It depends on how it is implemented. Those articles are about invading the privacy of the user by detecting what software is on the machine. However, you don'

It depends on how it is implemented. Those articles are about invading the privacy of the user by detecting what software is on the machine. However, you don't need to detect that someone has an ad-blocker to circumvent it. First, you can simply disguise your ads. You can disguise them for everyone. That doesn't require you to peek into what software is on the machine. You've simply made it difficult for an ad-blocker to detect the ads. Second, you don't have to peek into the software on a machine to determine that content isn't loading. Let's say that someone is blocking my ad server via /etc/hosts. I shouldn't be able to write code that tries to read a user's /etc/hosts, but I can load content via AJAX, see that the content isn't loading, and then take action based on that (maybe loading from a secondary source, maybe not displaying the page if not all content can be loaded). I'm not saying that Facebook is circumventing ad-blocking in this way, but you certainly can attempt ad-block circumvention without reading a user's local configuration. Facebook can install the ad-blockers like anyone else on their own machines and see how they can disguise the ads as content so that the ad-blockers don't detect them.