back to article Boffins prove the existence of jet-setters

According to researchers at Northeastern University in Boston, Massachusetts, most people rarely venture more than a few miles from home. And these researchers know what they're talking about. Two years ago, in some unnamed country outside the US, they acquired six months of cell phone records describing the daily movements of …

COMMENTS

This topic is closed for new posts.

Page:

  1. Anonymous Coward
    Flame

    Let me be the first...

    ...to welcome our mobile-tapping overlords. No one has anything to hide - unless the slow users were also using other information techniques like calling from petrol stations (dangerous they tell us) or somehow buying credit in a nearby country who happens to operate across the border. Only reds and commies would protest!

  2. Anonymous Coward
    Gates Horns

    Well

    It has to be the UK, seeing as Heathrow and a 100 mile circle would be a great starting point. Not withstanding the UK has laws allowing infringement of privacy even Singapore rejected as too invasive.

  3. Steven Knox
    Stop

    "Boffins" don't know science.

    "The researchers are very reluctant to describe how the data was obtained and who was involved,"

    Then the research is by definition bad science. Any study without full disclosure of methodology is invalid. All of their conclusions are dependent on the methodology being sound, and without full disclosure that is not proveable.

  4. Drew
    Black Helicopters

    Not in the EU, er, right!

    Here's a recent article describing similar digital cell data collected from Telecom Italia.

    http://www.abc.net.au/catalyst/stories/2008/03/20/2192852.htm

    Read it or watch it, RealTime Digital City, its still interesting and disturbing.

  5. Anonymous Coward
    Anonymous Coward

    Sounds like Iceland Telecom

    The numbers and situations resemble Iceland Telecom strongly. They have somewhere close to 100 thousand mobile phone users and they did have this stupid voluntary tracking service a couple years ago you could sign up for to be tracked by your friends, and I bet only 200 people signed up for that ever, hence they've dropped it now.

  6. Steve Roper
    Go

    I don't see the problem here

    Since the researchers' data would have looked something like this:

    Person 0124123456: 05:33:40GMT: 51°32'45"N:0°5'25W

    Person 0125321654: 05:33:40GMT: 51°33'48"N:0°6'32W

    Person 0128987653: 05:33:41GMT: 51°32'22"N:0°5'46W

    ...

    Person 0124123456: 05:34:05GMT: 51°32'27"N:0°5'03W

    ...etc

    Of course this tells me that Person 0124123456's name is Jonathan Andrew Citizen, married to Emma Anne Citizen on 18-6-1989 in St. Mary's Church Nottingham, has 2 children Michael Andrew Citizen DOB 12-4-1992 and Christine Jane Citizen DOB 23-9-1994 who attended school in.... you get the picture.

    There comes a point at which privacy paranoia goes too far. Now any Reg regular will know from my many vociferous posts that I am adamantly opposed to the police state the West is becoming, and I'm very concerned about my privacy. However, in this case, the researchers have no way of finding out anything about Person 0124123456 other than that they were in central London at half-past five in the morning. If that number can indeed be tied to an identity record somewhere, then yes, that IS a cause for concern. But if it's just an internal reference number intended to track a particular dot moving on a map, then I can't see an issue with it. And research like this can provide us with useful and interesting information, which I'm all in favour of.

  7. Anonymous Coward
    Coat

    Wait till they get hold of the accelerometer data...

    could be used to tell the monkey spanking habits of the population sample.

  8. Anonymous Coward
    Anonymous Coward

    Not science

    Unless the researchers fully disclose their methods and data to outside reviewers and other researchers then their research is not just bad science but is simply not science.

    Perhaps they could join up with most of the climate modellers to create a new subject where they can hoarde data without letting anybody review it and nobody bats an eyelid.

  9. Solomon Grundy

    Research Location

    Well, the names of the primary researchers are Marta González, César Hidalgo, and Albert-László Barabás. That pretty well narrows down places with a high concentration of "Spanish(ish)" names that aren't in the EU.

    Anybody want to take a guess?

    (1) Mexico

    (2) South America

  10. lglethal Silver badge
    Thumb Down

    @ Steve Roper

    The problem is that we dont KNOW that the data they received is just as you say because they wont tell us exactly what data they received and in what format. Hence why people are not happy with this.

