> Is it laziness that stops publishers from shipping targeted ads from their own servers, inlined in the content in a way that cannot be distinguished from arti

> Is it laziness that stops publishers from shipping targeted ads from their own servers, inlined in the content in a way that cannot be distinguished from article images or text? No, it's about fraud. With your proposal advertisers have to trust publishers really will show their ads, and unless the publisher is very large and has a reputation to protect advertisers pretty reasonably don't trust them. (We do actually see what you're proposing, with YouTube, Facebook, Google Search, etc) You could do this by having the ad networks serve the pages, reverse proxying to the publisher and inserting the ads. And then the publisher page can run its verification scripts, instead of the ads running verification scripts. But that gives us a web served by Google, Criteo, AppNexus, etc. which would be pretty sad.