"No shady privacy policies or back doors for advertisers" proclaims the Firefox homepage, but that's no longer true in Firefox 128.
Less than a month after acquiring the AdTech company Anonym, Mozilla has added special software co-authored by Meta and built for the advertising industry directly to the latest release
I think there is a big misunderstanding about this feature. People are throwing their arms up in disappointment but in reality this is a helpful feature for privacy.
This post doesn’t even explain what the feature is or how it works. If you take the time to go read what the feature actually does, you’ll see it’s a good feature to have and it really does improve your privacy when you don’t have an ad blocker.
Just because Meta participated doesn’t mean it’s bad. If they only participated as consultants to understand the advertisement system so they can better protect us against it, it’s not bad.
From my understanding of their implementation, you have to give a Mozilla server all of your traffic history, and then they feed a curated, sanitize topic list of that activity to the advertisers.
So now we’re trusting Mozilla with your full browsing history, that seems like a really bad idea. Even if I love and trust Mozilla, I don’t want to add yet another thing to the critical path
PPA does not involve sending information about your browsing activities to anyone. This includes Mozilla and our DAP partner (ISRG). Advertisers only receive aggregate information that answers basic questions about the effectiveness of their advertising.
The explicitly say if the aggregator is controlled by hostile party, and in my scenario that would be Mozilla, they could have full access to the deanonymized data. It’s out of scope for their protocol.
And while the DAP draft is nice, it doesn’t change my threat model, it just introduces extra steps. As the absolute hunger of AI inputs for models have shown us, if a company has the capability to get data, they will. Mozilla has demonstrated they are hungry for data and money. I don’t want to give them the capability
All your data is encrypted on our servers so we can’t read it – only you can access it. We don’t sell your info to advertisers because that would go against our data privacy promise.
I’m not clear on how this system works, but I would like to know how it’s supposedly better than Google’s Topics. Especially if, as comments elsewhere in the thread suggest, Mozilla’s solution involves potentially exposing your entire browsing history to someone. Topics doesn’t do that, since it’s entirely handled in your own browser and only sends vague categories. (And even fuzzes them by potentially sending a random category you didn’t actually visit.)
It’s better because PPA isn’t about targeting ads at all. It doesn’t share any browsing history, topics, or any information for ad targeting to advertisers at all. What it does do is provide a way for a website to tell your browser which ads are relevant to an action you take - for example on a checkout confirmation screen the site may tell your browser “here’s a list of ad IDs for the shop you just bought from”. Your browser then checks if it’s seen any of those ads, checking completely using local data that doesn’t leave the browser, then to an aggregator it reports which ads possibly led to your purchase. The aggregator increments a counter for each ad in its database and relays the totals to the advertiser. There are no unique identifiers or any information about your habits or interests involved.
When I initially heard about PPA I also thought it was related to FLoC / topics, but it has nothing to do with ad targeting or sharing information about habits / interests, it’s just a way to tell advertisers “Ad XYZ was effective and led to a sign up/purchase” without revealing who saw the ad or any personal information about them, just the total number of people.
I think there is a big misunderstanding about this feature. People are throwing their arms up in disappointment but in reality this is a helpful feature for privacy.
This post doesn’t even explain what the feature is or how it works. If you take the time to go read what the feature actually does, you’ll see it’s a good feature to have and it really does improve your privacy when you don’t have an ad blocker.
Just because Meta participated doesn’t mean it’s bad. If they only participated as consultants to understand the advertisement system so they can better protect us against it, it’s not bad.
From my understanding of their implementation, you have to give a Mozilla server all of your traffic history, and then they feed a curated, sanitize topic list of that activity to the advertisers.
So now we’re trusting Mozilla with your full browsing history, that seems like a really bad idea. Even if I love and trust Mozilla, I don’t want to add yet another thing to the critical path
Source.
https://datatracker.ietf.org/doc/html/draft-ietf-ppm-dap#name-security-considerations
The explicitly say if the aggregator is controlled by hostile party, and in my scenario that would be Mozilla, they could have full access to the deanonymized data. It’s out of scope for their protocol.
And while the DAP draft is nice, it doesn’t change my threat model, it just introduces extra steps. As the absolute hunger of AI inputs for models have shown us, if a company has the capability to get data, they will. Mozilla has demonstrated they are hungry for data and money. I don’t want to give them the capability
If you have syncing on, you are already trusting Mozilla with your history.
deleted by creator
You are correct. My mistake.
https://www.mozilla.org/en-US/firefox/features/sync/
Oh yeah, agreed, if your syncing then your security model doesn’t include worrying about tracking.
https://support.mozilla.org/en-US/kb/privacy-preserving-attribution
https://hackertalks.com/comment/4359282
How are they different from any other VPN service or even uBlock? They all have access to your browsing info and can potentially use it for profit.
https://www.privacyguides.org/en/basics/vpn-overview/
You think I don’t know how a VPN works?
I think you misunderstood what I meant.
I’m not clear on how this system works, but I would like to know how it’s supposedly better than Google’s Topics. Especially if, as comments elsewhere in the thread suggest, Mozilla’s solution involves potentially exposing your entire browsing history to someone. Topics doesn’t do that, since it’s entirely handled in your own browser and only sends vague categories. (And even fuzzes them by potentially sending a random category you didn’t actually visit.)
It’s better because PPA isn’t about targeting ads at all. It doesn’t share any browsing history, topics, or any information for ad targeting to advertisers at all. What it does do is provide a way for a website to tell your browser which ads are relevant to an action you take - for example on a checkout confirmation screen the site may tell your browser “here’s a list of ad IDs for the shop you just bought from”. Your browser then checks if it’s seen any of those ads, checking completely using local data that doesn’t leave the browser, then to an aggregator it reports which ads possibly led to your purchase. The aggregator increments a counter for each ad in its database and relays the totals to the advertiser. There are no unique identifiers or any information about your habits or interests involved.
When I initially heard about PPA I also thought it was related to FLoC / topics, but it has nothing to do with ad targeting or sharing information about habits / interests, it’s just a way to tell advertisers “Ad XYZ was effective and led to a sign up/purchase” without revealing who saw the ad or any personal information about them, just the total number of people.
https://support.mozilla.org/en-US/kb/privacy-preserving-attribution
Please explain to me how sending additional data from your private computer to Mozilla servers gives me more privacy and not less.