MemoryWiki: The Encyclopedia of Memories

Marshall Poe

Though we are hardly aware of it, there is a populist revolution sweeping the Internet. The most prominent example of this ongoing transformation is Wikipedia, a free encyclopedia produced on an open source platform. Until recently, encyclopedias have been created by professional editors, experts and publishers, each working in serial stages: the editors organized production, then the experts wrote the entries, and then the publishers distributed the books. Making an encyclopedia was expensive, time consuming, and the tomes themselves were, all things considered, neither complete nor very well distributed. And as things changed, they couldn't be updated. Wikipedia has changed all that by empowering anyone to contribute to the encyclopedia at anytime. Want to help? You may not have any formal training, but there is doubtless something you can contribute. You can edit existing entries, add new entries, or publicize the project. And you can help right now, so long as you have an Internet connection. The results of this remarkable experiment in simultaneous on-line collaboration have been nothing short of remarkable. The English version of Wikipedia (there are over one-hundred non-English versions) has one million plus entries, the vast majority of which are quite accurate.

Wikipedia is an example of "distributed content production" (DCP for short). Thanks to the read-write web (that is, web applications you can interact with), DCP is everywhere. Ebay's reputation system, Netflix's preference profiles, MySpace's networking utilities and the social bookmarks of Del.icio.us' (http://del.icio.us/ ) all rely on DCP. In each of these cases, you transmit what you know through a website to a central database that is then made available to all users. As more people contribute, the DCP resource becomes more valuable to all users in a kind of "virtuous cycle."

Historians have not joined the DCP revolution. To a significant degree, the way historical sources are gathered today mirrors encyclopedia creation of old. In the stead of editors, experts and publishers, we have archivists, organizations, and repositories. The archivists decide what should be kept, the organizations duly transfer the relevant documents, and the repositories store them. Nearly every modern government and large institution has a formal mechanism by which their "significant" papers (and increasingly other media) are put away for posterity. Though far better than nothing (as any premodern historian will tell you), this system suffers from a cardinal deficiency, namely, it does not capture the lived experience of regular people. People not only make history, they also experience it. But what is left after the dust settles is all too often nothing but dry sources "from above," that is, the jetsam of big institutions and the people who run them. Perhaps this is as it should be: from a purely explanatory point of view, some people and paper is more important than other people and other paper. The Declaration of Independence is surely more interesting than Jefferson's laundry list. That said, it does seem a shame that we so systematically miss the stories of ordinary people as they experienced "historic" events. And no amount of oral history gathering or organized attic rummaging is going to capture a goodly portion of that lived experience, though it does give us a tantalizing glimpse of what is forthcoming.

There is now, however, a way to ensure that everyone has an opportunity to share their stories of historic events with the future. In the fall of 2005, a class I was teaching at American University set out to use DCP to gather common stories of historic events and create thereby an "encyclopedia of memories." We began with nothing more than the basic software that runs Wikipedia and a vague notion that, properly implemented, this technology might allow us to craft a web portal that would permit individuals to record their memoirs of everything. The code in question is a species of "content management software" called a wiki. Designed by a software engineer named Ward Cunningham in the mid 1990s, the wiki is part blackboard and part database. It permits any untrained user to create and edit pages right from a web browser--no complicated downloads or registration required. You just go to the wiki and type. That's it. Moreover, it allows those pages to be saved in multiple versions (so reversion to earlier or better pages is possible) and categorized (so a kind of index can be built from the bottom up). What's in the wiki and how it's arranged depends on the community of users, not an editor (though there are editors in some cases, and they are important). The wiki was first used by communities of programmers to design open source software, but the utility of this format to other tasks slowly became clear. Wikis spread. Today there are thousands of wikis being used for collaborative purposes by hundreds of communities, the most notable of which is the huge Wikipedia community.

Our community began with me and twenty-five undergraduates. We configured the wiki to permit anyone with an Internet connection to add memoirs and categorize them. We opted for an "open" model of content production: anyone could add a memoir--short or long--about anything. In order to preserve the integrity of memoirs, we instituted a policy by which memoirs would be copyedited and categorized by "site stewards" after which they would be locked, that is, no further edits could be made without the express permission of the author.

We launched the website in October 2005 under the name "MemoryWiki" and began to publicize the site. Getting the word out proved quite difficult (the web is huge, and it's hard to find anything new), but we persevered. We wrote our own memoirs just to get the ball rolling and to provide examples to those who might stop by. And slowly people began to stop by, people from all over the planet. Word spread with the help of friendly bloggers, and soon we were getting several hundred unique visitors every day. People, it turned out, had been "there" when "that" happened, and they wanted to record their impressions for posterity. One hundred memoirs became two hundred, two hundred became three hundred, and three hundred became four hundred in a matter of six weeks. At presstime, we have over 435 memoirs. We have plans to launch separate non-English language sites soon.

Historians use archives to write history. But perhaps, as our experience with MemoryWiki shows, they should be creating archives so that future historians can write better histories. People want to tell their stories, and the means to allow them to do so is now available. Go to <http://www.memorywiki.org> and see. Write a little something; remember, the site becomes more valuable the more people add to it. And if you want to lend a hand, please contact us.

Marshall Poe is the cocreator and editor of MemoryWiki and a contributing writer at The Atlantic Monthly. For more information, visit <http://www.memorywiki.org/>.