A lot of people felt Facebook should do more to proactively stop people from scraping and aggregating data from their site after the CA debacle. Which is what they are doing here, and you are labeling “absolutely preposterous”.
Which way is it? Should they let people do whatever they want with their accounts as you suggest, and risk a repeat of the CA fiasco? Or try to proactively stop it like they are now?
> Which way is it? Should they let people do whatever they want with their accounts as you suggest, and risk a repeat of the CA fiasco? Or try to proactively stop it like they are now?
No, they should assume all the data they serve about people is being collected and indexed by all the people they serve it to, and then restrict what they serve accordingly. Suing people for asking their computers to record what Facebook served them is insane.
This looks like a strawman dressed in false dilemma. With Cambridge Analytica debacle, as I understand it, it was a facebook quiz hosted by facebook and the data collected from facebook directly and used for nefarious purposes.
This is a an extension developed by researchers asking users to install it on their machines and used exactly as advertised: scrapes advertisement data that facebook shows them.
CA data was from a quiz, developed by a researcher at Cambridge University, hosted by that researcher, which scraped FB APIs after being authorized by a user.
This data is from a browser extension, developed by researchers at NYU, hosted by those researchers, which scrapes the FB site after being installed by a user.
The situations are in no way analogous. CA gathered data on users who did not opt in. They paid 270k users to take a quiz, and gathered information on 87MM of their Facebook friends who never consented to sharing their information.
>A lot of people felt Facebook should do more to proactively stop people from scraping and aggregating data from their site after the CA debacle. Which is what they are doing here, and you are labeling “absolutely preposterous”.
Maybe not the best usage of the word scraping.
CA had access to the data without having to scrape the front end.
For this extension, it is by consent of the user
For scraping, I agree. Facebook should try their best to stop people accessing personal data of people they don't know. They do make reasonable attempts, it's less than trivial to set up fake accounts on scale but not impossible. Their "bulk uploads" feature is designed in such a way that it doesn't link email addresses to profiles (or at least as easily), unlike LinkedIn and Twitter. Saying that, it's up to the user to set their privacy settings but I would much prefer the defaults (if they still aren't, I don't use FB) were automatically set to non-public. I've seen an implementation that used headless browsers and thousands of FB accounts to scrape millions of profiles.
They harvested the data about users without their consent. This is about users wanting to protect against said data collection, so I think your angle comes from the wrong perspective and the juxtaposition (Which way is it?) is wrong.
> Should they let people do whatever they want with their accounts
That CA didn't have the consent of users was the scandal.
Which way is it? Should they let people do whatever they want with their accounts as you suggest, and risk a repeat of the CA fiasco? Or try to proactively stop it like they are now?