Hacker News new | past | comments | ask | show | jobs | submit login

I suspect that in a few years it will not matter.

Google has enough might to fix this and nobody could compete.

All google needs is to have sites talk via a backend connection to google - no adblocker could fix that - and it could train AI backend supercookies based on your behavior. You could be indentified at key points and it could pick you vs your spouse or kids all on the same ip address all contacting multiple sites and know it is you.

Current CDNs already do some of this backend tracking. People think that CDNs are about caching content for faster load times.

What is actually happening is that they are individually identifying people in an unblockable unstoppable way across multiple sites and compiling data on the backend.

Seriously, look at any cdn request. Maybe stack overflow maybe steam. anything with edgecast or akamai* (especially akamaihd) or fastly or more. you are being identified and tracked. google could take this all the way where others couldn't.




Agreed on the fact that Google could definitely do that. I'd be surprised if they didn't have at least one person working on such a solution as a backup plan.

> Seriously, look at any cdn request. Maybe stack overflow maybe steam. anything with edgecast or akamai* (especially akamaihd) or fastly or more. you are being identified and tracked.

Do you have sources on that?

I mean, technologically, it's feasible, but I have never heard of CDN providers actually doing it.


I have had friends who have worked at CDNs.

but just look at wikipedia:

> CDN providers profit either from direct fees paid by content providers using their network, or profit from the user analytics and tracking data collected as their scripts are being loaded onto customer's websites inside their browser origin. As such these services are being pointed out as potential privacy intrusion for the purpose of behavioral targeting and solutions are being created to restore single-origin serving and caching of resources.

https://en.wikipedia.org/wiki/Content_delivery_network#Secur...


How could a policy exist to counter this? Ban fingerprinting as a whole? The CDN actually has to know very little, right?


Collecting, storing, selling, or otherwise using any digital data in any way traceable to a person, entity including device or location, or other single identifying point of information should require informed consent and contract with minimum dollar value set by law to like $10. Add a bypass where entities do not have to sign to this contract to use the app or site, opting out of all tracking and data collection beyond those specified in ToS for subscribers or registered users of the service.

Now every service in the US that can identify any individual entity must agree to a contract with minimum monetary value for every entity or stop collecting data that has any use beyond generic trends, much like Google trends itself. Maybe eliminate third parties without their own user consent and contract so if I consent to the site having my data (for $10) then any deal with third parties must reflect a smaller amount (say $2) per third party data pull. Fold that cost into the cost of selling user data.

Add exceptions for good faith research, education, healthcare, and similar groups (not all nonprofits but select types or areas of work) so they do not have a minimum dollar amount but maintain a contract with informed consent by both parties. Force opt out by default behavior by making auto-check options to subscribe, agree to terms, etc all unchecked with disclaimer about what you are signing up for by checking including email volume or third parties. Add language for a simple overview of any contracts or ToS acknowledging the law and entity rights including ability to use the service without data collection.

Basically, F u Pay me but for digital data with further third party deals returning on the investment that is your site vising, sales, etc habits. Make it so companies either have to fork over real money (not pennies on their revenue) or stop collecting data.


> All google needs is to have sites talk via a backend connection to google - no adblocker could fix that

Some ad tech companies already do this. They exchange and aggregate user tracking data with content websites on the backend.


> All google needs is to have sites talk via a backend connection to google

Trading of user information should be outlawed.


Trading within what context? If between corporate entities, I think that would give Google and other large companies a larger advantage because sharing within the company is allowed.


That's what GDPR is for. Separate consent required for each usage.


Not if Google is broken up.


>> Trading of user information should be outlawed.

And collection of user information.

This will be slightly tricky because of real businesses that need to know certain things about you, but those are very few.


There are, indeed, very few of them. They just happen to be your employer (Background and credit checks), your landlord (Background and credit checks), and your bank (Credit checks).

I'm not sure Facebook, Google, etc, are quite as capable, or as motivated as any one of those three, when it comes to making your life very uncomfortable.


And doctors, but in the US they are covered by HIPPA.


When you say "trading of user information", you mean API calls, right?


Google's new cross platform development framework flutter has a mode on the web called "canvaskit", where the page is rendered as a giant canvas that you get to interact with. This makes all forms of extensions practically useless.

also notable, with firefox to longer allowing anything but a handful of carefully approved extensions on mobile, mobile users haven't much of a chance at all to navigate the web in their preferred fashion. this attack on mobile users, this non-user-agent-ified browser, makes me very sad.


> Seriously, look at any cdn request. Maybe stack overflow maybe steam. anything with edgecast or akamai* (especially akamaihd) or fastly or more. you are being identified and tracked.

Do you have sources on that?

I mean, technologically, it's feasible, but I have never heard of CDN providers actually doing it.


Have you heard of AMP?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: