- 注册
- 2002-10-07
- 消息
- 402,179
- 荣誉分数
- 76
- 声望点数
- 0
What data do social media companies such as Facebook, Twitter and Google know about you and why does it matter?
It’s a question that’s become more relevant since the revelation that British consulting firm Cambridge Analytica may have mined the information available through social media channels to sway the outcome of the U.S. election in 2016.
The scandal and its fallout are shining a bright spotlight on the data that social media firms have been collecting.
And that should scare people into demanding better privacy protection laws immediately, according to a former Ontario Information Privacy Commissioner.
“We should be totally concerned about companies and their unauthorized use of your personal data,” said Anne Cavoukian, a former privacy commissioner for the province of Ontario and currently the head of the Privacy by Design Centre of Excellence at Ryerson University in Toronto.
“The way it is now, if you want to preserve your privacy, you have to go through all the Terms of Service and all their legalese and the privacy policies to find the opt-out box that says ‘Do not use my information for any other purpose other than the primary purpose intended.’ We know nobody does that. But, that doesn’t mean people don’t care about privacy.”
Social Network applications including Facebook, Instagram, Slack, Snapchat, Twitter, Skype, Viber , Teamsnap and Messenger, are on display on a smartphone.
To illustrate just how much personal information has been collected on average Canadians by the three big social media networks, the Citizen used publicly available online tools to download the data records of a senior editor. The Citizen’s deputy editor, Keith Bonnell, volunteered for the experiment and proceeded to collect his personal data file from the Big 3: Facebook, Google and Twitter.
The information he downloaded is easily accessible for anyone who wants to replicate the exercise with their own data. The companies offer any user the chance to download their data footprint. A quick Google search shows how to do it for each platform.
Combined, Bonnell’s downloaded data totalled more than 1.66 Gigabytes of data. For comparison’s sake, a digital copy of the King James Bible is only 13.48 megabytes in size — more than 100 times smaller than the combined size of his data.
The files contained a massive hodge-podge of data, including a decade’s worth of Facebook status updates, messages, photos, tweets and an extremely comprehensive catalogue of everything Bonnell has used Google to search for online.
Every event he has indicated an interest in attending — concerts, parties, art shows — dating back to 2007, was listed.
Facebook also had a detailed database of every advertisement he had ever clicked on and when.There’s a long list of advertising related search terms and topics that he had expressed interest in over the years — from Journalists for Human Rights, to Glenfiddich whisky, to tattoos, to airlines, to Star Trek.
The file contains a list of every message Bonnell has ever sent using Facebook Messenger, including every picture he has ever sent or received. A list of “friends” named every person Bonnell has befriended on Facebook, including the ones he is no longer friends with and when he “defriended” them.
A woman poses in front of a computer displaying the Facebook logo.
Bonnell’s list of personal contacts was empty seeing as he had, at some point, chosen to bar Facebook from accessing to his contact lists on other platforms. Other users who have downloaded their private Facebook information have found hundreds of names, email addresses and phone numbers (including cellphone contact numbers) for people who aren’t on Facebook. The information in Facebook’s files appears to be taken directly from a user’s cellphone contact information.
Google offered similarly in-depth knowledge of Bonnell’s personal life, including a list of every address he had searched using the service’s map application over the past decade. There were searches for directions and searches for apartments at times when he’d been moving. Google also collected information on every advertisement that Bonnell has clicked, search terms he has typed into Google, his preferred sources of news and articles and a detailed index of appointments and meetings he has attended, after accepting invitations to the meetings through email, going back to 2007.
Twitter proved to be least intrusive of the three. The company did provide a detailed file which included every Tweet that Bonnell has ever made, as well as a list of Tweets he has liked and Retweeted. It also sent a lengthy list of the people who follow him and which people he follows.
“It’s not blackmail material — nothing horrible, nothing awful — but it’s definitely stalker material,” Bonnell said. “It’s like that obsessive-compulsive diary I never took the time to keep and never really wanted to exist.”