  11. E

    Metrics

    If the data came from a metropolitan area one should keep in mind that there is very little reason for a citizen of such to travel more than a few miles.

  12. Andy Bright

    Anonymised Data?

    "How could they not know who they were studying if they had latitude and longitude?"

    I have a simpler question and probably more relevant. How can you tell if a person only moved 10 miles or less from their place of residence if you don't know who they are or where they live? Let me put it another way, even if you didn't know their names, you would at least have to know their address. Now tell me that marketing couldn't use the same information for mass mailing to consumers, especially the more affluent that they could identify as moving 100 miles away from home on at least a weekly basis.

    Sure having a name for personalised mailing is proven to be more likely to end in a sale, but even "Dear Sir/Madam" will yield a 1% response rate, of which 1% will purchase something. For those that didn't know, that's the maths that calculates the value of names and addresses to mass mailing / marketing companies.

    Personalising those names and address - a task by the way performed by minimum wage employees, using a post code book to add post codes and a telephone book to verify names - increases the odds of an actual sale dramatically (however I can't remember how much).

    As you can probably guess I worked for such a company once upon a time, and my knowledge is from first hand experience. Yes it does pay to have people looking up names to match address, as well as their post codes (post code books can be obtained from local post offices at little or no cost). The Post Office uses it's own numerical equivalent to a post code, and if a mass mailing company pre-sorts and pre-bags it's mass mailing it receives a significant discount on the cost of postage. This is then partly passed on to the magazine / junk mail / whatever company that wants all this crap posted. Probably more information than you wanted, but it does explain how these things become profitable and how having no name is hardly an obstacle to profit, nor a good way of anonymising someone.

  13. Anonymous Coward
    Thumb Up

    I always leave my mobile phone at home when...

    ...I'm out committing some dreadful crime.

  14. Peter Gold badge

    Of course they don't move around much..

    .. with current fuel prices. Duh.

    :-)

  15. Scott
    Joke

    Location

    Nah it was based in Australia

    The 2-3% was just the country folk going to buy milk

  16. Anonymous Coward
    Anonymous Coward

    Still not seeing a problem

    Every day I go to the coffee shop at the same time for 30 minutes, every day I see the same people passing the window at the same time, the same people in the coffee shop. The same office staff going into the corner shop opposite for ther daily milk before returning to the office space across the road a few minutes later. I don't know these people, where they live, but I recognize their face. Occasionally when I'm somewhere else and I recognize someone I'll actually say "hey, don't I see you every day at such and such" and conversation ensues. In my head I have clearly established a pattern for these people. Am I invading their Privacy.

  17. Big Al
    Boffin

    Non-representative sample

    Since the sample comes from only one country, it is not representative.

    Example: Czechs are notoriously opposed to the idea of commuting to work, preferring to move closer to their workplace than spend 2 hours a day travelling. Compare this to the situation in the UK, where getting workers into London from the south-east every day is a massive logistical operation.

    In other words, the cultural biases inherent in the sample make it impossible to apply the lessons 'learned' from one country to another in any meaningful way.

  18. Steve
    Stop

    @Steve Roper

    Steve,

    When they see a number that stays in one location for 10 hours every night, would it really be that difficult to check the electoral register and find out who lives at that address?

    The data might be anonymous to start off with, but de anonymising it would be trivial.

    And in case it's not on public record, they can see where you spend 8 hrs Mon - Fri and find out who you work for.

    They can match up any addresses you visit more than once, and generate a list of businesses you use and friends and acquaintances.

    But looking at the results of the study, is there anything of interest being shown.

    Duh, most people work relatively close to where they live, they have friends in the area, they shop in the area and they play in the same area.

    Some people have to travel for business, others make trips at the weekend.

    Nothing to see here (in terms of interesting results), move along.

  19. Dr Patrick J R Harkin

    @Andy Bright

    "I have a simpler question and probably more relevant. How can you tell if a person only moved 10 miles or less from their place of residence if you don't know who they are or where they live? Let me put it another way, even if you didn't know their names, you would at least have to know their address. "

    I think if someone stayed in a ten mile circle for over two weeks, we'd be fairly sure their home was somewhere in that ten mile circle. You don't need to know their address, or where in that circle their home is to make that deduction. If location detection got down to GPS resolution (which is doesn't) and you traced me over two weeks, you'd find there's a location where I spend 8 hours a day, five days a week, mostly during the day and another where I spend every night. Figuring out which is home and which is work isn't Nobel-prize level research.

