[mikepasek]

Ad blocking via DNS is relatively easy right now because a content provider like CNN.com will use a domain like “ads.evil-surveillance-media.com” to load their ads into your browser. But what happens if all these companies switch to just using their own domain to load ads? If the ads as well as the content BOTH come from CNN.com then there will be no easy way to filter the ads out. This will be the next stage in this war between ads and adblockers.

[uponcoffee]

With the prevelence of ad blocking tech, the question becomes why haven't they already?

The answer being that content providers can't be trusted to self report metrics that determine how much advertises pay. At least not for pay per view/client/etc models.

The people that self select themselves from viewing advertisements might be doing advertisers a favor. They're perhaps less likely to make purchases based on impressions//click ads on purpose; per dollar, ad campaigns might be more effective without said people.

[matoro]

This has been warned about for a while now and it's not actually terribly difficult for sites to implement. Why don't they? Simple - click fraud. The ad networks don't trust the sites to accurately report click numbers, so they insist on running their own code. For this reason, the number of truly first-party ads will be limited for a long while.

[ben509]

Adblocker plugins already do content-level filtering because of this, not just for ads but various tracking and other annoyances.

I do that for the sites that have the banner that hides if you scroll down, but pops out the moment you scroll one pixel back, or sites that put up "please don't leave me" modals the moment your cursor strays out of the window.

[nerdponx]

I'm already seeing this happen in some cases. Or they serve it from an opaque CDN alongside the functional site code.

[Lxr]

This. Router-level blocking will become impossible without intercepting https, and endpoint blocking also gets harder.