@edsu It won't work, because while it might be able to momentarily bump up the amount the publishers get paid, it will reduce the value of the ads to the advertiser. And that assumes the fraud detection algorithms don't detect the impression and click as fraud. They operate after the fact, not in real time, so they have a tremendous amount of data available to them for detecting fraud. They don't even need to detect which specific clicks were fraudulent; they just have to estimate the number.
@freakazoid yeah, the project itself seems pretty misguided -- but using zoom to provide transparency into browser automation was pretty crazy to see
@edsu A very simple means of fraud detection that you can assume is being applied is to filter out anyone who clicks on a significantly larger fraction of ads than usual. The referer probably also won't match the way most people come in. And the tracking cookies will probably also be unusual. So many ways to detect fraudulent ad clicks, and all of them are used.
@freakazoid I mean, clearly, yes :-)
As I was trying to say in the last two posts, whether or not the automation is effective wasn't something that interested me very much...
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!