If I were Omaha World-Herald Publisher Terry Kroeger, I’d be booking my post-holiday flight immediately.
As Gina reported here last week, Omaha’s employee-owned metro daily just bought WikiCity, an Omaha-based Web startup that wants to provide mini-Wikipedias for every city in the country. Creating a cheap platform for evergreen, user-generated local Web content has been tried, um, once or twice before. But with some notable exceptions, corporations have turned out to be really, really bad at this.
Philip Neustrom hasn’t.
Today, the quirky 500-page wiki Neustrom launched with fellow UC Davis math student Mike Ivanov in 2004 has 14,000 pages and drew 13,000 edits by 3,300 users last month, averaging 10,000 unique visitors daily. More importantly, it’s the best way in town to find a lost cat, compare apartment rental prices or get a list of every business open past 10 p.m. Operating budget, not counting its founders’ part-time volunteer labor: about $2,000 a year.
What’s the secret? Neustrom, who now wrangles code for the Citizen Engagement Lab in the Bay Area, was nice enough to tell us.
Wikis need content to breed content. Or, as evergreen-content guru Matt Thompson put it last week, a wiki written primarily by robots will appeal primarily to robots.
“Starting anything is hard,” said Neustrom, now 25. “The issue is predominantly an issue of outreach, of coordinating people and making sure people understand that they can’t just put something up there and add 50 pages and walk away, and then come back in a month and hope that it’s taken off.”
Instead, Neustrom, Ivanov convinced some of their friends to spend four summer months writing snippets about things that only exist in Davis, like drunken biking through late-night fog, oversized playground equipment and the smell from the cow farm on the edge of town.
“We were just trying to do something that we liked,” Neustrom said. “We certainly weren’t trying to do anything that was very useful.”
Business information is the holy grail. Pages about your local toad tunnel are dandy, Neustrom said, and quirky content kept the site from feeling generic to early users. But the feature that made DavisWiki take off was what the traditional media calls “consumer reporting.”
“After we’d sort of seeded it with 500 pages or something like that, we opened it up to the public,” Neustrom said. “First, it was pretty slow going. Nothing really happened.”
Then, sometime in late 2005, pages on things like lunch specials and Davis’s nicest bathrooms started filling up. Local business coverage has been “a big driving force” ever since, Neustrom said. Today, he said, retail businesses in town often keep their own information on DavisWiki up to date.
A wiki’s strengths kick in after one year. The web craves news like kids crave sugar. Blogs and tweets are gobbled fast and burn quick. But wikis are the whole grains of the web: One year after news breaks, someone will want to find and link to it again — and a wiki is likely to be the only place it’s still hanging around.
“All of the existing online resources for sort of cataloging anything about the town were sort of time-based,” Neustrom said. “After about a year and a half, these things would sort of disappear, even if they’d been around for 100 years, like the local newspaper…So we became the resource of record.”
Start with a subculture, then build out to a general audience. DavisWiki has always aspired to cover its whole town, but it’s always served students best.
That’s all right, Neustrom thinks. If he’d tried to please everybody who showed up, no one would have come back.
“When building something like this, you can’t just aim for this wide spectrum at first,” Neustrom said. Some companies try to launch wikis by writing programs that “crawl through a database, that spit out statistics and create 13 million pages and put that out there and hope that it’s going to stick. You can’t do that. It’s just not going to work.”
Neustrom, who spent 2004 sharing a house with musicians, found his base among the artsy, but he thinks any subculture would do. “You could have, like, a physics grad student start a community for their town, and it’s a bunch of physics nerds,” he said. “And that could spiral out and out.”
Keep your content open source, no matter what. Don’t do it for marketing reasons or out of the kindness of your heart. Do it because it’s the only way to guarantee to your users that if you fold, all their hard work won’t die with you.
Good wikis inspire rabid devotion — if they don’t, they never become good wikis. Neustrom and Ivanov keep their budget online and think of the project as a user co-op. Their users did, too. “There are people on there who literally spend four hours a day looking at DavisWiki,” Neustrom said. “People had free [computer lab] pages every quarter, so they would use their excess printing to print out 400 fliers and staple them to every room on campus.”
People don’t do that for sites they think are “neat,” Neustrom said. They do it for sites they own.
Don’t get hung up on mimicking Wikipedia. Sure, it may be the most useful object ever created by human beings. But as Marshall Poe showed in his terrific biography of Wikipedia’s youth, its rules — universal editorship, neutral point of view, no original research — were forged out of year-long flamewars among the early Wikipedians. Neustrom and his friends didn’t think NPOV was suited to an inherently Davis-centric site, so they ditched it.
“People want to be able to search for all elementary schools within a certain radius of a certain point, or all of the restaurants that serve vegan food,” Neustrom said. “MediaWiki suffers the same issue [as Sycamore] — it was written before the advent of modern web framework.”
Neustrom is yearning for a modern wiki platform. That’s why he’s been messing around with Django this year. It’s also why he’s incorporating Wikispot, the nonprofit he set up to reproduce DavisWiki for other towns and topics, as a 501(c)3.
Looking for a tax write-off, Terry?
Photo by Arlen used under a Creative Commons license.