Information central

I remarked a couple of days ago on Dave Winer’s observation that the enormous popularity of Wikipedia may represent a new kind of centralization of information. An analysis of the recently disclosed AOL search data, by SEO Blackhat, adds a further gloss to this phenomenon. According to the analysis, the #1 ranked search result garners, on average, 42.1% of all clickthroughs. The clickthrough rate falls off precipitously from there, with the #2 result representing 11.9% percent of clickthroughs. If current trends continue, Wikipedia will likely come to occupy the top spot for a very wide range of topics – in fact, it already holds the top spot for many common terms.

Combine the growing dominance of web searches over the way people find information with the growing dominance of Wikipedia over search results, and you do seem to have the makings for an unprecedented dominance of a single information source. Could it be that, counter to our expectations, the natural dynamic of the web will lead to less diversity in information sources rather than more?

And could it be that Wikipedia will end up being Google’s most formidable competitor? After all, if Google simply points you to Wikipedia, why bother with the middleman?

UPDATE: Patrick Ross goes further down the long tail of topics and finds that Wikipedia’s dominance over Google results only gets stronger.

4 thoughts on “Information central

  1. Tom Lord

    Which is worse as a central source of information? The transparent, not-for-profit (at least formally) but clearly f’ed up direct control structure of Wikipedia, or the opaque, for-profit, clearly f’ed up indirect control structure of Google’s ranking algorithm? Centralization is there one way or another and the question of how it plays out against, say, the values of The Englightenment is present in either case.

    Which is to say, you are right. We have various questionable entities competing, with all attendent economic efficiencies, to impose an unprecedented kind of hegemony. Ick.

    Hmm.

    -t

  2. Brad

    If Wikipedia is one of the “best” places to go on the web for info, that says something about the web qua library — and it isn’t good.

  3. seamusmccauley

    And could it be that Wikipedia will end up being Google’s most formidable competitor? After all, if Google simply points you to Wikipedia, why bother with the middleman?

    Because Wikipedia’s own search doesn’t work? Hardly insoluble, but since the solution would involve a deal with Google or equivalent…

    I also wonder how far the inadequacy of Wikipedia’s own search has driven the apparent popularity of the site on Google. I expect the Wikipedia results on Google get a lot more clicks than other sites that have functioning interal search facilities, and as users become habituated to the necessity of using a search engine to navigate around Wikipedia that’s a lot of clicks and a lot of possibly (relative) phantom search popularity.

  4. Scott Wilson

    I would be suspicious citing those click-through numbers as any sort of support for the “One World / One Wikipedia” argument (which doesn’t invalidate it on it’s own merits, of course). I think you would have to do a much more granular analysis of those searches to draw any such conclusions out of them. Taken from a different perspective, they may simply indicate that search engines are becoming very good at finding for you exactly what you were looking for.

    If there were some way to differentiate between searches where a person was simply casting about for general information (of the sort that Wikipedia offers) as opposed to a specific website or product, then that logic might work. But as it stands, there is nothing telling us whether those 42% clicked on that first link because it really was the most relevant information about that Shelby Cobra they’ve been wanting to buy, or if it simply happened to be the first thing on the page when they were poking about curiously for information on the history of Israel.

    Someone with more motivation than I could probably cast through the AOL search data and figure that out; until they do, I don’t think this argument holds a lot of water.

Comments are closed.