Live data from Hacker News

Google acknowledges XKCD #1361

news.ycombinator.com

31–40 of 44 posts

Re: Google acknowledges XKCD #1361

#31
post #21

Earlier quoted context omitted.

They could use it to build more accurate profiles of their users in order to monetize them.

> We don't correlate or combine information from our temporary or permanent logs with any personal information that you have provided Google for other services. -- https://developers.google.com/speed/public-dns/privacy

They also don't work with the NSA.

Re: Google acknowledges XKCD #1361

#32
post #21

Earlier quoted context omitted.

> We don't correlate or combine information from our temporary or permanent logs with any personal information that you have provided Google for other services. -- https://developers.google.com/speed/public-dns/privacy

They also don't work with the NSA.

Define "work with"

Re: Google acknowledges XKCD #1361

#33
post #26
post #25

Earlier quoted context omitted.

Interesting. I tend to believe it for some reason, although I don't know why myself. Does somebody really check if such promises are true? Are they trustworthy? It reminds me, that case with collecting wifi traffic — how did it get revealed at all?

Nine days ago the data protection authority (DPA) in Hamburg, Germany asked to audit the WiFi data that our Street View cars collect for use in location-based products like Google Maps for mobile, which enables people to find local restaurants or get directions. His request prompted us to re-examine everything we have been collecting, and during our review we discovered that a statement made in a blog post on April 2…

How do you _mistakenly_ collect "samples of payload data"? Someone mistakenly wrote/used a packet logger, compiled/installed it on the Street View cars and mistakenly executed the service to be running in the background? How does that happen?

Re: Google acknowledges XKCD #1361

#34
post #14

Maybe a stupid question, but is it monetized somehow? I mean DNS servers. After all, it's a pretty heavy load… I still don't really understand what is (is known to be) monetizable and what isn't.

People use Google much more if they can access it fast, which means more revenue. A lot of ISPs have slow DNS servers, and Google needs a good production DNS server for internal purposes anyways, so it was not that hard for the engineers to make it a public project.

Re: Google acknowledges XKCD #1361

#36
post #13

The command is only: dig +short TXT google-public-dns-a.google.com by the way, the " http://xkcd..." part is the response, but since @asdafa didn't put \n\n after the command, it's appearing at the same line.

They probably did put it in but forgot about markdown.

Re: Google acknowledges XKCD #1361

#37
post #36
post #13

The command is only: dig +short TXT google-public-dns-a.google.com by the way, the " http://xkcd..." part is the response, but since @asdafa didn't put \n\n after the command, it's appearing at the same line.

They probably did put it in but forgot about markdown.

That is exactly what happened :(

Re: Google acknowledges XKCD #1361

#38
post #26

Earlier quoted context omitted.

Nine days ago the data protection authority (DPA) in Hamburg, Germany asked to audit the WiFi data that our Street View cars collect for use in location-based products like Google Maps for mobile, which enables people to find local restaurants or get directions. His request prompted us to re-examine everything we have been collecting, and during our review we discovered that a statement made in a blog post on April 2…

How do you _mistakenly_ collect "samples of payload data"? Someone mistakenly wrote/used a packet logger, compiled/installed it on the Street View cars and mistakenly executed the service to be running in the background? How does that happen?

http://googleblog.blogspot.jp/2010/05/wifi-data-collection-u...

links to

http://static.googleusercontent.com/media/www.google.com/en/...

which explains how it happened in the executive summary and then in way more detail in the body.

Re: Google acknowledges XKCD #1361

#39
post #19
post #14

Maybe a stupid question, but is it monetized somehow? I mean DNS servers. After all, it's a pretty heavy load… I still don't really understand what is (is known to be) monetizable and what isn't.

I guess, for Google at least, getting all DNS requests of their users is worth paying for a popular DNS server. That's how they know which sites your like even if those sites don't use Google Analytics.

So if I want to bias their search in favour of my favourite sites, I should use their DNS?

Re: Google acknowledges XKCD #1361

#40
post #26

Earlier quoted context omitted.

Nine days ago the data protection authority (DPA) in Hamburg, Germany asked to audit the WiFi data that our Street View cars collect for use in location-based products like Google Maps for mobile, which enables people to find local restaurants or get directions. His request prompted us to re-examine everything we have been collecting, and during our review we discovered that a statement made in a blog post on April 2…

How do you _mistakenly_ collect "samples of payload data"? Someone mistakenly wrote/used a packet logger, compiled/installed it on the Street View cars and mistakenly executed the service to be running in the background? How does that happen?

The logger was obviouslly there, it was deliberatly collecting the SSIDs and MAC addresses.

Possibly a debug option to log the whole packets added during development, and it was accidently left on in production.

Or the whole packet was always logged, a second process would then skim just extracting the SSID/MAC (correlating with GPS), and another process was deleting the raw logs. That third process failed.

A few big drives in teh data collection devices, and possibly nobody noticed where filling up a little too quickly.

Post reply on HN