Wikitravel: Functional Copyleft Content for Travelers

Detractors from the idea of Copyleft Content often point to the fact that non-software expression is not functional in the same way that software is, and therefore doesn't need the same user-maintenance mechanics that software does. Oh, well, except software documentation. But that's it, right? Is any other type of non-software information functional? Perhaps Wikitravel is.

The idea of Copyleft Content, that is, freely redistributable and editable non-software information based on the Free Software model, has gotten off to a rocky start. A number of Copyleft Content licenses have been developed, and the Creative Commons project has been started, but the number of texts, musical pieces, visual arts and videos available under open licenses has been less than impressive.*

This dearth of Copyleft Content has a number of possible causes. One may be that there is a greater infrastructure for independent publishing of proprietary pieces than for software, meaning there's less of a need for the Free Software model from the creator's point of view.

Another might be the greater personal and emotional investment that writers, musicians and artists put into their work than programmers put in their programs -- although there are definitely counterexamples. Few artists can stand the idea of someone tinkering with their brainchild; when the artist has put so much of themself into the piece that the artist is the art, it's a very personal violation.

A last may be the subjectivity of non-software information. Most programmers are happy to see bugs in their software corrected, but how many writers are willing to see a short story "fixed" to have a different ending? It's easy (ish) to determine when a program is operating poorly, but whose opinion matters in determining how a musical piece could be improved?

A frequent, and I believe factious, reason posited for the lack of Copyleft Content is that most content isn't functional. End users don't have the same physical need to improve content the way they need to improve software. Users don't have to "scratch an itch" to fix a song they're not happy with; they merely shrug their shoulders and hit fast forward to move on to the next one.

In fact, there are many forms of content that are very functional -- besides software documentation. In the area of reference material, Copyleft Content development has advanced by leaps and bounds. The poster child for Copyleft Content reference works is, of course, Wikipedia, a free encyclopedia. It's collaboratively edited by thousands of contributors, using the Wiki model, where any Web reader can edit any page.

A wiki development model helps break down another barriers to Copyleft Content. Wikis tend to discourage personal "ownership" of content -- in the emotional rather than legal sense. But, most importantly, there's less subjectivity in reference works than in fiction or art. A reader can clearly tell that World War I ended on November 11th and not November 12th or 10th; the factuality of the work makes bugs more obvious.

This article's point, however, is to discuss another Copyleft reference work and its domain of content: Wikitravel. Wikitravel is a project to create a free (as in freedom), complete, up-to-date and reliable world-wide travel guide. I believe that travel is an area ripe for development by copyleft content advocates.

My fiancee and I got the idea for creating Wikitravel on a trip through Southeast Asia last winter. We arrived on an island in Thailand by ferry in the dead of night, and set off on a dirt road to find a cheap and quaint hotel listed in our guide book. After a few miles of walking with our heavy packs, we arrived at the site -- and found only a few broken 2x4's sticking out of the weeds. There may have been a hotel there once, but it was long gone.

I was angry about the inaccuracy, and of course having to trudge back to the docks to stay in one of the shabby rooming houses there, but my fiancee was even angrier. "You know the worst part about this?" she asked. "We're not the only ones who are going to do this. There are going to be hundreds of people -- maybe thousands -- coming out to this empty lot for years, because the guidebook isn't up-to-date. The guidebook won't be updated for another year, and the chances are slim that they'll actually come out and check that this place still exists. Nobody else can learn from our mistakes!"

Travel guides are a classic example of the Cathedral model of development. A small editorial staff contracts with one or two writers to go visit a country, city, or region. Those writers mostly check for new attractions or things to do, and don't spend a lot of time verifying existing listings. They just can't -- it would take years for most regular-sized guides. So there are thousands or tens of thousands of travelers depending on the information provided by just a handful of writers and editors.

What we needed was a Bazaar model for travel guides -- a way for those tens of thousands of travelers to add, delete, update, and enhance all those restaurant reviews, siteseeing listings, hotel guides, and train schedules. These people are out verifying -- and cursing! -- the information in the guides *every day*. Why shouldn't they be correcting the guides as well?

I had experience with Wikipedia, so of course I saw a solution in the wiki-plus-copyleft model of content development. We were on holiday, of course, and we didn't settle down for a while, but once we were back in a stable spot -- Montreal, July 2003 -- we installed some Wiki software and began the experiment.

It was instantaneously popular, and for good cause. We had stumbled into a culture that was about as close to classical hackerism as can be gotten outside of Cambridge 1972. Independent travelers are much like hackers in a lot of ways. They have similar needs and motivations:

• Travelers have a culture of sharing knowledge. When travelers meet in train stations, airports, restaurants or bars, they share data about where they've been, ask for info about where they're going. Travelers feel comfortable -- even proud -- sharing data about where they've been.

• A traveler's value and status is based on what they know. The more they know about one destination, that is, the more deep their travel knowledge is, the more it's apparent that they take travel seriously. The more multiple destinations they know about, that is, the more broad their travel knowledge is, the more it's apparent that they have traveled far and wide, and that they make travel an important part of their day-to-day life.

• Travelers distrust authority. There isn't a traveler in the world who hasn't been burnt by so-called "official" information, much as we had. If it isn't public sector information -- subject to political influence that mandates that Redding is as attractive a destination as San Francisco or Los Angeles -- it's private sector info, up for the highest bidder, where the hotel or restaurant that "sponsors" the data gets the best reviews and fawning link location.

• Travelers put greater value on peer acceptance than on any recognition of authority (whatever that authority would be in the travelers' world).

After a few months in operation, we have over 500 destination guides in Wikitravel, with 10-20 new guides added every day. Each guide is available under the Creative Commons Attribution-ShareAlike 1.0 license, a copyleft license that ensures that data added to Wikitravel will be Free (as in freedom) as long as it's findable on the Web, or elsewhere.

We think that Wikitravel is a great addition to the world of Copyleft Content available on the Web. We plan on taking our first trip with only Wikitravel guidebooks as our info source over Thanksgiving weekend this year. We have only to wait and see whether the rest of the Internet finds it as valuable as we do.

*That is, of course, if you don't count public domain works as Open Content. Up until the Sonny Bono Copyright Extension Act, public domain content was increasing exponentially. It's still trickling in, of course, but it's generally not new.