  20. Anonymous Coward
    Boffin

    right...

    ...and the problem is them having access to that "anonymous" information ?

    In my view, the problem is that information is there in the first place.

    Even if the government, the telco or the researchers aren't "allowed" to legally use that information - the information is still there and that's a problem in my view, although I rather not expose my paranoid view of the world.

    Quoting Kraftwerk: Information wants to be free ;)

    "all you have to do" is hack the billing system (which in some countries usually has all the information you need, since on every billable interaction with the network gets logged) - and that might be easier than you'd think.

  21. Anonymous Coward
    Alert

    Anonymous ID Codes

    Think about it...

    like the random number on your car...(not used to trace you! no never..)

    NOT ANONYMOUS! BUT A PSEUDONYM!!

    I really wish some agency or other would hang at least some of these idiots out to dry. wake up people your supposed to be academics. LEARN SOMETHING.

    While it is not Personilised Data it is also not Anonymous Data, it is Pseudonymised Data! using an numeric Identity instead of a personal Identity.

    These people have been using Phorm's Definition! and that is bad phorm!

  22. Anonymous Coward
    Alert

    @ Steve Roper

    Person 0125321654: 05:33:40GMT: 51°33'48"N:0°6'32W

    Person 0128987653: 05:33:41GMT: 51°32'22"N:0°5'46W

    HOW FAST!!...

    but even so from data like that:

    Plane manifest list 64 persons... two have Nottingham addressess

    One attends st marys church every sunday (coordinate Data) and is listed in parish records. Bingo You were right its Mr Citizen listed (electoral register) at Number X Xstreet the Lace market Nottingham...

  23. Anonymous Coward
    Black Helicopters

    reluctant

    I have a feeling that the reason they are reluctant to disclose the "methods" is that the data was not so anonymised as they claim. In my previous experience, the monkeys who are supposed to do the data-scrambling can't be arsed, and it's much simpler for them to email (!) the raw database with a wink.

  24. Anonymous Coward
    Thumb Down

    Re:Research Location

    @ Solomon Grundy

    "Well, the names of the primary researchers are Marta González, César Hidalgo, and Albert-László Barabás. That pretty well narrows down places with a high concentration of "Spanish(ish)" names that aren't in the EU."

    What the hell does that matter? They all work in US universities, and the data they studied was all numbers. It doesn't matter what language people were speaking on their mobiles, because they weren't listening in!!!!

  25. Nigel Sedgwick
    Thumb Up

    Very Doubtful 'Anonymisation'

    I am concerned that the raw data reported to be 'anonymised' cannot be 'anonymised' in a very large proportion of cases.

    This is because location itself is largely an identifier. Even if the locations given are not particularly precise, a high prevalence of the same locations (eg home and work) could quite easily lead to identification of the individual person, with a very high probability of being correct.

    Then, obviously, further locations, even if also approximate, could disclose private information about the individual concerned.

    Best regards

  26. Anonymous Coward
    Anonymous Coward

    @Andy Bright - defining home

    I would assume the way to do this is to look at those times when a particular mobile is stationary for a long time, especially overnight. You can then extrapolate a home base from the cell data going in / out of that location.

    I would assume most people have a fairly regular pattern such:

    19:00 - 07:30 area X

    07:30 - 08:00 travelling between X and Y (probably via same route)

    08:00 - 16:30 area Y

    16:30 - 17:00 travelling between Y and X (probably by reverse route to that taken earlier)

    17:00-19:00 general tavel, multiple locations (hobbies, shopping, vandalising stuff)

    Given the nature of X and Y and the times you can work out which is home and which is work, just in case someone works shifts.

  27. MarmiteToast

    mmm

    I wonder who was number 5.

  28. Anonymous Coward
    Dead Vulture

    Woosh!

    This sounds like a very useful bit of research that must could have thrown up some interesting data. It's a shame the writer is seemingly so intent on uncovering a scandal and grabbing a Pulitzer that he couldn't see past the totally irrelevant (here) 'privacy issue'...

