Biggest issue with this argument - advertising supported businesses are fine, contextual advertising is fine, targeted cross site advertising is a pointless red queen race that is undermining our society in multiple ways.

Can you say more what you mean by "pointless red queen race"?

Let's say someone wants to sell fishing equipment. The traditional way of doing this is to buy ads on fishing sites. So now my fishing equipment purchases make there be more writing about fishing; yay!

Then one of the fishing websites decides to put a tracking pixel on their site to drop "fishing website visitor" cookies (or, in a future without third-party cookies, a turtledove interest group). They make a deal with a third party provider and get paid a small amount per visitor. Then fishing retailers have a new choice: instead of buying ads on fishing sites they can instead buy ads on any site for users who have one of the "fishing website visitor" cookies. If there were a monopoly fishing site, then this would increase their earnings: while the ad space on their site isn't as valuable, they will set the pixel price high enough that they come out ahead. It's not a monopoly, though, so the price of the pixel gets driven down through competition, and money that would go to fishing sites instead goes to the publishers that people who spend money on fishing equipment visit.

In this case I see how it's worse for fishing sites, but not how it's bad for consumers: their willingness to buy fishing equipment translates into support for all the sites they visit, and not just the fishing sites.

But there are also many niches that don't have economic tie-ins, or have ones that are far weaker than "writing about fishing" and "buying fishing equipment". In a world with targeted advertising, these niches do better, because of overlap between audiences. A "let's have better housing policy" blog can show ads for fishing equipment, vacations, HVAC supplies, or whatever else visitors have shown interest in on other sites.

Additionally, targeted advertising increases the total amount of funding available for online content, because people with niche interests are available to be advertised to in more places. Seeing ten fishing ads once a week when you visit a fishing site vs seeing twenty fishing ads spread over the course of the week, etc.

So while niche publishers in lucrative niches would likely make more money if we only had context-based advertising, I don't think niche publishers overall, publishers overall, or consumers would be better off.

(This is modified from a comment I originally posted on https://news.ycombinator.com/item?id=21620763)

> n this case I see how it's worse for fishing sites, but not how it's bad for consumers: their willingness to buy fishing equipment translates into support for all the sites they visit, and not just the fishing sites.

It's bad for the consumer because their privacy is being violated and their metadata is being sold, in order for advertisers to track them everywhere they go online, so businesses can try and extract all of their spending money as efficiently as possible.

It also is bad for the consumer because it's bad for the collective whole: instead of quality content online everything is being driven by outrage and clickbait in order to serve as many targeted ads as possible.

Personally I don't even want to support that ecosystem or those sites.

It's also bad for the fishing site because now their niche targeted ad slots that used to pay decently in order to target people with an interest in fishing are pushed into the same race to the bottom low return ads that are being automatically targeted. So they lose too.

Only winners are huge publishers that don't have any niche audience to target because now they effectively target every niche. And Google of course. Basically the two groups who I want to win the least.

This is exactly what I'm working on changing; have a look at the second half of the post?

(Or read https://github.com/WICG/turtledove)