MORE: Roy: The Facebook affair shows democracy in peril
The amount of data available from the big three social media companies came as no surprise to David Nickerson, an associate professor in the department of political science at Temple University in Philadelphia. Nickerson was a co-author on a 28-page research paper released in 2013 that examined the impacts that big data could have on an election campaign.
He said he doesn’t think the impact that Cambridge Analytica had on the U.S. federal election in 2016 is as big as some people are alleging. Cambridge Analytica is alleged to have released personality quizzes, games or other applications that mined personal data from Facebook users and their contacts or friends.
The entrance of the building which houses the offices of Cambridge Analytica.
It’s believed the information harvested was used to create “psychographic profiles” of users, which is essentially a snapshot of each users’ interests, lifestyle and political leanings. The profiles could be used by marketers or political consultants to better target an intended demographic.
Nickerson said while the data that the company collected may have been personal in nature, it wouldn’t offer much more insight into a voter’s intentions than what would traditionally be available through political polls, which ask people what they plan to do come election day. However, where the company could have used Facebook to influence voters is by using the quizzes or apps to mask partisan ads and information.
To use the quizzes and apps, Facebook users had to give permission to the company to “push” notifications to them, which gave Cambridge Analytica a direct pipeline to the person. Opting to allow the app to send notifications, allows that application to send news articles, surveys or other information right to a person’s Facebook news feed, making it difficult to ignore. People are more likely to look critically at information that comes directly from a political party. However, if information appears to come from a news website and appears in their newsfeed, it could be perceived as carrying more weight.
“The thing about a game app or a quiz app is that these are not explicitly partisan. So, it’s possible it’d become a trusted source. If you have partisan news, you wouldn’t recognize it as partisan news,” he said. “If I hear something that is partisan, I can label it as ‘the Republicans are saying this.’ Or, ‘the Democrats are saying this,’ and I down-weight it. As opposed to, me saying ‘Oh, this newspaper, which I have no idea whether it is partisan or not.’ I’m more likely to view it as honest fact and not spin.”
That message gets amplified when the user passes the information onto their network of friends. Friends may be more likely to believe information shared by people they trust.
But, Nickerson said, the real targeting is done by the social media giants themselves, with Facebook and Google chief among them. Using Bonnell’s data as an example, the companies have collected so much personal information that they know what content or advertisements a person is likely to click on, even before that person is presented with the ad.
“Facebook has way more data on people than Cambridge Analytica does and they’ve got an entire team of data scientists that have been trying to tweak their algorithm to get maximum engagement with the ads,” he said. “My understanding of it is, they’ve created very broad profiles of people. They’ve built up so much data to access. When they roll out an ad they say, ‘OK, we have John that clicked on, and Mary did not.’ Then they say, as an ad rolls out, ‘We want more people that look like John’. It’s all based on the behaviour of people.”
A Facebook employee walks past a sign at Facebook headquarters in Menlo Park, California.
In essence, Facebook can use the reams of information it has about each of its users to target specific ads based on needs and desires. Fans of sports teams could find they are inundated with ads for sports jerseys. Fans of a specific TV show may find their feed is full of places to buy Blu-ray box sets. Real danger could occur should Facebook choose to limit political messaging or news based on a person’s preferences. It could limit a person’s exposure to opposing views which could affect the way they see a political debate.
Facebook has toyed with manipulating its users in the past. The company came under fire in 2012 after it was discovered that Facebook had partnered with data scientists to manipulate the news feed of 689,003 Facebook users in order to see if it could impact the content that those users then posted to Facebook. The study, which lasted a week and was published in the journal the Proceedings of the National Academy of Sciences, found that negative status messages led to more negative status messages throughout a person’s network. Whereas networks that shared more positive status messages, bred more positivity. Researchers conducted their study by reducing the positive or negative responses that could be seen by the allegedly unwitting participants.