  29. Andy Turner

    @Jaowom

    You're a stalker! Someone phone the cops

  30. Bronek Kozicki
    Coat

    mostly agree

    @Steve Roper - indeed. Also pls note that, as the data were not collected from GPSs (or the raport missed important information), the location would be just rough estimate from signal strengh. This means that there is was privacy invasion, as from such little data there is no way of telling where someone lives or works (unless population density was one home per square mile or such - very unlikely).

    @Big Al - interesting observation

    Which comes to conclusion - useless report and useless data, only mildly interesting from privacy PoV (as mighty El Reg noticed).

    One last point - people usually do not object to having there faces and locations, along with other details, "scanned", "processed" and "memorized" by best known distributed data processing network, i.e. other people and their social networks. It is only when data are stored and made accessible via electronic means they spook.

    Mine is the one with long hood

  31. Anonymous Coward
    Anonymous Coward

    I left my mobile at home today

    And yesterday. Nobody needs to get me so why shorten its lifespan by carrying it around in a pocket with other things that might scratch it?

  32. Anonymous Coward
    Anonymous Coward

    I daily travel more than 100 miles away

    I am a lorry driver, my colleagues also mostly travel more than 100 miles from our homes and back every day with stops on the way.

    Jet-setters? Business travelers? Nah, we are just the people who fulfill the final part of the promise so we don't count.

  33. Sir Runcible Spoon
    Unhappy

    @Steve Roper

    I see your point Steve, but how about this one..

    Person 0124123456: 01:40GMT: 51°32'45"N:0°5'25W

    Person 0124123456: 02:40GMT: 51°32'45"N:0°5'25W

    Person 0124123456: 03:40GMT: 51°32'45"N:0°5'25W

    Person 0124123456: 04:40GMT: 51°32'45"N:0°5'25W

    Person 0124123456: 05:40GMT: 51°32'45"N:0°5'25W

    Hmm, I wonder where the owner of this particular cellular device lives.

  34. Sven Coenye

    Proof of concept

    This project proves that it is possible to continuously track the movements of individual members of a population over a long period of time.

    Jaowon, you're almost there.

    The anonymized data establishes the base line of average dwell times. What if a non-coffee shop area of interest is defined on the map, with attached reporting criteria when a number deviates from the base line? There would still be a lot of data to wade through, but it is all already conveniently packaged for automated analysis. Now you know who to put in the database for terrorist proclivities. So much easier than spotting Transformer's T-shirts.

    You could even establish criteria that automatically defined new points of interest: cross off all areas of congregation that map to already known venues, and you're left with all the new terrorist recruiting stations (or meeting places of the political opponents of the powers that be, but surely it would never be used for something like that...)

  35. Eddie
    Thumb Down

    The data is not that anonymous....

    The full paper has demographic information - age, sex etc. Coupled with relatively precise geographical data, it would be trivial to find the subjects.

    I'll print out a copy and leave it in the common room here for opinions.....

    For those of you with a masochistic trend, the full paper is here (Pdf):-

    http://www.nd.edu/~chidalgo/Papers/HidalgoRodriguez2008Extended.pdf

    Only 22 pages....

  36. Shonko Kid
    Black Helicopters

    Depends on the precision of the information

    @Andy Bright

    If the data is merely information about which cell a particular handset was attached to then this is less invasive. Sure, the TelCo will be able to match that to a billing address, but I'm guessing that this wasn't supplied, hence 'Anonymised'. Also, if it is just Cell information, then there wouldn't be enough precision to match that to a definite building, only an area.

    You can then assume that the 'home address' is the Cell most frequented, perhaps at specific times (morning, evening) They could use there extra sample of 200 to verify that assumption, but 200 is a very poor sample size.

    @Jaowon

    Yes, you did. Technically at least. I'm sure some fascist regime will be along soon enough to punish you for trying to interract socially with a complete stranger, (and by your own admission) completely unsolicited!

    @Big Al

    I'm still not sure what the 'lessons' were? What exactly has this study contributed to the world? Apart from confirming the bleeding obvious - some people travel a lot, most people hang around in towns, and TelCos will sell anything if they think they can get away with it.

  37. Dan Shannon

    Don't be Daft

    Eh ?

    A brief check on some "other" news indicates the (phone) numbers where encrypted to protect privacy and the assumption was based on a cell tower providing approximate coverage (about 3km) without any direct location data being used (they didn't know where you where within the the 3km zone).

    So how exactly is this a problem ?

