In February, it came to light that through the “Find Friends” function on its mobile app, Twitter is able to download the full address books of mobile users, including phone numbers and emails. The iPhone app does not alert users that their information is being downloaded, but rather says they can “scan” contacts for people they already know.
“The main concern is over users’ ability to control the disclosure and use of their own personal information,” David Jacobs, a consumer privacy fellow at the Electronic Privacy Information Center, said. “Any time that a company takes that away or lessens that control you’re going to have a privacy issue.”
Danielle Citron, a law professor at the University of Maryland, said this leaves Twitter users with few options other than to “take it or leave it.”
Without technological skills to prevent information downloading, “I think you vote with your feet and leave,” Citron said. “And that’s really the tough thing about so much of the world, take it or leave it. You don’t like our policy stance, too bad, so sad. You have to go somewhere else.”
According to Frank Pasquale, a law professor at Seton Hall University, the downloading of phone numbers and email addresses is not a major concern because this information is increasingly available on the Internet. For Pasquale, the bigger concern is the direction data collection may be going.
“It could be that in the future you might see people that are trying to construct reputations for individuals based on their social graph or the people they know if those people have opted in,” Pasquale said. He added that health profiles or consumer profiles could be created for users simply by accessing their social media data.
Recently, Twitter took a step in this direction in a deal with DataSift Inc. that could alter the way social media data is analyzed. The deal, according to company’s Chief Marketing Officer Tim Barker, makes DataSift one of Twitter’s two “re-syndication” partners. DataSift has access to tweets from up to two years ago and can sell the rights to this data to companies seeking to analyze it for varying purposes.
“When tweets go into DataSifts service, we enrich the information that’s held about those,” Barker said. “One good example of that is sentiment. We have algorithms that will look at what the sentiment of tweets is – is it positive, is it negative, is it neutral?”
According to Barker, customers interested in using Twitter data range from news organizations to financial service companies or even education establishments.
Though there has been some concern about the privacy implications of accessing tweets from several years in the past, Barker said DataSift only accesses information already publicly available. The fact that tweets are public lessens the privacy concerns of this deal, according to Jacobs. Yet, he added that there is the lingering issue of “privacy through obscurity.”
“Even if something might be publicly available, the fact that it’s practically inaccessible because you really can’t go back and find what someone posted two years ago functions as sort of a privacy protection…” he said. “It’s one thing for [tweets] to sort of trickle out over the course of two years and … followers only see them one at a time and it’s another thing to have them available all at once and open to be mined and analyzed.”
Barker, however, stressed that though individual tweets are available, they are not the focus of companies seeking data on Twitter. Instead, they are focused on major trends in the data.
“Brands want to understand how people talk about their companies. Financial service firms want to understand what the sentiment of conversation is around stocks,” Barker said. “It’s those kinds of things that our customers are focused on, not individuals.”
Despite the fact that the data mainly focuses on trends, Citron said analyzing people as part of a larger trend could create a very specific “mosaic” of information on individuals. Citron said she believes privacy in public can be beneficial but it is not part of the current culture.
“There’s no ability to be obscure and be forgotten anymore,” she said. “Is there any law to protect us? No … Should we have privacy in public? I think so. I think there are a lot of important things about privacy in public. There are a lot of other values at stake, not just obviously commerce, and the ability for Twitter to sell it. But is that the prevailing rule now? No.”
According to a recent Pew Research Center study, a higher percentage social media users deleted others from their networks in 2011 than in 2009 while a lower percentage of people post material they later regret. Jacobs said he thinks this research may point to more awareness of privacy issues concerning social media.
“I think that users are increasingly taking steps to try and protect their privacy and to restrict the availability of information that they think they’re only sharing with a select group of people,” he said.
Because social media continue to expand at a rapid pace, Citron said she does not believe comprehensive privacy legislation will be possible in the near future. And as tools such as Twitter, Facebook and Google continue to grow, Pasquale said the concern has to be about finding ways to protect individuals.
“Those are utilities that you can’t get by in modern life without them … Good luck trying to be a kid without a Facebook,” Pasquale said. “It’s got to be about reputation in the future. It’s got to be about protecting the integrity of reputation.”