Over the past couple weeks Facebook has been rolling out an “upgraded” version of user profiles which connects their interests to Facebook Pages. While the concept is an interesting one, the current implementation has been frustrating a lot of users. For one, when you go through the process of shifting to the new design and associating your interests with Pages, many of the new “community Pages” are not properly synced. There are a number of other issues as well.
Ineffectively Structured Data
While a lot of the interests in users’ profiles are correctly associated with popular Facebook Pages, many of the interests and/or jobs are not properly associated. For example, I state that I’m working at “Social Times Inc.” which was a network within Facebook. Now Facebook has shifted my association to a randomly generated community Page, not our official Social Times page.
The result is that there is now more than one Page for “Social Times”, a problem that William Beutler effectively pointed out on our site earlier this week. Now that there are easily hundreds of thousands misattributed Pages, Facebook needs to come up with a solution.
Connected Profiles Create Privacy Issues
One of the other issues that Sarah Perez of ReadWriteWeb effectively articulated is that once your interests become connected with Facebook Pages, that data immediately becomes public. It’s a decision which is becoming increasingly controversial in the press, however most users don’t appear to care about the shift on a privacy level.
Despite users not caring (as many of them are not aware of the privacy implications of this new change) it doesn’t mean that it’s a justifiable switch. Then again, the concept of a connected social graph based on user interests is a pretty powerful one. At the least, Facebook could enable users to edit the associations to various pages. Given that there are now 5 pages for each interest, I’m sure users could effectively curate those associations rather than Facebook forcing it.
Are you a fan of the “upgraded” profiles or not?