News for hackers. Stuff that matters.

teamjimmyy

We now have trackers acting like first-party properties, but where do you draw the line between first-party and third-party? What I mean is, if I build my own in-house analytics app that does a lot of what Adobe's product does, should that be blocked too?

I specifically mean for site analytics, like Experience Cloud or GA, not serving ads. Ad block is different IMO.

If this is hosted on a first party subdomain you're already blocking the ability of it to set third-party cookies and track you across sites. So, in practical terms, what's the difference between this CNAME trick and building the same thing in-house to run your own analytics?

Re:Do Not Track is not respected on mozilla.org on Wednesday July 31, 2013 @04:02PM (#6136474)
Yeah, I saw the bit about turning off all 3rd party cookies, which made me happy as I already do that myself.

As for the ubiquity and potential for data sharing among Google services, I suppose I hadn't though that entirely through. I know there was one analytics company claiming it could track individuals between devices using some fancy statistics, but I assumed it was snake oil (it was not GA claiming that).

Anyway, I hear ya, and thanks. I can see a case against GA specifically, though I have a hard time swallowing it against all analytics. I suppose it's a question of trade-offs that people are willing to make.

Re:Do Not Track is not respected on mozilla.org on Wednesday July 31, 2013 @03:37PM (#6136314)
I'll say here what I said in the other reply, but briefly.

There's a difference between a 3rd party doing the analytics and a 3rd party cookie. GA can (and should) use a 1st party cookie for this, which would make it impossible for them to correlate between sites. As a bonus, turning off 3rd party cookies also breaks ad retargeting, which makes everything better.

At that point, it's the same as Mozilla doing it themselves, but your concerns about JS being more potentially intrusive is valid.

note: i may be wrong about GA using 1st party cookies. if so, that's really not cool.

Re:Do Not Track is not respected on mozilla.org on Wednesday July 31, 2013 @03:31PM (#6136270)
This isn't strictly true, which is why I made the differentiation above between 1st and 3rd party cookies. With the 1st party cookie you'd get a new GA cookie on each site (e.g. mozilla-GA, ycombinator-GA, etc), making those correlations impossible. In the case of 3rd party cookies, yeah, I totally get that they can be used for some seriously evil things.

It's possible GA could try to correlate IPs or browser fingerprints between 1st party cookies over multiple sites, but proxies and mobile devices would make that difficult. The fact that all the data is together in GA's warehouse doesn't change the fact that the data isn't there to be correlated.

As for JS being able to be more intrusive, sure, I get that. At that point, I suppose you have to trust the site you're on that they wouldn't use a service that was intrusive. Perhaps this is a bridge too far for some, which is reasonable.

I guess I just don't get wanting to ban the tool entirely when it could but is not currently be used nefariously. (working on the assumption that if GA started fingerprinting browsers someone would've seen the traffic by now. it's not easy to hide.)

Re:Do Not Track is not respected on mozilla.org on Wednesday July 31, 2013 @10:14AM (#6133785)
I don't get it. Why is this different from inspecting your web logs? Sure you lose the first-party cookie aspect, but I bet you can get awful close just looking at the request IPs. There's "tracking" inherent in how everything works, so why does it matter if collection is contracted to a 3rd Party?

Does the poster expect the web server to not write a log line because he sent a DNT header too?