World’s Largest Social Network: The Open Web
Guest Post by Randall Stross (Originally published in the NYT May14, 2010.
ON its Web site, Facebook says it’s “giving people the power to share and make the world more open and connected.”
But the online world outside of Facebook is already a very open and connected place, thank you very much. Densely interlinked Web pages, blogs, news articles and Tweets are all visible to anyone and everyone. Instead of contributing to this interconnected, open Web world, the growing popularity of Facebook is draining it of attention, energy and posts that are in public view.
Every link found on the open Web, inviting a user to click and go somewhere else, is in essence a recommendation from the person who authored the page, posted it or broadcast it in a Tweet. It says, “I’ve taken the trouble to insert this link because I believe it will be worth your while to take a look.”
These recommendations are visible to search engines, which do far more than just tally how many recommendations point to this or that item. The engines trace backward to who linked to the recommender, then who linked to the recommender of the recommender, and so on. It’s a lot of computation to derive educated guesses about which recommendations are likely to lead to the best-informed sources of information and then placed at the top of a search results page.
No “friending” is needed to gain access; no company is in sole possession of the interconnections.
The size of the open Web — built without Facebook’s help — is hard to appreciate. In 2008, Google announced that its search engine had “crawled,” that is, collected and indexed material from, one trillion unique URLs, or Web addresses.
“The beauty of the Web is that it is open, and anyone can crawl it,” says Matt Cutts, a software engineer at Google.
But Facebook does not permit Google to reach most categories of information placed on the site, says Mr. Cutts, adding, “Google can only know what it can crawl.”
Susan Herring, professor of information science at Indiana University, sees it this way: “What the statistics point to is a rise in Facebook, a decline in blogging, and before that, a decline in personal Web pages. The trend is clear, she said — Facebook is displacing these other forms of online publication.
Barry Schnitt, a Facebook spokesman, said his company provides Google with access to public profiles of members and status updates for public Facebook pages, formerly called “fan pages.” He said it also has announced plans to work with Microsoft on its Bing search engine, allowing Bing to publish the status updates of individual members whose privacy settings permit display to “everyone.”
The Facebook model of organizing the world’s information involves a mix of personally sensitive information, impersonal information that is potentially widely useful, and information whose sensitivity and usefulness falls in between. It’s a tangle created by Facebook’s origins as the host of unambiguously nonpublic messaging among college students.
People often talk about the two leading social networking sites in a way that sounds like they’re a single entity: FacebookandTwitter. But the two are fundamentally different. Facebook began with a closed, friends-only model, and today has moved to a private-public hybrid, resetting members’ default privacy settings. By contrast, most Twitter users elect to use the service to address the general public.
Facebook has redefined the way its users go about obtaining information.
“Information is becoming less of a destination that we seek online,” says Anthony J. Rotolo, assistant professor of practice in the School of Information Studies at Syracuse University. “Instead we are expecting it to come to us in a social stream.”
In the Facebook stream, friends, not search engines, are the trusted sources.
“Just because someone belongs to your social network, it doesn’t make them a good source,” Professor Rotolo says. “But there’s a natural inclination to assume that a person possesses reliable information because it’s person-to-person.”
This is what Professor Herring calls a “recommender model” of getting information. And she sees it as replacing the search-engine model. She points to the recent introduction of the Facebook “Like” button at Web sites, which allows Facebook to note recommendations of those sites among one’s friends. The record of who clicks that “Like” button, however, is not part of the open Web; it’s Facebook’s. The public visibility of users’ Likes on Facebook depends on their privacy settings.
DEFENDERS of the Facebook information stream argue that it doesn’t displace the open Web, but that it merely adds a new layer of information to it. Yet there is a cost: more time spent dispensing recommendations among friends on Facebook means less for similar contributions elsewhere. Members now spend more than 500 billion minutes a month on Facebook, a company fact sheet says.
The links on the trillion Web addresses found by Google, and within the billions of Tweets that have followed, form an incomparably vast, truly worldwide, web of recommendations, supplied by fellow humans.
In this sense, the open Web has a strong claim to being more “social” than does Facebook.
Randall Stross is an author based in Silicon Valley and a professor of business at San Jose State University.