Google kills off FLoC, replaces it with Topics
techcrunch.com
techcrunch.com
Same snake with different color.
Yes and no.
Yes, it still amounts to building user-tracking functionality into user agents.
No, it’s not the same approach, because (IIUC) FLoC was doing unsupervised clustering of users, whereas Topics allocates users into predetermined clusters. This makes it potentially more transparent and also seems designed to work around the “protected categories” objection to FLoC, although in a world of proxy metrics and indicators I’m doubtful.
"For sites that it hasn’t categorized before, a lightweight machine learning algorithm in the browser will take over and provide an estimated topic based on the name of the domain."
yeah right
So it is better. I just think that it will not achieve the endgoal because every user will just deactivate it and the ad business will find another way to track me.
Most mainstream people don't care.
In addition to the business model question about whether enough people would actually pay directly to make it worth building and participating in the system, there are also serious challenges around how to prevent blockchain privacy issues from turning it into a great tracking system, scaling the number of transactions the system is capable of up by many orders of magnitude, and dealing with various avenues for abuse.
Why would a website need to know my interests for me to be able to pay for content?
I don't get the scenario, if I got linked a particular article then you already know what article I want to unlock. If I landed on the frontpage of the site then showing snippets/headlines/previews will guide me to what article I want to unlock.
What is the scenario where there is no signal on what a visitor wants to read but that visitor would still be willing to pay for it?
I'm not sure why Google/Facebook/advertising even need to enter the conversation here.
That's exactly what I'm saying.
>I'm not sure why Google/Facebook/advertising even need to enter the conversation here.
Because I like to experiment with business models and because I wanted to show you that established business models don't change that easily.
The barrier to micropayments is friction. People don't want to have to put in a CC number or sign into Paypal to read a website. At the same time, publishers don't want to nickel and dime users, because the transaction costs of a $0.20 article aren't worth it.
Google's changes don't affect either of these issues. The changes are purely about how audiences are targeted for Google ad products.
Hard to sell ad targeting to your customers if iPhone users or Firefox users or Safari users can't be targeted.
1. Not implement it at all and depend on web developers having heard of error handling or feature detection
2. Implement it as a stub and return no data
3. Implement it to return random / meaningless data.
If Google is aware that no other browsers were going to implement FLoC and that's why they're putting the kibosh on it, why would their alternative then be just another system that those same browsers wouldn't implement?
Maybe this is more of a rhetorical question, since others in this thread have suggested all this amounts to is a name change? I dunno, this all just seems so... dumb.
I think Google is protecting their business model of targeted ads and perhaps also as a reason to lobby for tracking ("look, we tried to propose alternatives but nobody adopted it so we need tracking").
I personally don't like targeted ads even if I am anonymous so I'm not happy with the changes still. If there's an option for generic ads I'll always choose it.