The wild, wild Web
IF you go searching for God on Google, here’s the first thing you’re likely to find: a video essay titled “The Interview With God,” which serves up a breezy Q&A; with the Author of Creation punctuated by New Age-y piano riffs. (Spanish speakers making the digital pilgrimage will get, as their No. 1 search result, an invitation to a gig by a Hawthorne band called “Dios.”)
Why “The Interview With God” and not the Book of Job or the Bhagavad-Gita? Because Google’s busy little algorithmic brain tends to give priority to websites that link to the most other sites with similar content, and satisfy other mystic criteria known only to Google’s inner circle of geeky utopians.
It may well be that “The Interview With God” reached its lofty perch because it’s profound, or at least profoundly popular. But at this point in the Web’s evolution, it’s no stretch to guess that, the site’s merits aside, someone connected with it probably knew how to, ah, finesse the Web’s search-engine mechanisms.
For one naive, shining moment in the ‘90s, the assumption was that on the Web, popularity would be democratic, earned one enthusiastic click at a time. Pure. Simple. Untainted by Billboard, Hollywood, Nielsen or other mainstream media usual suspects. But that was before clicks meant cash, and before a flood of tools and communities brought millions of new, mainly nonprofessional content providers online, jostling to get their videos watched, audio clips downloaded and blogs and Web pages linked to bigger, more popular blogs and websites.
This intensifying contest has stoked the imperative to be “most viewed,” “most e-mailed,” “most played.” And that, in turn, has led to a gamut of strategies for one-upping the competition.
Today, the name of the game is gaming the system. And there are so many ways to do it that, whether it’s your son’s alleged number of MySpace “friends” or a confab with the Almighty, if the ranking is inordinately high, a certain amount of caveat emptor is probably called for.
Over at YouTube there were accusations this year that certain videos were pushed up most-viewed lists by viewers using fake account names. Spammers keep finding ingenious ways to clutter up our e-mail in-boxes by using code words that do end runs around filters. One site launched pop-up windows, then counted them as hits.
Then there’s “Google bombing” or “Googlewash,” in which individuals or groups (say, members of a political party) attempt to boost their own Web page rankings, or else discredit their enemies by linking and cross-linking them to negative sites (which is why, for the last two years, a search for “miserable failure” has turned up a bio of George W. Bush). Some Wikipedia entries are periodically sabotaged -- that is, rewritten -- by so-called “trolls” with ideological axes to grind.
A cottage industry has sprouted up around “search engine optimization,” more commonly known as boosting Google rankings. The trick: seeding a website with key terms that will show up in text hyperlinks, regardless of the site’s actual importance or relevance. And there’s always the fallback of enlisting actual humans to help click you up the list.
All this just compounds the endless, everyday flow of disinformation, evasion and falsehood abetted by one of the Internet’s defining features: users’ ability to disguise their identities. Thus, would-be Hollywood ingenues impersonating real-life teenagers (see: Lonelygirl15 and her ersatz video diary), and any number of creeps or bored kids posing as Brangelina look-alikes on Internet dating services.
Blame ourselves
BUT if the Web has failed to live up to some of the Arcadian hopes that launched it, perhaps that has more to do with our inflated expectations than the Web itself, says Bruce Bimber, a professor in political science and communication at UC Santa Barbara.
“So, big surprise, human nature reveals itself to be pretty much the same, even though the technologies change,” says Bimber. “People behave with sort of the same mixture of crass motives and integrity online as they do elsewhere.”
As Bimber, the author of books about technology and politics, points out, “One of the things that’s so compelling about the Web is you find all kinds of informational strategies there.” For better and worse, the Web commands our attention because it solicits our participation, even if not all its “informational strategies” fall within neat ethical lines -- which is equally true of the Web’s mass communication ancestors: newspapers, television and radio.
It’s a feature of the Web’s protean, interactive nature that duplicity and deception may be rewarded rather than reviled. When it was revealed that the witty YouTube poster Littleloca wasn’t actually a teenage East L.A. Latina homegirl but a 22-year-old, non-Latina, actress-wannabe from Victorville, the hate e-mails started rolling in. But so did interest from traditional entertainment/news media outlets (including Fox, which did a long segment on the send-up), who’ve been sniffing around the Internet’s new “stars” like hyenas circling a gazelle.
Surreptitious self-promotion, after all, is an old tool for grabbing at the American Dream, and in many circles, the attitude seems to be “no harm, no foul” -- with extra points for gonzo audacity and creativity. Digital libertarians argue that the Web always has been a Wild West where rules were made to be broken and imposing restrictions could stifle free expression.
Yet some of the more devious Web stunts and manipulations raise questions about the quality of the information the Web generates and the popular culture it gives rise to. Rather than the all-access Garden of Knowledge or the Virtual Democracy envisioned by its early champions, the Web is a contested space roiled by competing claims and the pressure to stake out valuable turf. Especially as the Internet economy explodes.
When Alinta Thornton, a senior consultant on Web design and intranet applications at the Hiser Group in Sydney, Australia, wrote and posted online her master’s thesis about the Web’s creeping commodification in 1996, she says, “People pooh-poohed me and said, ‘Don’t be ridiculous.’ ” The prevailing idea then was that the Web would “be a public space free of interference, both from government control and commercialism” and wedded to “a larger narrative of progress,” as Thornton wrote.
But like past technologies, the Web so far has proven to be no more innately pro-democracy (as opposed to merely populist) than cellphones or BlackBerrys.
