Escaping the Googlearchy

by Simon St. Laurent

Related link: http://www.princeton.edu/~mhindman/googlearchy--hindman.pdf



A few days ago, Len Bullard posted a link to "Googlearchy": How a Few Heavily-Linked Sites Dominate Politics on the Web. After reading it, it occurred to me that the situation they describe is precisely one I'm trying to escape.



It's not a new story, really - it's yet another case where network effects create winners and losers. I've been writing about network effects since 1991, when I was researching an article on the failure of the Susan B. Anthony that eventually ran in The Journal of Money, Credit, and Banking. John Caskey, the professor I was working with, suggested that network effects might be a good explanation of why it's so difficult to introduce new coins unless you pull the bill it replaces.



Network effects have come under fire when they sound like they undermine a free market and praised to the skies when the benefit adventures like the Internet and the Web. They're relevant to standards development, the interpretation of monopolies, and a wide variety of situations which start out as an open field but eventually become much less open.



The distributions which take place as patterns settle in have been discussed before in the context of weblogs, in Clay Shirky's Power Laws, Weblogs, and Inequality. I think Clay is right about what happens in such systems but wrong to shrug his shoulders and say "There just weren't enough blogs to have really unequal distributions. Now there are." The writers of Googlearchy don't quite shrug their shoulders; they're researchers pointing out that this field is worth studying, and that naive rhetoric about technology improving communications to create a more level playing field for different perspectives needs to be reconsidered. My favorite quote:



But whether this result is surprising or not, it is clear that in some ways the Web functions quite similarly to traditional media. Yes, almost anyone can put up a political Web site. But our research suggests that this is usually the online equivalent of hosting a talk show on public access television at 3:30 in the morning.


So how can people escape this dire world? Both Shirky and the Googlearchy researchers seem to recognize that community scale is an important factor. As people know less and less about each other, they tend to forge connections through people they do know, and those people effectively gain the ability to serve as a nexus. On the Web, that means links. As Shirky notes:



people's choices do affect one another. If we assume that any blog chosen by one user is more likely, by even a fractional amount, to be chosen by another user, the system changes dramatically. Alice, the first user, chooses her blogs unaffected by anyone else, but Bob has a slightly higher chance of liking Alice's blogs than the others. When Bob is done, any blog that both he and Alice like has a higher chance of being picked by Carmen, and so on, with a small number of blogs becoming increasingly likely to be chosen in the future because they were chosen in the past.


Think of this positive feedback as a preference premium.


The Googlearchy folks do note some categories of exceptions to this pattern, and this may suggest a direction for those of us who would like to contribute despite coming late to the race:



This study suggests that
communities where most sites have substantial numbers of inlinks are the exception, not
the rule. The communities that have previously been studied at length - public companies,
universities, newspapers - are all unusual, in that they represent groups in which there is
a high degree of mutual recognition among the actors (Pennock et al. (2002)).


These are all communities with a relatively small number of members, a sense of equality fostered by there only being a limited number of choices.



While the enormous bazaar of the web, globalization, and all these other delightful 21st century visions of glory may sound exciting, I think it's time to recognize that we're losing things people took for granted when there were fewer people in a larger world. Lots of small fish in big ponds growing more and more alienated from each other and from their own possibilities seems like a perverse result for something we've called progress.



That said, I think there's still room to escape this fate, though maybe it has its own downside. "Groups in which there is a high degree of mutual recognition among the actors" are generally pleasant places to be.



Small groups have different dynamics from large groups, and frankly, those dynamics are often less alienating, even when they're hostile. Geographic proximity, tying ourselves back to a sense of place rather than joining vast and unknowable communities, has its appeal. Similar interests can also bring people together, and I have to say that I've found some of the most interesting conversations on sites where people discussed matters of deep concern to them - but "them" was a small cluster of people. (I think most geeks have an unusual interest or two and may recognize the phenomenon.)



I'm not sure that the Googlearchy and the celebrity culture that power laws seem to propel are necessarily that important. They'll be there, and they'll have influence, but I can think of lots more exciting things to do, even on the Web, than read A-list bloggers I've never met.



Ever find a smaller community that feels comfortable? Start one?


2 Comments

mchampion
2004-01-10 19:12:42
I wonder if Vivissimo, et al. will change anything
I've very recently discovered www.vivissimo.com 's search result clustering technology. (You can use it online via that site, and there is some Windows-only toolbar thing too, but I haven't tried it). The point is that it doesn't sort results by something akin to Googlerank, it classifies them on the fly according to some statistical/linguistic analysis (apparently not a predefined ontology). This means that you see the "communities" that actually exist out there, not the contaminating effects of the power law.


My personal test of a search engine is how fast it takes to find me. You might think I have a fairly unusual name, but there are a number of "Mike Champion"s out there with a lot more celebrity than I have, and I usually have to add some qualifier to find myself in Google in the first few pages. Vivissimo puts the actor, baseball player, and XML geek named Mike Champion in separate categories. (Unfortunately, it also puts Mike Tyson in the list, apparently because he's both a Mike and a "champion"!)


Likewise, a certain prolific weblogger regularly notes that he is the "Dave" at the top of the Googlearchy. The same query in Vivissimo finds the various bloggers named Dave, but a lot of other Daves are easily found and put into reasonably descriptive categories.


Presumably, with Vivissimo, Grokster, etc. on google's heels, they'll be doing something similar real soon now. (And if not, well, Father Darwin is waiting with open arms and their post-IPO stock won't be worth much, because the Googlearchy is becoming less and less useful in the real world). In any event, one can at least hope that Googlearchy's will open up a bit to us non-celebrities who don't give a rat's patootie about celebrity culture.

johnwcowan
2004-03-17 12:09:59
Mike Champion ueber alles
Funny, the top Google hit for "Mike Champion" with or without quotes right now is xegesis.org, and the next two are to a couple of xml-dev postings. Only then do we get to the baseball player.


The Googlearchy at work?