  38. Anonymous Coward
    Anonymous Coward

    So they've proven what we all knew all along? Great! Mindblowing stuff.

    I'm interested in how an AMERICAN University that studied people OUTSIDE the US comes under "internal research purposes" of a PHONE service, especially when information about it is then published.

    Anyone ?

    Looks like they took advantage of idiots who signed their information away without knowing, due to vague wording. Even the whopping 200 out of 100,000, were not told that their information would be analyzed this way, only that their movements would be tracked- supposedly for their benefit.

    Also- I am not liking this new trend of companies illegally (or at least questionably) using people's data and destroying the information after they've used it, which they seem to think gets them off the hook because how convenient! They couldn't tell you if they wanted because the non-anonymised part of the data is now destroyed.... and all that's left are the results! (like Phorm!)

  39. Dr Patrick J R Harkin

    The concept of "anonymous data" causes a lot of confusion.

    If I take the medical records of 1000 people, and pass a researcher the ages, heights and weights with an index number I've created (so he knows that height A belongs with weight B, then to him the data is anonymous. I would keep (locked away) a record of who those numbers belong to, so that if he rings me up and says "Patient 32 is 18 metres tall according to your data - could you check?" "Oops, I meant 1.8 m".

    But the data he has is totally anonymised, and no amount of analysis he can do will break that. I admit that in medicine we're paranoid about confidentiality (once upon a time I'd go to meetings and see slides describing "Patient ASJ" and "Patient JT" but even that's not permitted now. It's subject 1, subject 2, subject 3. Poor subject 1 - every time I go to a meeting he's had some new disease befall him. And quite a few sex changes...) and other fields may not be as cautious.

  40. Spleen

    Assumption

    People assume that if data is "anonymised", i.e. their name is removed, that means the data is no longer related to them at all. One moment it's personal, then *poof* by removing one piece of data it no more relates to them than all the other records in the set. This is exactly the same mistake that people in the Middle Ages made when they believed that if someone knew your name, they could use it to hurt you in black magic rituals. Your name is not "you", it's just another piece of data about you. If instead of "anonymised" they said "removed one piece of data" maybe more people would be concerned.

    Having someone's name may make it easier to find other data about them than, say, their height, but it also works in reverse - as has been pointed out, given enough data, working out someone's name isn't at all hard. Or even narrowing them down to a small group (such as their family).

    Didn't we learn anything when AOL released its not-at-all-anonymous "anonymised" search engine data?

  41. Anonymous Coward
    Alien

    Has any El Reg reader seen the 'raw data'?

    For those of you claiming the conspiracy nuts are wasting their time, think on this. The researchers refused to reveal the source of their data, and refused to reveal their data.

    So unless one of you has seen it, how do you know if it was just Lat&Long co-ordinates, a list of info such as "Phone A moved from point XA1YA1 to XA2YA2", or if it did give name, address and telephone number?

    If the data is truly anonymized then there would be no problems with them revealing it; despite the behaviour of the Green Lobby, "scientific proof" actually requires - no DEMANDS - that the 'theory' be demonstratable every time the experiment is run and that the test data is freely available for other people to check - remember the 'cold fusion' debacle where only two researchers ever saw it work?

    If you're prepared to take this report at face value, then why not cold fusion?

    So who's gonna build the first FTL starship and prove Einstein was talking out his butt (the poor old chump)?

  42. ImaGnuber

    The Real Danger

    ...is that legitimate* research might be constrained by social issues such as, as is indicated by the comments re this case, the fear of even a possibility of a privacy breach.

    *'legitimate' does not mean or require that it must be seen to be useful by all and sundry.

  43. Christopher Smith
    Go

    I'm with Steve Roper...

    "In most cases, they approximated customer movements by pinpointing the cell towers routing their various calls and texts."

    Pinpointing location to the nearest cell tower doesn't seem terribly accurate or terribly threatening to me.

    Although to be fair, the data could well have been truly anonymised - phone identified by a unique and unattributable number, all location information transformed to some random ground zero within the geographical area covered. Dates removed and replaced with offset from day zero. So ...

    #1 N5.1km+/- 2km, W1.8km+/- 2km, Day #1 7:30am

    Its not as if we're not talking supercomputers here...

