10 ms·
It's been a long time since this incident occurred, but from what I remember, Safari had/has a (non-default?) setting that blocked third party cookies, but allo
by Ajedi32 8y ago
It's been a long time since this incident occurred, but from what I remember, Safari had/has a (non-default?) setting that blocked third party cookies, but allowed them upon user interaction. At some point Google added +1 buttons to their ads, and some aspect of how they implemented that tricked Safari into thinking the user interacted with the ad even when they didn't and allowed cookies to be sent. After this was pointed out Google updated the +1 buttons to no longer trigger this behavior in Safari. (Correct me if I'm wrong; this all happened years ago so I may be misremembering.)
- yuhong 8y agoI mentioned it in the essay: http://yuhongbao.blogspot.ca/2018/04/google-doubleclick-mozilla-essay-final.html http://yuhongbao.blogspot.ca/2018/04/google-doubleclick-mozi...
- ballenf 8y ago> After this was pointed out Google updated the +1 buttons Has Google ever claimed that the code had an innocent purpose and the privacy-bypass was purely accidental in nature? The descriptions of the code imply that not breaking Safari's privacy features required much less and much simpler code than what was in place. If they didn't, then it wasn't so much as having it "pointed out" as "found out". The context at the time (as best I recall) was this came on the heals of the wifi-snooping street view cars that cracked weak wifi encryption and recorded traffic, so maybe people weren't in the mood to listen to any defense.
- abraham 8y agoThe street view cars didn't crack any encryption. https://www.theguardian.com/technology/2010/may/15/google-admits-storing-private-data https://www.theguardian.com/technology/2010/may/15/google-ad...