“It doesn’t matter what the technology is, whether it’s the Internet or a horse-drawn carriage or carrier pigeons,” Thornton says. “People will still lie and cheat. They misrepresent themselves to mates, they’ll try to get as much sex as they can. And on the positive side, they’ll try to collaborate with each other and help each other as much as they can.”
Beyond the hit parade
IT’S that last element that leaves researchers hopeful that Web searching can move beyond sometimes bogus “popularity” to genuine originality, authority and expertise -- that instead of conjuring up “The Interview With God,” our future Web searches might possess a collective cognitive power that’s truly divine.
Perhaps the most significant byproduct of the Web’s obsession with “most this” and “most that” is the emergence of a new mass culture assembled from thousands of fragmented, narrow-cast audiences. A major challenge will be figuring out how to make that culture reflect not just the quantity but the quality of what’s available via the Internet.
Right now, the Web is akin to a global, digital Miss Universe pageant in which the “most linked” and “most e-mailed” sites get the tiara, even if they won partly by sabotaging their rivals backstage. How, some researchers are asking, do you help Web surfers separate the Web’s rare gold from the mass dross without imposing restrictive filters or replicating the superficial ranking mentality of the old mass culture?
Oren Etzioni, who specializes in the future of Web searching, believes that one solution is developing new models for “unsupervised information-extraction systems” that would offer search results based on a more sophisticated, composite analysis of the content of new products and services, rather than looking simply at how popular they seem to be.
“I think the fundamental motivation for a lot of the work we do is information overload anxiety. Everybody’s feeling it these days,” says Etzioni, a professor of computer science at the University of Washington. “It’s very natural when you’re overwhelmed with all this stuff to start backing away and to say, ‘Hey, just give me the number, just give me my favorite source.’ ”
One such system called Opine, developed as part of a larger initiative between Google and the University of Washington, will take advantage of ever-more-powerful computers to sift through dozens of online reviews of, say, Los Angeles-area hotels. Then, it will develop a model of their key features, plus an overview of how these hotels have been evaluated by large numbers of bloggers, newsgroup posters and other Web users.
Etzioni, the project leader, says that Opine is smart enough to extract value judgments not just from key words but also from complicated syntax (“This hotel is ridiculously expensive”), and to organize search results to give users a more nuanced, less reductive picture of what they’re looking for, accurately, authoritatively and concisely.
“The search-spam is an arms race, so it’s not like Opine is an antidote to that. The arms race is going to continue. It’s a fact of life,” Etzioni says. But he believes that programs such as Opine will save Web surfers hours of time while also broadening their search horizons because it “is less about what site is at the top than about synthesizing information from hundreds of sites.”
Given the enormous size and scope of its operations, the Web has some rankings that naturally are going to be more logical and useful than others. Google, Yahoo and other search providers are constantly tinkering with and refining their engines and algorithms, making them more sophisticated and scam-proof. The Web as a whole is busily developing more safeguards and hashing out the various legal and ethical quandaries facing it.
“We’re in the early days of a whole new institution, and there are going to be loopholes,” says Thornton, who compares the Web’s current vulnerabilities to those of banks before safeguards such as alarm systems and surveillance cameras were developed. At the same time, she says she finds it “reassuring” that the broadly reliable Wikipedia so often turns up in the top five results for virtually any Google search, given that studies have shown most people rarely look past the first “page” of results.
Thornton is more concerned with subtler kinds of cultural biases that tend to skew Web search results, such as the continuing dominance of the United States in generating Web content and the related, disproportionate dominance of the English language. (According to some estimates, about 40% of all Web content worldwide comes from the U.S.)
She also worries about the complicity of search providers in allowing foreign governments to censor results in their home countries, whether it’s the Germans barring Nazi-related material or the Chinese version of Google showing only flowery tourist images of Tiananmen Square. In the long run, she suggests, these issues will have far more bearing on world culture and freedom than whether Littleloca or one of her v-blogging rivals is pumping up her YouTube tallies.
What’s needed, says Bimber of UC Santa Barbara, isn’t so much a new way of policing the Web to verify its contents and make sure no one is “cheating,” but “a new kind of literacy and a new kind of relation to authority.” As information consumers, he says, we’ve been conditioned for generations to defer to putative authorities, arbiters and referees such as newspaper editors and the Federal Communications Commission.
But the great gift of the Web has been to turn millions of us from being primarily consumers into content providers and authors. While he thinks “it’s a bit optimistic” to imagine the Web ever could be fully “self-policing,” Bimber believes the medium could develop “pretty good self-governance,” if not “definitive or authoritative self-governance.”
As for the manipulation of search results or search-generated information, Bimber thinks “the best response is the outing of the practice” by bloggers and other Web users.
“Norms change slowly, and they typically require more than one generation,” Bimber says. “I think we’re a little bit impatient. Probably none of the people reading your article will be alive when these things are ironed out.”
Etzioni, similarly, says that he’s “a big believer in the idea that ‘We’re minutes away from the big bang -- you ain’t seen nothing yet.’ ”
“It’s very early in the evolution of these tools,” he says. “In 10 years -- which is a huge time in ‘Internet time,’ as they say -- we’re going to have radically better tools. Which is good, because there are going to be radically more effective ways to manipulate us.”
Next Sunday: How people in the industry manage to keep up or fail to keep up with the explosion of pop material.