  44. Anonymous Coward
    Anonymous Coward

    Cell Phone Tracking

    A few points worth making:

    1. It's a fairly safe guess that they were doing Cell Id tracking, not latitude/longitude as some commenters have suggested. There are several reasons for this: 1) it's easy to collect as part of normal network operations (and however the network may feel about your privacy, they *do* care about the money they'd have to spend supporting this research); 2) the cost of tracking 100,000 people using any of the better positioning techniques (Angle of Arrival, Time Difference of Arrival, etc.) is prohibitively high; 3) if they collected anything more accurate then you can bet that the company would become a bit concerned about the privacy risks as well.

    2. I think that Barabas is either being slightly disingenuous or rather naive when he says that some (hexadecimal?) hash is anonymous and therefore 'secure'. As someone else has pointed out, that's *pseudonymous* data, and that means much lower levels of privacy protection. There are a number of well-known vectors of attack against this class of information and if they'd done anything particularly clever to protect the anonymity of the users then you can bet they'd be touting it to the heavens. I'll be quite interested to see what would happen were I to request access to this data for research purposes.

    3. The data from Rome (@Drew), on the other hand, actually is anonymous. That research uses Erlang (there was an article on this research project in the IEEE), which is a measure of bandwidth usage that evolved from analog cellular networks. 1 Erlang is one person-hour of usage, so you don't really know if it's 1 person on the phone for a *looooong* time or 60 people on the phone for a minute. The GPS data for that research came from buses and taxis, which is hardly what people should be getting up in arms about.

    4. I'd be rather unpleasantly surprised to find out that the network had shared billing location with the researchers since you can deduce home and work locations simply by looking at where people spend the most time. On the theory that nearly everyone returns home on weekends, and can often be found at home on weeknights, it's not too hard to guess where 'home' is for any randomly-selected user. This hardly resolves the underlying privacy issue with this data, but if you combine having to guess at someone's house using statistics with the diameter of the average network cell (100m in downtown areas, 5km in rural areas) then this does make it *slightly* harder to figure out who someone might be in an automated way (you'd get a lot of false positives and false negatives unless you were looking for a specific person whose habits you already knew).

    It also behooves me to talk about some of the positives of this type of research and some of the permission-based issues that arise. For things like transportation planning and infrastructure provision, it helps a *lot* to understand where people are and when, as well as where they are trying to go. Most of this type of planning is based on random samples of a few hundred people, but if you could figure out that a lot of people are trying to get from A to G then you could redesign your bus routes so that an express bus did that segment directly rather than wandering through B, C, D, E, and F. You can also makes guesses about travel mode by tracking the speed of a phone, so this (again) can help you to figure out how to deliver services better (maybe a bus-only lane to encourage a switch to public transit, or maybe the placement of government offices or public service announcements at locations where people are likely to see them/encounter them at convenient times). Or in the event of a terrorist attack or natural catastrophe you could determine pretty quickly how many people were affected... For all of this type of work if you use mobile phone data then you can start to work with samples of hundreds of thousands so your findings and predictive models get much, much better. The challenge is to do it in a way that is provably private... I don't think that this research lives up to that standard, so interesting as it is I have some serious qualms about how it was done, and Barabas' reticence suggests that he does as well.

    The problem is that securing permission from several million people is, frankly, impossible, as is managing opt-ins and opt-outs for each of them individually (at what point in the process do you filter out the opt-outs? how do you manage permissions in the first place?). So there's a debate that needs to be had around whether the public value of research using this type of data outweighs some arbitrary level of concern about privacy. If I could promise you a 1 in 1,000,000 chance of being reidentified would that be enough if I could also promise you better public transit or public services? How about 1 in 100,000? Right now permission is an all-or-nothing game on both sides of the debate, and I don't see that as very constructive because this data really could be used for *your* benefit (and not just for advertising and surveillance).

    And on a final note, I'd guess that this research used the same operator (and possibly data set) as an article on social networks that appeared in the Proceedings of the National Academy of Sciences (PNAS) last year. I've heard rumours about this data set and all I can say is that I'm 99% certain it didn't come from the UK (small comfort, I'm sure).

  45. Anonymous Coward
    Black Helicopters

    Fear Data Mining Gov, not researchers.

    So....

    You, the gubment, have 3 unregistered SIMs making their way from one of a few possible streets, to one of a few possible offices (geographically grouped) every day, along with the hoardes of other signals going about wherever.

