The Oil Drum

» see theoildrum.com as it lived, on the wayback machine

~ the eulogy ~

Almost nothing on the internet gets to die deliberately. Sites are sold, or the founder stops paying, or a company buys them and quietly deletes the servers on a Tuesday. The Oil Drum announced a closing date months in advance and kept publishing until it arrived, thanked its writers by name, and then froze. Whatever you think of what it argued, that ending is close to unique.

It began in March 2005, run by Kyle Saunders, a political science professor at Colorado State who posted as Prof. Goose, and Dave Summers, a mining engineering professor in Missouri who posted as Heading Out. It was published under a small Colorado nonprofit and it described itself as an energy and peak oil research site. Roughly twenty-five contributors wrote for it from around the world, most of them working geologists or engineers writing under handles.

The subject was peak oil, the argument that global production would top out and decline while demand kept climbing. The site got its first real audience after Katrina and Rita tore through Gulf production in 2005, and by 2007 Nielsen had it inside the top five sustainability blogs. Paul Krugman read it. So did James Howard Kunstler. In 2008 it won an energy education award from the American branch of the peak oil association.

Energy is not my field and peak oil was never a subject I followed. What I recognize in it is the shape, because engineers are the same animal everywhere: give them a public technical problem and a place to argue about it in front of each other and they will produce something better than what they were hired to produce. The Oil Drum is the best documented case of that I know about, and the case is the summer of 2010.

~ the dossier ~

~ the origin ~

For most of its life the site ran on a rhythm that no longer exists. Long posts with charts, published by people with actual field experience, followed by comment threads running to hundreds of entries where the disagreement was substantive and the corrections stuck. Contributors argued with each other in public under the article rather than somewhere else on a different platform.

It was wrong about the timing and it is worth saying so directly. The production peak the site spent years modeling was pushed out by hydraulic fracturing and American tight oil, and when The Oil Drum closed there were gloating obituaries in Forbes and Reason treating the shutdown as a verdict on the whole idea. The analysis of how hard the remaining oil is to get, and what it costs in energy to get it, held up considerably better than the dates did.

~ the summer of the spillcam ~

On 20 April 2010 the Deepwater Horizon burned and sank, and for the next three months there was a live video feed of a wellhead a mile underwater leaking into the Gulf of Mexico. Everyone on earth could see it. Almost nobody could read it.

The Oil Drum ran live comment threads through the whole thing. Working drilling and completion engineers watched the same ROV footage as CNN and explained, in near real time, what the top kill was doing and why the junk shot was failing, and what the flow rate actually looked like against BP's public number. Its posts on the blowout ran for months, thread after thread, and people who had never heard of the site arrived through search and stayed. It was, briefly, the best coverage of the biggest story on the planet, produced by unpaid people in a comment box.

~ the actual end ~

On 3 July 2013 the editors posted the notice. Eight years, and they were stopping. The reasons were unglamorous and completely honest: submissions of the quality they wanted had dried up and the contributors had drifted off into their own lives. Hosting a site with that much traffic and that much archive was expensive for a nonprofit carrying no advertising.

The original plan was to go static at the end of July. They extended it to 31 August so people could say goodbye properly in the threads, which is the most in-character possible thing for that site to have done. After that, no new posts and no new comments. More than ten thousand articles going back to April 2005 stayed exactly where they were.

~ the long tail ~

The archive is still up in 2026. theoildrum.com redirects to its own archive index and the whole corpus sits on Amazon S3 as static files, which costs almost nothing and requires nobody to maintain a database or patch a Drupal install. That was the point of freezing rather than closing.

It is a strange thing to visit. Every link works and every chart loads. Every argument about 2012 oil prices is preserved with the replies still underneath it, and not one word has been added in thirteen years. Most dead websites are dead because somebody stopped caring. This one is dead because everyone involved agreed it was finished and then took the trouble to leave the building standing.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »