You Think Knol Sucks? I’ve Got the Perfect Knol Article For You…

I typically agree with Slate technology colunnist Fargad Manjoo about just about everything. True Enough resonated particularly with me. It seems at first to be just another book about culture wars and the decline of journalistic integrity, but it pushes the discussion farther than I’ve ever seen it go.

As Eric wrote in a July review of the book,

Manjoo observes that we, the body politic, used to agree on what was happening and the problems we were facing, but had different ideas about how to address those problems. Now we can’t even agree on what reality is.

Unfortunately, Farhad is way off on his criticism of Google’s Knol product. He compares Knol to Wikipedia, using the worst possible point of comparison: the consistency of quality.

Wikipedia has developed strict rules and a deletionist community culture specifically with the goal of maintaining a consistent quality of article content.

On the other hand, Google doesn’t do this it. It doesn’t try to do this. If Wikipedia is a farm (or a plantation), then Knol is a jungle.

Yes, there’s tons of spam. Yes, some articles are a lot worse than others. But Google’s search engine was built primarily to “rank” pages and filter out the junk, if I’m not mistaken. “Perspective and style don’t scale,” Farhad writes. These qualities do indeed scale. It’s just that they’re lost in the midst of junk, and search engines should help us filter out the noise.

I’m living proof that there’s some good stuff on Knol. I have exactly two articles published, and I’m proud of both of them.

• The Pasteurization of Knol

  This also isn’t really an article, but rather a satirical children’s story about the Semantic Web: Alice is a partner at prestigious Silicon Valley VC firm Alice & Cooper.She’s been approached by the hyped semantic web search engine company Looking Glass, Inc. to participate in their series A multimillion dollar fundraising round. The company promises a “killer app” service that will take questions from users about people, places, things - just about anything - and return accurate factual answers, instead of just a list of relevant sites. The engine is able to “reason” by processing user queries against a huge dataset of semantically linked data.

• A Perplexed Guide For the Dumb

  Less of an article than a nano-sized Samuel Beckett play, this old college assignment stands among my personal favorites of anything I’ve written. It starts off with a simple premise:

  Having read all the books, they forgot where had the books gone.

  …and it goes from there. The 50 footnotes are all real literary quotes, each from a single book. I did use electronic copies of these fifty books, which is obviously cheating (and as a result I didn’t have page numbers to cite) . This story was indeed entirely plagarized from works of fiction. This literary conceit itself is stolen from Jonathan Letham.

Farhad is right that Knol is in need of some tuning up before it can be taken seriously. Here are two easy suggestions:

1. Junk needs to quickly fade out, and search should be as good or better than it is for your web searches.

2. Google Analytics for Knol articles.

In short, Knol is not solving any important need, or pushing farther the quest to organize the world’s information. How about adding a few microformat templates? How about opening up a microformat developer platform? How about SearchMonkey style results for Knol content?

But Farhad doesn’t make these kinds insightful suggestions. He instead concludes by once again unfairly comparing Wikipedia and Knol: “The problem is that we don’t need the next Wikipedia. Today’s version works amazingly well.”

Farhad, I implore you to read The Pasteurization of Knol. It just may change your mind.