Newsless.org

@NicholasAllen asked me today what I thought about the Omaha World-Herald’s acquisition of the hyperlocal wiki site WikiCity.

When Gina Chen, who wrote up this bit of news on NiemanLab, first wrote about it in August, Perry Gaskill left a comment I think is still trenchant:

Sorry, Gina, but it strikes me that WikiCity could serve as a poster child for what’s generally wrong with the direction of hyper-local news efforts. Once again, what we’re seeing is a quasi-franchise business model based on selling low-CPM ads against freely generated content. Nothing special.

Spend any time wandering around WikiCity, and what you find is the same dog who doesn’t bark. No sense of each town’s quirkiness; no sense of place. Instead of a local cafe where the cook knows you like your eggs scrambled, you get an Egg McMuffin.

I’ll allow myself some snark here, ’cause I think it’s deserved. I would bet that most of what you need to know about this acquisition can be gleaned from this sentence in the World-Herald’s article about it: “WikiCity (http://www.wikicity.com) has more than 13 million Web site pages and is one of the largest ‘wikis’ in the world.”

How many of those millions of “Web site pages” do you think was ever touched by a real person? And how many will ever be seen by a single person?

WikiCity in its current state strikes me as a textbook example of a site built by robots. Such sites tend, in my experience, to appeal mostly to other robots.

Contrast it to Wikipedia, whose every page was built, word by work, link by link, on the actions of individual people. Or to Everyblock, whose pages run on powerful algorithms, lovingly engineered and hand-polished by a brilliant and careful team of makers. These are large sites built on millions of niches, but neither were built that way to start. Wikipedia began as a small collection of pages that became a massive collection over time. Everyblock started as a selection of data sets in a handful of cities, and has grown over the years to encompass hundreds of data sets in more than a dozen cities. They started small and built up, like every success story I know, rather than the reverse, which is the WikiCity approach.

“Scaling down” remains a problem for the Web, on site after site. Sites such as Wikipedia and Delicious function beautifully in domains where they can garner enough attention. If a Wikipedia topic is significant enough to draw the interest of even a dozen editors in a few months, chances are it will be pretty decent. But the more niche you get on Wikipedia¹, the shallower and spottier the pages become. Look for a popular topic like “usability” on Delicious, and you’ll find a wonderfully curated selection of links, courtesy of the wisdom of crowds. But for a significant topic outside the site’s core niche of designers and techies, Delicious underperforms.

Howard Owens has written passionate criticisms of approaches to “hyperlocal” news that start with a giant, anonymous maze of computer-generated pages, all alike, all imagining that users will spontaneously arrive to populate their pages with genuine, quality material. Everything I’ve seen tells me Howard’s criticisms are right. These efforts are attempts to bring a mass mentality to a niche world. I’ve never seen a successful wiki that wasn’t built like Wikipedia, from the bottom up, page by page.

If I were advising the World-Herald, I’d tell them to reboot WikiCity and start building a wiki just for Omaha. Better yet, start with just one of the city’s six regions. Build on what you can from Wikipedia – giving proper attribution, of course – but begin with the understanding that it’s not going to be very complete just yet. Assign someone to add as much information as they can to the site every day. Create a content plan to prioritize what information you’ll pursue first. Early on, create pages for the most trenchant issues affecting the neighborhood; diligently and prominently link to those pages when the issues appear in your coverage.

For months, I expect this exercise will seem like a neverending, pointless slog, and no one will join in. After a few months, your traffic will still be underwhelming, but you’ll notice a tiny stream of fellow-travelers who’ll timidly participate here or there. Keep at it, and in a year, you’ll have a small but dedicated community. And you will probably have built something more significant than you had realized. After two years, it will begin to seem like it was worth the investment.

Come to think of it, that last paragraph could probably be applied to most successful businesses on the Web.

1. That’s my neighborhood [↩]

Thanks to everyone who voted for my SXSW session! It was confirmed among the first batch of panels to be included in the festival.

Now comes the fun part. Over the next couple of months, I’ll be setting up a website for the panel, which I hope will be a great resource for anyone looking for what’s being tried and what’s needed to create a more contextual Web. There, we’ll begin collaboratively setting the agenda for the panel. I hope you will all participate in that process, and I hope to see many of you in Austin in March! Thanks again for voting.