The depth of information collected on individuals and the way that information has been shared with third party firms such as Cambridge Analytica is kickstarting the debate in countries all over the world about how to best update privacy laws, Cavoukian said.
The former privacy commissioner of Ontario said if there is any silver lining to take from this, it’s the fact that regulators all over the world are seeing red, are demanding answers and many are now releasing new legislation aimed at culling the data that companies are allowed to collect and better policing how its shared online.
“Facebook is in such trouble. Regulators in all the major countries, in the U.K., in the E.U., in the United States, the Privacy Commissioner of Canada is investigating, the Federal Trade Commission in the U.S. has called them to appear before them,” she said. “Because of this privacy debacle from Facebook and Cambridge Analytica, Facebook has turned a page. They’re scared. They’ve changed their tune overnight.”
Earlier this week Facebook released new privacy tools, which it says are aimed at helping users better police the information they share and how it is used.
ALSO: Searching for news on Google can return victim and offender names under strict pub ban
The company is under the microscope, particularly in the U.S. In 2011, the company signed an agreement, known as a “Consent Decree,” which specifically stated how the company was to protect the personal information of people using its services. If the company is found to be in breach of that agreement, it could be fined $40,000 US for each American user of its services.
“It’s beyond billions (of dollars),” said Cavoukian. “The implications of that, off the scale. Huge.”
Facebook is having one of its worst weeks as a publicly traded company.
Facebook is also facing a class action lawsuit in a California court from users of its “Messenger” app over the data that application had been collecting.
She also said, European regulators are preparing to roll out new privacy laws in the European Union aimed at better protecting the data of its residents. The roll out of the stringent new European laws, coupled with Facebook’s recent issues could see Canada revisit its privacy laws in the very near future.
“We have to upgrade it. We want our laws to be considered to be equivalent to the new (European law). In the past, our laws were considered to be essentially equivalent to EU law. Now, we no longer have essential equivalence,” she said. “More people care about privacy than ever before. I’ve never seen Internet survey results in the 90 percentile consistently and trust is at an all time low. When you compare those factors its in everyone’s interest to upgrade our privacy laws.”
查看原文...
It’s a question that’s become more relevant since the revelation that British consulting firm Cambridge Analytica may have mined the information available through social media channels to sway the outcome of the U.S. election in 2016.
The scandal and its fallout are shining a bright spotlight on the data that social media firms have been collecting.
And that should scare people into demanding better privacy protection laws immediately, according to a former Ontario Information Privacy Commissioner.
“We should be totally concerned about companies and their unauthorized use of your personal data,” said Anne Cavoukian, a former privacy commissioner for the province of Ontario and currently the head of the Privacy by Design Centre of Excellence at Ryerson University in Toronto.
“The way it is now, if you want to preserve your privacy, you have to go through all the Terms of Service and all their legalese and the privacy policies to find the opt-out box that says ‘Do not use my information for any other purpose other than the primary purpose intended.’ We know nobody does that. But, that doesn’t mean people don’t care about privacy.”
Social Network applications including Facebook, Instagram, Slack, Snapchat, Twitter, Skype, Viber , Teamsnap and Messenger, are on display on a smartphone.
To illustrate just how much personal information has been collected on average Canadians by the three big social media networks, the Citizen used publicly available online tools to download the data records of a senior editor. The Citizen’s deputy editor, Keith Bonnell, volunteered for the experiment and proceeded to collect his personal data file from the Big 3: Facebook, Google and Twitter.
The information he downloaded is easily accessible for anyone who wants to replicate the exercise with their own data. The companies offer any user the chance to download their data footprint. A quick Google search shows how to do it for each platform.
Combined, Bonnell’s downloaded data totalled more than 1.66 Gigabytes of data. For comparison’s sake, a digital copy of the King James Bible is only 13.48 megabytes in size — more than 100 times smaller than the combined size of his data.
The files contained a massive hodge-podge of data, including a decade’s worth of Facebook status updates, messages, photos, tweets and an extremely comprehensive catalogue of everything Bonnell has used Google to search for online.
