5 ms·

You hit the nail on the head. Google Analytics is on a substantial proportion of the Internet. 65% of the top 10k sites, 63.9% of the top 100k, and 50.5% of th

by Smerity 12y ago

You hit the nail on the head. Google Analytics is on a substantial proportion of the Internet. 65% of the top 10k sites, 63.9% of the top 100k, and 50.5% of the top million[1]. My own results from a research project I did using the Common Crawl[2] corpus estimates approximately 39.7% of the 535 million pages processed so far have GA on them. The real key to tracking is the referrer data. For the vast majority of clicks, you land on a site that has Google Analytics or you've just left one that did. As Google Analytics tracks your referrer, that means they still have your full browsing history if you jump from GA => !GA => GA => !GA => ... According to my research[3], Google gets activity information on 51.43% of the 42 billion links analyzed in the 535 million page corpus as either the start or end of the link uses Google Analytics. This activity means they can accurately track browsing history on most sites, even those that don't use GA, simply as timing information, referrers, and knowledge of the web graph end up leaking user activity. Used in an anonymized fashion, this is beneficial as it helps Google understand real world web traffic and hence rank search results accordingly (far better than simulated activity based upon PageRank or similar). In the theoretical situation you drop anonymization is where this gets troublesome. If you're interested, there are more details at "Measuring the impact of Google Analytics"[3], though much of the discussion is on Hadoop + Common Crawl. For a privacy focused write-up (primarily worried about the NSA using Google Analytics), refer to "Google, make Google Analytics HTTPS by default"[4]. P.S. Everyone who notes "Google Analytics is easy to evade" are correct but missing the broader point -- the majority of web users will never do that. [1]: http://trends.builtwith.com/analytics/Google-Analytics http://trends.builtwith.com/analytics/Google-Analytics [2]: http://commoncrawl.org/ http://commoncrawl.org/ [3]: http://smerity.com/cs205_ga/ http://smerity.com/cs205_ga/ [4]: http://smerity.com/articles/2013/google_analytics_and_nsa.html http://smerity.com/articles/2013/google_analytics_and_nsa.ht...

MattHeard 12y ago
Could a user "fake" additional site visits in between each real visit in order to obfuscate the actual visits to GA sites?
- Smerity 12y ago
  As Google Analytics is "self reporting" (i.e. your browser tells the GA servers what it's doing) you can avoid reporting or erroneously report whatever you'd like. It'd likely be easier for you just to block Google Analytics though if that is of concern to you. In the unlikely event that fake activity became a problem, Google's well equipped to deal with it. They have a great deal of tech and brains in place to detect fake ad click activity, which is vaguely related.
- andreasvc 12y ago
  That doesn't really help if $incriminating_website is among the real visits. I think obfuscation is a waste of time given the machine learning techniques at their disposal.
atmosx 12y ago
Excellent comments and links!
tombrossman 12y ago
In the theoretical situation you drop anonymization is where this gets troublesome. Here's something even more troubling. Take an 'anonymous' crime reporting site[1] and put Google Analytics on it. Put it on every single page, even on the page with the forms to submit anonymously. Not bad enough? How about a similar site, only this one aimed at reporting corruption[2] and try the same thing. What could possibly go wrong? Both sites are well aware of the issue and have written me back when I pointed this out. This level of trust in an American ad company is curious. All I can do now is hope that whistle-blowers wanting to report corruption are savvy enough to avoid the web forms. Imagine you are a government worker somewhere and you see evidence of corruption and report it. From the same machine you signed in to GMail with. Now consider that your local government can order Google to secretly hand over tracking data and forbid them from notifying the crime reporting site(s). [1]: https://crimestoppers-uk.org/give-information/give-information-online/ https://crimestoppers-uk.org/give-information/give-informati... [2]: https://forms.theiline.co.uk/integrityline https://forms.theiline.co.uk/integrityline
ronaldx 12y ago
> P.S. Everyone who notes "Google Analytics is easy to evade" are correct but missing the broader point -- the majority of web users will never do that. This is an important point here because it's in stark contrast with Google e-mail. Google's e-mail servers are exceptionally difficult to evade. I am (apparently) able to evade Google's attempts to track my web browsing habits with Google Analytics. I can take responsiblity for that myself. This is a technically difficult opt-out process is IMHO against the spirit of "don't be evil", but at least there is an option. On the other hand, I am totally unable to prevent Google from building an accurate profile of my e-mail habits. The only way I can opt-out here is by significantly curtailing my e-mail habits (or by insisting on PGP, which will have the same effect). This is significantly more frightening to me - I have no choice.
mike-cardwell 12y ago
Isn't it about time we dropped the HTTP referer header? If we lived in a World where that header didn't exist, and somebody came along today and proposed that we add it to Firefox, Chrome or IE, there would be absolute outrage. If the proponents then argued: "Yeah, but it will make tracking users easier, and we might be able to target advertising better to make more money", people would not accept that as a valid argument.
- slashdotaccount 12y ago
  I agree, cross-site referrer headers are harmful for our privacy, but same-host referrer headers can be legitimately useful.
- rubinelli 12y ago
  The HTTP referer hearder was a very valuable tool for webmasters to correct broken links or prevent abuse years before ad targeting became a thing.
dalek2point3 12y ago
I'm a PhD student at MIT interested in doing research using the Commoncrawl dataset. Are you interested in working on this together or at least getting together and chatting about it? drop me a line: nagaraj@mit.edu

