Wikipedia meet Google

December 13, 2007

Marshall Kirkpatrick has the scoop on the knol project.

What does it mean in the long run? I don’t have the foggiest idea of the ultimate impact, but speculation is always easy. I have been critical of Mahalo from the start, arguing that a human powered search engine would never have the breadth of an algorithmic one (Google), and that despite gaming and SEO crap, a hyperlink-powered web was the most efficient way of identifying relevance.

Obviously Wikipedia does put some holes in my theory, but even Wikipedia becomes relevant only when people link to it. So Google’s new move is interesting. IMO, it’s value comes into play in scenarios similar to the one on Wikipedia and Science. In other words, algorithms find information while the human filter provides expertise. Google plans to share revenue with authors as well, which seems to be their differentiator from Wikipedia.

As an expert on a topic, would you contribute to the knol project? What do you think this is going to end up as? How does this impact Wikipedia? Will it end up being the comprehensive science resource we all talk about? Only time will tell. For now this appears to be a direct assault on Wikipedia and the planned search project.


Technorati Tags: ,

Posted by Deepak Singh | Filed Under Audio/Video, Search, Software & Internet 

Amazon Recommends

Comments

Viewing 5 Comments

    • ^
    • v
    It's interesting, isn't it. Although hard to see it as anything other than an attempt to compete directly with Wikipedia at the moment, as you say.

    I don't know that I would contribute to knol - simply because writing long, authoritative articles takes a lot of time. In fact this is where a lot of systems fall down - they rely on substantial contributions and only a tiny fraction of the audience are willing/able to contribute. It takes almost a borderline autism to write long articles and submit them to the ether, unsure if anyone will ever use them. Trust me, I'm a blogger ;)

    Perhaps a better approach to "fact checking" is some sort of voting system. Rather like that Google Images site where you're presented with images and assign them tags. Imagine "Google Factoids": present users with statements such as "the earth orbits the sun", vote true or false, get a consensus [1]. Now extend that to bioinformatics: get a broad idea of the users background, then present them with "colicin E9 interacts with TolB" - yes, no, don't know. Count the votes and you have a true wisdom of crowds.

    [1] OK, this might only lead to despair at how little people know, but you get the idea :)
    • ^
    • v
    Speaking as a Wikipedia editor - we're not actually all about running a hugely expensive website with no ads. It's about creating a resource for the future which anyone can take and reuse freely, not just making a cool website today.

    It looks like they're bending over backwards not to make this actually freely reusable content. Which, y'know, they could easily do (all editors agree to release their work under GFDL or CC-by-sa or something). So that immediately makes it less interesting to Wikipedia in terms of what we're actually doing. This immediately places Knol with about.com and Scholarpedia. I wonder precisely what rights over other people's work they're going to try to claim.

    If the quality of the work is good, we'd probably use it for references, like we do about.com.
    • ^
    • v
    Speaking as a Wikipedia editor and Wikimedia Foundation volunteer - The PNG example (http://www.google.com/images/blogs/knol_lg.png) shows a CC-by-3.0 tag. As far as I'm concerned that's a BIG WIN for Wikipedia and what we do - making free content *normal and expected*. If they require contributions to be under a proper free content licence, then I'm a BIG FAN of this endeavour. Same reason Citizendium succeeding would be a big win for what we do - it's not competition, it's expanding the pool of unencumbered knowledge.
    • ^
    • v
    David,

    As someone who wears a creative commons shirt twice a week (and has contributed to Wikipedia), you can imagine how much I admire the Wikipedia effort. I don't know the consequences of this entire effort, but you have to admit, this seems to be a counter on Google's part to Jimmy Wales' wiki-based search engine (not Wikipedia per se having slept over it). One could hardly think that they would be going after Mahalo or Squidoo.

    Is there room for Wikipedia Knol. Sure. In fact, if Google did not use the CC or similar license, I would continue to favor Wikipedia as a resource and NOT be a contributor. Call's for another blog post
    • ^
    • v
    Actually, "knol" was just the codename for this project.

    According to some sources, the actual name is "Unipedia"

    Sounds like a better name than "knol".

    I quite like the idea that Unipedia will share revenue of any ads with the authors.

Trackbacks

close Reblog this comment
Powered by Disqus · Learn more
blog comments powered by Disqus