Every event he has indicated an interest in attending — concerts, parties, art shows — dating back to 2007, was listed.
Facebook also had a detailed database of every advertisement he had ever clicked on and when.There’s a long list of advertising related search terms and topics that he had expressed interest in over the years — from Journalists for Human Rights, to Glenfiddich whisky, to tattoos, to airlines, to Star Trek.
The file contains a list of every message Bonnell has ever sent using Facebook Messenger, including every picture he has ever sent or received. A list of “friends” named every person Bonnell has befriended on Facebook, including the ones he is no longer friends with and when he “defriended” them.
A woman poses in front of a computer displaying the Facebook logo.
Bonnell’s list of personal contacts was empty seeing as he had, at some point, chosen to bar Facebook from accessing to his contact lists on other platforms. Other users who have downloaded their private Facebook information have found hundreds of names, email addresses and phone numbers (including cellphone contact numbers) for people who aren’t on Facebook. The information in Facebook’s files appears to be taken directly from a user’s cellphone contact information.
Google offered similarly in-depth knowledge of Bonnell’s personal life, including a list of every address he had searched using the service’s map application over the past decade. There were searches for directions and searches for apartments at times when he’d been moving. Google also collected information on every advertisement that Bonnell has clicked, search terms he has typed into Google, his preferred sources of news and articles and a detailed index of appointments and meetings he has attended, after accepting invitations to the meetings through email, going back to 2007.
Twitter proved to be least intrusive of the three. The company did provide a detailed file which included every Tweet that Bonnell has ever made, as well as a list of Tweets he has liked and Retweeted. It also sent a lengthy list of the people who follow him and which people he follows.
“It’s not blackmail material — nothing horrible, nothing awful — but it’s definitely stalker material,” Bonnell said. “It’s like that obsessive-compulsive diary I never took the time to keep and never really wanted to exist.”
MORE: Roy: The Facebook affair shows democracy in peril
The amount of data available from the big three social media companies came as no surprise to David Nickerson, an associate professor in the department of political science at Temple University in Philadelphia. Nickerson was a co-author on a 28-page research paper released in 2013 that examined the impacts that big data could have on an election campaign.
He said he doesn’t think the impact that Cambridge Analytica had on the U.S. federal election in 2016 is as big as some people are alleging. Cambridge Analytica is alleged to have released personality quizzes, games or other applications that mined personal data from Facebook users and their contacts or friends.
The entrance of the building which houses the offices of Cambridge Analytica.
It’s believed the information harvested was used to create “psychographic profiles” of users, which is essentially a snapshot of each users’ interests, lifestyle and political leanings. The profiles could be used by marketers or political consultants to better target an intended demographic.
Nickerson said while the data that the company collected may have been personal in nature, it wouldn’t offer much more insight into a voter’s intentions than what would traditionally be available through political polls, which ask people what they plan to do come election day. However, where the company could have used Facebook to influence voters is by using the quizzes or apps to mask partisan ads and information.
To use the quizzes and apps, Facebook users had to give permission to the company to “push” notifications to them, which gave Cambridge Analytica a direct pipeline to the person. Opting to allow the app to send notifications, allows that application to send news articles, surveys or other information right to a person’s Facebook news feed, making it difficult to ignore. People are more likely to look critically at information that comes directly from a political party. However, if information appears to come from a news website and appears in their newsfeed, it could be perceived as carrying more weight.
“The thing about a game app or a quiz app is that these are not explicitly partisan. So, it’s possible it’d become a trusted source. If you have partisan news, you wouldn’t recognize it as partisan news,” he said. “If I hear something that is partisan, I can label it as ‘the Republicans are saying this.’ Or, ‘the Democrats are saying this,’ and I down-weight it. As opposed to, me saying ‘Oh, this newspaper, which I have no idea whether it is partisan or not.’ I’m more likely to view it as honest fact and not spin.”