    Location data cannot pinpoint a specific house or office. You cannot dedicate human attention to every single signal, so this mining has to be automated. There's no problem, but you'd like to know who is using those phones, because, you know... just in case!

    You mine just a few categories of your data (you have names and addresses of registered SIMs, employment/taxation records, electronic purchases and airline passenger name records) and you're laughing.

    An office fire drill will show a load of signals moving together, unexpectedly. Tally those movements with registered SIMs, look up the employment records of those phone users, and you know what office the unregistered users are working at.

    A quick check of flight records for the people living in the residential areas the SIMs spend most nights, and check against roaming records or lengthly periods spent switched off, and you're probably all sorted. If there's still 2 or 3 you're not too sure of, check out credit/debit card purchases. And the vict^H^H^H^H citizens are none the wiser.

    Moral of the story: The more data sources you have, the less detailed (and therefore easier to "market" to the sheeple) those sources have to be.

  46. Anonymous Coward
    Thumb Down

    Big Bro.

    If you think this does not happen in UK on a contiuous basis, think again.

    It has been done for years by XXX and XXX.

    The ONLY way to prevent it is to REMOVE your mobile battery when not in use.

    Always pay in cash.

    Oh, and always microwave your car number plates for 30 seconds before use.

    UNLESS you have something to hide ?

  47. Charles Smith
    Black Helicopters

    Private Enterprise

    I'm thinking of offering a service where pretty assistants (girls), for a small fee, take your mobile phone when you arrive at the local commuter rail station and lend you an alternative phone.

    Your phone will be loaded on a Thames barge that floats up and down the river all day and then your phone is returned to you at the end of the day. The men in black will be convinced that there are many more river commuters and that their green objectives have been achieved.

    A magnificent side effect is when you employer calls you (rather than vice versa) they only disturb the peace of the interior of a barge.

    A variation on this service is to be able to drop your phone off at a motorway service station, whereupon it will spend the day travelling in the opposite direction or round the M25.

  48. Trygve

    @steve

    "When they see a number that stays in one location for 10 hours every night, would it really be that difficult to check the electoral register and find out who lives at that address?"

    What address? what are you talking about? Making the huge assumption that they did the study in the way they claim, the only "address" information they would have is the location of the cell tower the phone is spending most of it's 'home' time talking to.

    Given that cells are typically inversely sized to user density, the only time they would be able to narrow it down to a small geographic area would usually be when that small area is full of apartment buildings and such - so there's still no way of figuring out individual identities without an NSA-sized correlation exercise against other data sets. In my London suburb, even a picocell or whatever covering only a few hundred square metres would cover at least half-a-dozen residences and probably twenty GSM devices. An actual tower would probably have several hundred devices associated with it at any time.

    They might have done the study in some crazy country where cells regularly only had a handful of devices in them - but then I think they would have been seeing much longer journes as people trek through the wilderness on a daily basis.

  49. Anonymous Coward
    Anonymous Coward

    Internal Research

    @Anonymous Coward: simple, you assign a bunch of internal staff to the project, they hire in some 'contractors' (i.e. university researchers) to do the legwork (possibly on-site, possibly not), and presto, it's internal. I've seen at least two research projects structured this way.

  50. Anonymous Coward
    Anonymous Coward

    I'm with Steve Roper too

    People seem to believe that ALL phones have GPS switched on, and report their GPS coordinates to the cell tower. They don't.

    The cell tower only knows that the phone is in its signal zone... where in the signal zone it is, is irrelevant to the signal tower. If you have three signal towers from your mobile company around, and you switch towers when you move 20 feet, then yes, you could be triangulated to a zone in which the towers who receive your mobile signal have dual coverage. It doesn't go down to street level, unless you're in an unfortunate spot in which both towers have equally crap signal and you bounce back and forth between them, and it's the block of flats you live in.

    The data only tracked which towers you moved between, and of course, the finer the tower coverage (many towers with short coverage provide a much more fine picture of your movements), the better the tracking.

    Tower data that shows no movements overnight only proves that people leave their mobiles on overnight in that tower's signal area.

    I have to agree with EPIC though that the definition of 'internal research' is stretched here. However, if a company's definition is "data can be given to researchers who work for us, but are not actually employees of us", then that's all good.

Page:

This topic is closed for new posts.

Other stories you might like