A tracker is a script included as part of a webpage's content, often utilizing some combination of tracking pixels (an http request for a 1x1 image file from the tracking script's providers domain), a persistent cookie, and increasingly some form of browser/device fingerprint which is used to identify a particular machine.
The user's machine presents back to the tracking network the cookie and a bunch of http params to the tracking provider whilst interacting with pages that support the script, which the tracker stores in a database to sell access to.
It gives developers/businesses a way to collect metrics while offloading the trouble of keeping track of and maintaining the infrastructure to do so to someone else.
Firefox will probably be enforcing a cross-origin isolation constraint, requiring that all material be hosted by the domain you're requesting from in the first place, which doesn't really fix the problem since people will probably just try to build ways around the limitation.
Until the industry breaks itself free of it's current fetish for wholesale data collection, it's just going to be an arms race.
"Disconnect Private Browsing automatically detects when your browser tries to make a connection to anything other than the site you are visiting. We call these other attempted connections “network requests" https://disconnect.me/help
Moving everything to first-party tracking could potentially solve a lot of the problems.
It could reduce the across-the-entire-web tracking, categorizing and labeling of users, and instead limit analytics to useful things like "what percentage of my users are on mobile".
Whatever they feel like it being. Including images.
For some reason, this keeps getting flipped to "on" for me, and I have to keep turning it off, to get images to load correctly in both my RSS reader and via a convenience user script.
The user's machine presents back to the tracking network the cookie and a bunch of http params to the tracking provider whilst interacting with pages that support the script, which the tracker stores in a database to sell access to.
It gives developers/businesses a way to collect metrics while offloading the trouble of keeping track of and maintaining the infrastructure to do so to someone else.
Firefox will probably be enforcing a cross-origin isolation constraint, requiring that all material be hosted by the domain you're requesting from in the first place, which doesn't really fix the problem since people will probably just try to build ways around the limitation.
Until the industry breaks itself free of it's current fetish for wholesale data collection, it's just going to be an arms race.