That message gets amplified when the user passes the information onto their network of friends. Friends may be more likely to believe information shared by people they trust.
But, Nickerson said, the real targeting is done by the social media giants themselves, with Facebook and Google chief among them. Using Bonnell’s data as an example, the companies have collected so much personal information that they know what content or advertisements a person is likely to click on, even before that person is presented with the ad.
“Facebook has way more data on people than Cambridge Analytica does and they’ve got an entire team of data scientists that have been trying to tweak their algorithm to get maximum engagement with the ads,” he said. “My understanding of it is, they’ve created very broad profiles of people. They’ve built up so much data to access. When they roll out an ad they say, ‘OK, we have John that clicked on, and Mary did not.’ Then they say, as an ad rolls out, ‘We want more people that look like John’. It’s all based on the behaviour of people.”
A Facebook employee walks past a sign at Facebook headquarters in Menlo Park, California.
In essence, Facebook can use the reams of information it has about each of its users to target specific ads based on needs and desires. Fans of sports teams could find they are inundated with ads for sports jerseys. Fans of a specific TV show may find their feed is full of places to buy Blu-ray box sets. Real danger could occur should Facebook choose to limit political messaging or news based on a person’s preferences. It could limit a person’s exposure to opposing views which could affect the way they see a political debate.
Facebook has toyed with manipulating its users in the past. The company came under fire in 2012 after it was discovered that Facebook had partnered with data scientists to manipulate the news feed of 689,003 Facebook users in order to see if it could impact the content that those users then posted to Facebook. The study, which lasted a week and was published in the journal the Proceedings of the National Academy of Sciences, found that negative status messages led to more negative status messages throughout a person’s network. Whereas networks that shared more positive status messages, bred more positivity. Researchers conducted their study by reducing the positive or negative responses that could be seen by the allegedly unwitting participants.
The depth of information collected on individuals and the way that information has been shared with third party firms such as Cambridge Analytica is kickstarting the debate in countries all over the world about how to best update privacy laws, Cavoukian said.
The former privacy commissioner of Ontario said if there is any silver lining to take from this, it’s the fact that regulators all over the world are seeing red, are demanding answers and many are now releasing new legislation aimed at culling the data that companies are allowed to collect and better policing how its shared online.
“Facebook is in such trouble. Regulators in all the major countries, in the U.K., in the E.U., in the United States, the Privacy Commissioner of Canada is investigating, the Federal Trade Commission in the U.S. has called them to appear before them,” she said. “Because of this privacy debacle from Facebook and Cambridge Analytica, Facebook has turned a page. They’re scared. They’ve changed their tune overnight.”
Earlier this week Facebook released new privacy tools, which it says are aimed at helping users better police the information they share and how it is used.
ALSO: Searching for news on Google can return victim and offender names under strict pub ban
The company is under the microscope, particularly in the U.S. In 2011, the company signed an agreement, known as a “Consent Decree,” which specifically stated how the company was to protect the personal information of people using its services. If the company is found to be in breach of that agreement, it could be fined $40,000 US for each American user of its services.
“It’s beyond billions (of dollars),” said Cavoukian. “The implications of that, off the scale. Huge.”
Facebook is having one of its worst weeks as a publicly traded company.
Facebook is also facing a class action lawsuit in a California court from users of its “Messenger” app over the data that application had been collecting.
She also said, European regulators are preparing to roll out new privacy laws in the European Union aimed at better protecting the data of its residents. The roll out of the stringent new European laws, coupled with Facebook’s recent issues could see Canada revisit its privacy laws in the very near future.
“We have to upgrade it. We want our laws to be considered to be equivalent to the new (European law). In the past, our laws were considered to be essentially equivalent to EU law. Now, we no longer have essential equivalence,” she said. “More people care about privacy than ever before. I’ve never seen Internet survey results in the 90 percentile consistently and trust is at an all time low. When you compare those factors its in everyone’s interest to upgrade our privacy laws.”
查看原文...
