Opinion: We need to build systems into user-supplied content sites that provide the necessary checks and balances to create quality.Not surprisingly, I managed to kick up a bit of a tempest in the blogosphere last week when I wrote that Weblogs (and Podcasting and video and RSS and social networking apps) were "fads of the future."
I wasn't going to (and am still not gonna) write to defend my position this week, but I was surprised by this little item in The Register reporting that Wikipedia co-founder Jimmy Wales has "acknowledged that there were real quality problems with the online work."
Putting two and two together, I began to think about the limitations of user-supplied content and how they impact what we do in our Web development lives, especially as we careen wildly towards "Web 2.0."
I work at a college and know for a fact that many of my colleagues are constantly fighting an uphill battle to get students to use reliable sources for their research (much less cite those sources).
Barely a day goes by when some professor doesn't report that one student or another has turned in a paper citing some unverified Web source. When called out on it, the students usually look mystified: "But it was on the Internet" they moan.
One of the problems with the Web (and one of the benefits) is the "flattening" effect it has on content. For the most part, every site appears the same, glowing on our screens with perfectly displayed text, glossy pictures, and set in the context of all the other sites we see.
Unlike the analog world, where cues as to authenticity are usually pretty apparent in physical features such as craftsmanship, feel and smell, on the Web things tend to look the same, whether they're from an authoritative source or some crackpot cooking up conspiracy theories in their basement.
What are the pros and cons of building a corporate blog? Click here to read more.
This isn't to say that there aren't online cues to credibility—Consumer Reports' Webwatch has done some pretty interesting research in this area—but those cues are usually harder to detect to the untrained (or uncritical) eye.
Enter Wikipedia. Billing itself as "The Free Encyclopedia" and presenting a unified, reasonably well-designed face to the world, it has all the cues of "authority" that people are looking for. A naive student (or adult or human of any kind) coming to Wikipedia is going to see "encyclopedia" and see that it wasn't designed by a lunatic and assume that the information inside is correct. Some is. Unfortunately, some isn't.
The problem is that unless you're an expert on a topic, there's no way of knowing. And if you were an expert on a particular topic, you'd probably have no need for Wikipedia, right?
Wikipedia (and other user-contributed projects) are based on the open-source model of software development that has worked so well to bring us Linux and countless other wonderful tools and toys.
The theory is that putting stuff out there in the global Web provides sort of a hive-mind, checks-and-balances system where bad content is identified, ferreted out and replaced by good content. There's a lot of knowledge out there in the world, says the theory, and the bad stuff will be replaced by good stuff in the marketplace of ideas.
It's a good theory and one that has worked fairly well with software, but one that's a little tougher when you're talking about meaning. Computers are binary and a program either works or it doesn't. Feedback about bugs can return to a networked development team that then works to remove the bugs. They release a new version to the Net, people test it, feedback returns and the process continues.
All parties have a vested interest in working towards perfection: The social capital of the development team rests on producing good software, and the pragmatic needs of the users rely on having good software tools to get their jobs done (or amuse themselves with fun games). The loop works because everyone has a stake in the betterment of the project.
Software is one thing. "Meaning" is another. It's a lot slipperier and not quite as binary. Sure, untrue statements can be ferreted out pretty quickly, but where a statement slides into interpretation…well, that's a different story. The same goes for the quality of the writing: what constitutes "good" writing is open for interpretation. What constitutes a bug-free printer driver is much clearer. The "quality problems" with Wikipedia are a lot more to do with meaning and writing then they do with the veracity of the facts (though apparently there are a lot of facts in question).
Many breathless netizens have an almost religious fervor in their voices then they talk about how mass collaboration has brought "power to the people." And I'd say that I agree with them for the most part…just go check out some of the amazing stuff happening with video Podcasting (Tiki Bar TV is probably my new favorite). But just like the entries on Wikipedia, for every brilliant new production, there's also loads of dreck out there. Just because it's online and contributed by neophytes doesn't make it good.
Nicholas Carr calls the unquestioning, breathless love that many exhibit for user-contributed content "The Cult of the Amateur." Those in the cult see everything that's new, unfiltered by the profit motive, rough, and espousing the ideals (as Carr defines them) of "participation, collectivism, virtual communities, [and] amateurism" as an unquestioned good. You find this viewpoint in the glowing eyes of those who wax philosophic about Web 2.0, about Wikipedia, about social networking apps, and (yes, I'll say it) about blogs.
But it's not that new a concept. Stretching back to the optimism and celebration of simplicity which characterized the Romantic writers of the 18th and 19th centuries to Italian Neo-Realist films of the 40s and 50s to the utopian ideals of the 60s all the way to the breathless "Long Boom" hype of early cyberculture pioneers, amateur authenticity has always been a privileged state.
This is what's going on with much of the hype surrounding Web 2.0, blogs, RSS, Podcasting and videocasting, and, yes, Wikipedia. It's not that any of these things are "bad" in and of themselves. As far as I'm concerned, they're neutral as far as publishing mediums go. But what we need to do as we move forward into the lands of greater user-supplied content is to keep a critical eye, and be willing to say, "Yeah, that sucks."
There's wheat and there's chaff. There's gold and there's crap. We need to build systems into our open-source and user-supplied content sites that provide the necessary checks and balances (a la the whole Linux development methodology) to assure quality and accurate content. We can't just assume that because something is coming from "the people" that it automatically has value.
Online Media 
