Malamud, putting together the Open Library system for the Internet Archive.

Aaron liked my FAQ and decided to use the library system to start recycling at scale. Steve had written a simple PACER crawler and Aaron wanted to apply that. The courts had just set up an experimental service in 20 libraries around the country to see if “ordinary” people might want to use PACER. This was a bow to increasing pressure from Congressional officials who wanted to know why they kept getting letters from people asking about PACER. The courts thought a 2-year pilot investigation might be an easy stalling action.

Aaron took Steve’s code and wrote a bigger crawler. He also noticed that authentication for access on the library system was based on a “cookie,” which meant that the librarian would log in once at the beginning of the week at the terminal and then anybody could sit down for a week and use PACER. I’m still not sure exactly what Aaron did here, but I think he sent a buddy in once a week to the Sacramento library and copied the cookie and mailed it back to him. In any case, he got a cookie good for a week and was able to use it to crawl the system.