Google is working on a new technology called Privacy Sandbox to replace the need for advertisers to track individuals with third party cookies. This is interesting to me for a couple of reasons:
Google is an advertising company, so they definitely want the ability to continue being able to classify users into cohorts based on their behavior in order to deliver targeted ads to them.
But they are also a browser company (as the makers of Chrome), and there appears to be enormous public (and increasingly competitive) pressure on browser vendors to provide stronger privacy protections by default in browsers. Apple for example has rolled out various privacy measures in its very popular mobile browser Safari. I speculate that Google is concerned that if they don’t find a way to better protect user privacy (while still preserving the ability to provide targeted advertisements) that an increasingly privacy-conscious public might eventually reach a tipping point where folks might start abandoning Chrome en masse for a competitor browser that is viewed as more privacy-conscious.
So Google has twin incentives: to support tracking users to put them into cohorts that can be addressed with targeted advertising, and to do this kind of tracking in a privacy-conscious way. Hence I assume these incentives are how Privacy Sandbox has come to be.
So how does Privacy Sandbox work exactly? Similar to current advertising practice, there still exists an algorithm which is fed a user’s browsing history and that algorithm is used to put the user into an advertising cohort. However the interesting change with Privacy Sandbox is that this algorithm is run locally within the user’s browser, and never actually needs to transmit the user’s browsing history over the internet. Instead only a code for the identified cohort is transmitted, and a remote ad server can use that cohort directly.
On one hand I like that this scheme implies that third party advertisers aren’t constructing a detailed dossier of my browsing history themselves.
On the other hand I believe the transmitted cohort code could inadvertently reveal that I’m a member of one or more protected classes, which could cause online companies I interact with to treat me in ways that I don’t like or seek to predate me. Google has identified this as a risk and seeks to mitigate it by doing things like recognizing certain protected classes and removing features from any identified cohort that put it to close to a protected class.
Update: Everything discussed about Privacy Sandbox so far is only talking about the FLoC component of it. But there is also a TURTLEDOVE component which avoids the problem of transmitting one’s cohort code outside the browser by running the full ad bidding process on the local device. Neat! (Thanks to jefftk on Hacker News for the clarification.)
Now a number of advertising companies who like their tried-and-true methods of third party tracking are worried about Google closing the door on traditional third party tracking in favor of Privacy Sandbox and are now pursuing anti-trust actions against Google regarding impending removal of support for third party cookies (which is a core technology for enabling third party tracking).
I’m personally optimistic about the promise of Privacy Sandbox and removing the ability of browsers to broadcast my browsing history to third parties. If it can meet its stated goals, enabling businesses that rely on internet advertising to stay afloat while making the mechanism of advertising also be privacy-conscious, then I’m all for it.
Shame on the NSA - Back in 2013 I was particularly vexed with the kind of privacy-invasive tracking that Edward Snowden revealed the NSA to be conducting on all US citizens, and almost certainly still is.
Exploring Onionland: The Tor .onion Darknet - Tor
is a technology that provides strong anonymization when browsing the internet.
It also gives access to Tor Hidden Services (
.onion domains) which is like a
parallel version of the internet. Here I describe my brief adventures in
exploring Tor Hidden Services.