by Smerity 12y ago

MattHeard 12y ago

Could a user "fake" additional site visits in between each real visit in order to obfuscate the actual visits to GA sites?

Smerity 12y ago
As Google Analytics is "self reporting" (i.e. your browser tells the GA servers what it's doing) you can avoid reporting or erroneously report whatever you'd like. It'd likely be easier for you just to block Google Analytics though if that is of concern to you. In the unlikely event that fake activity became a problem, Google's well equipped to deal with it. They have a great deal of tech and brains in place to detect fake ad click activity, which is vaguely related.
andreasvc 12y ago
That doesn't really help if $incriminating_website is among the real visits. I think obfuscation is a waste of time given the machine learning techniques at their disposal.

atmosx 12y ago

Excellent comments and links!

tombrossman 12y ago

In the theoretical situation you drop anonymization is where this gets troublesome. Here's something even more troubling. Take an 'anonymous' crime reporting site[1] and put Google Analytics on it. Put it on every single page, even on the page with the forms to submit anonymously. Not bad enough? How about a similar site, only this one aimed at reporting corruption[2] and try the same thing. What could possibly go wrong? Both sites are well aware of the issue and have written me back when I pointed this out. This level of trust in an American ad company is curious. All I can do now is hope that whistle-blowers wanting to report corruption are savvy enough to avoid the web forms. Imagine you are a government worker somewhere and you see evidence of corruption and report it. From the same machine you signed in to GMail with. Now consider that your local government can order Google to secretly hand over tracking data and forbid them from notifying the crime reporting site(s). [1]: https://crimestoppers-uk.org/give-information/give-information-online/ https://crimestoppers-uk.org/give-information/give-informati... [2]: https://forms.theiline.co.uk/integrityline https://forms.theiline.co.uk/integrityline

ronaldx 12y ago

> P.S. Everyone who notes "Google Analytics is easy to evade" are correct but missing the broader point -- the majority of web users will never do that. This is an important point here because it's in stark contrast with Google e-mail. Google's e-mail servers are exceptionally difficult to evade. I am (apparently) able to evade Google's attempts to track my web browsing habits with Google Analytics. I can take responsiblity for that myself. This is a technically difficult opt-out process is IMHO against the spirit of "don't be evil", but at least there is an option. On the other hand, I am totally unable to prevent Google from building an accurate profile of my e-mail habits. The only way I can opt-out here is by significantly curtailing my e-mail habits (or by insisting on PGP, which will have the same effect). This is significantly more frightening to me - I have no choice.

mike-cardwell 12y ago

Isn't it about time we dropped the HTTP referer header? If we lived in a World where that header didn't exist, and somebody came along today and proposed that we add it to Firefox, Chrome or IE, there would be absolute outrage. If the proponents then argued: "Yeah, but it will make tracking users easier, and we might be able to target advertising better to make more money", people would not accept that as a valid argument.

slashdotaccount 12y ago
I agree, cross-site referrer headers are harmful for our privacy, but same-host referrer headers can be legitimately useful.
rubinelli 12y ago
The HTTP referer hearder was a very valuable tool for webmasters to correct broken links or prevent abuse years before ad targeting became a thing.

dalek2point3 12y ago

I'm a PhD student at MIT interested in doing research using the Commoncrawl dataset. Are you interested in working on this together or at least getting together and chatting about it? drop me a line: nagaraj@mit.edu