Aren’t the Like buttons distributed around the Web and used on purpose by publishers? If your reading the New York Times, there’s a like button on the story, if you click it, aren’t you explicitly telling Facebook your interested?
Perhaps you object to the page impression being tracked even if you don’t like it? This seems unavoidable unless you mandate publishers host their own copy of the code behind these buttons and use a public API only to record the likes.
Facebook was tracking users even if they didn't click the Like button. This has nothing to do with page view tracking.
And they were tracking even non-Facebook users this way. There's zero consent here no matter how you cut it.
Worse yet, Facebook did this for years, and in the first years it swore it doesn't do that, and when they were caught doing it they said it was "just a bug"...twice.
Facebook has 0% credibility at this point. Just look how they're ruining people's trust in 2FA security, too, now (although if they end-up ruining the trust in SMS 2FA, that wouldn't be too bad of an outcome actually...).
This is a good article (although not complete, and only until 2014) on Facebook's awful, no-good, history of lying and tracking users without their consent:
There's zero consent here no matter how you cut it.
Thats not true. You implicity consent to usage policy of that website which then gave permission to FB. Ofcourse GDPR makes it illegel but until May 25 its not.
As far as I know, Like buttons are partially served by Facebook, which then knows I've accessed the site (the NYT in your example). This happens even if I do not click the button, do not hover over it, and do not have a Facebook account.
In the case of this the theguardian.com page it looks like they've got a webbug as the last element of the body, which results in a request to https://www.facebook.com/tr?
Having facebook count views of your like button is probably considered a feature to publishers.
> This seems unavoidable unless you mandate publishers host their own copy of the code behind these buttons and use a public API only to record the likes.
If Facebook means no harm, they could make this the encouraged way to include the like button.
> This seems unavoidable unless you mandate publishers host their own copy of the code behind these buttons and use a public API only to record the likes.
Yes, that's the consequence (well, it does't really have to be an API triggered from the server, it can send the user there). Or a pattern where a user has to click a locally hosted link/button which then loads the Facebook one, and/or records permission to load it with further page visits.
It's not really that much difference from an integration perspective if you include code in your page that loads a facebook iframe, or code that only does that after user interaction.
Perhaps you object to the page impression being tracked even if you don’t like it? This seems unavoidable unless you mandate publishers host their own copy of the code behind these buttons and use a public API only to record the likes.