It Was Mrs. Peacock, In The Lounge, With A Text Message

from the CSI:-SMS dept

Information from mobile phones is already a well-established part of the police's forensic arsenal, as location data is commonly used to put criminals at the scene of a crime. But one problem is that location information only establishes a phone was in a particular place, and without corroborating information, it's possible to argue that a certain person may not have had that phone in their possession -- and hence, been somewhere else. In an effort to help investigators better determine who was using a phone at a particular time, in addition to where it was, researchers are now trying to determine ways to detect who wrote a particular SMS. Much like handwriting or linguistic analysis can be used to establish if the same person wrote two particular things, researchers think they can tell whether a group of messages were written by the same person by looking at how certain words are abbreviated, or the way messages are structured and written. While it remains difficult for law-enforcement bodies to intercept text messages in real-time, the growing push for data-retention laws that include them, as well as advances in forensic analysis, will likely make them a much more valuable tool for investigators in the future.


Reader Comments (rss)

(Flattened / Threaded)

  1.  
    identicon
    Egat, Oct 9th, 2006 @ 9:39am

    Data Retention now a good thing?

    the growing push for data-retention laws that include them, as well as advances in forensic analysis, will likely make them a much more valuable tool for investigators in the future.

    So data retention wouldn't obscure the problem by throwing more data at it?

     

    reply to this | link to this | view in thread ]

  2.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 9:52am

    yeay...another article about using sms to "track" criminals.

    i've said it before, and i'll say it again. even though handwriting can be studied, sms isn't the same

    it's an abbreviated language. whereas there are limited ways to abbreviate and talk and whatnot. just like programming, there's a limited ammoount of ways to do something...

     

    reply to this | link to this | view in thread ]

  3.  
    identicon
    Flamsmark, Oct 9th, 2006 @ 10:21am

    Re:

    even with the programming anlogy, that's not the be all and the end all. my high school computing teacher can ideantify which programs were written by which members of the class by the source code.

     

    reply to this | link to this | view in thread ]

  4.  
    identicon
    Faz, Oct 9th, 2006 @ 10:27am

    Re: Re:

    I agree, english teachers can tell english papers were written by whom. I just was thinkign about all the messages i send and they definately have simlarities in sentence structure

     

    reply to this | link to this | view in thread ]

  5.  
    identicon
    Tack Furlo, Oct 9th, 2006 @ 10:48am

    More like linguistics than handwriting

    The actual techdirt post compares this to linguistics and handwriting, but I think we all agree it's closer to lingistics. In a way, it allows you to rule out some people who would never use certain words or abbreviations, but I don't see this like a fingerprint.

    There are certain things text massages do that are based on both usage and which phone is used. For example, I might use one word on my SDA with T9 input disabled (because it's hard to hit the 5 key I might use a different word) versus a RAZR with T9 enabled. For that matter, if you sit someone in an interrigation room for 10 minutes and they use an ucommon woord 15 times, and that same word is in a text message, it's a pretty good bet they sent the message.

    Just the same, often the government will make people submit to a typing test. If a criminal sends 5 emails in 5 minutes and a suspect can't get more than 2 done, you can rule them out. The same can be applied to a text message - if a supect either can't type that fast, or simply can't use the phone (i.e. if a RAZR was used but I can't fgure out the menus on a RAZR) then such signs (though completely inadmissable in court) allow the police to rule out suspects even if it doesn't tell them a given person did it.

    I think this stuff is a far cry from evidence (though some of it can and probably will be one day). On the other hand, before they used it to aquit OJ Simpson, nobody would've ever considered using the fit (or lack thereof) of a leather glove to prove innosence or guilt of a person on trial for murder. DNA, just the same, wouldn't be admissable now if some lawyer in the early 90's hadn't decided to try it and then have the supreme court uphold it. I'd like to see how this plays into major cases of today (such as Rep. Foley's case if he used text messages). Surely, his attorney is intelligent enough to bring up "someone hacked his computer and sent these messages" as a viable defense. It'll be interesting to see how that plays out.

    That, plus wo the hell enters alcohol rehab right before you get acused of being a petphile? Call me stupid, but I'd immagine if someone accused me of that, the first thing I'd do is go get drunk!

     

    reply to this | link to this | view in thread ]

  6.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 10:51am

    Who am I

    This is a text message.

    I've put plenty of posts up here on Techdirt.

    Can you tell me what name I normally post under?

    Absofuckinglutely rediculous...

     

    reply to this | link to this | view in thread ]

  7.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 11:00am

    to the programming example, in a class i had, about 20 kids, we had "cheaters" on the final. 2 kids, never worked together before, wrote the exact some program. same variables, same format. line by line the exact same code. issue? they weren't sitting anywhere near eachother. how can that be? impossible? no. it happened. on a simmilar note, in other classes, the teacher noted that several programs were very simmilar, including uses of variables. they tried to claim cheating, however when writing simple code, it's pretty hard to distinguish youself from others.

    with txt messaging, it's the same way. most messages aren't more than 20 words. mainly, it's under 10. how can that be a good sample of how a person writes?

     

    reply to this | link to this | view in thread ]

  8.  
    identicon
    U don't know and I don't care, Oct 9th, 2006 @ 11:02am

    One thread

    It seems to be only one thread in a (hypothetical) fabric of evidence. Just because you were walking by the address of a crime scene, even while that crime was happening, and sending a text message, doesn't mean you committed the crime. Piecing together probable typing similarities with OTHER evidence can help to build a case to arrest.

     

    reply to this | link to this | view in thread ]

  9.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 11:02am

    no

    http://www.google.com/search?q=site:techdirt.com+Absofuckinglutely

    no results for the one uncommon word there (until this thread is crawled) -- so I guess not. But we'll be watching...

     

    reply to this | link to this | view in thread ]

  10.  
    identicon
    u no, Oct 9th, 2006 @ 11:17am

    crap

    next time i conspire to commit a crime i will have to use something other than a txt message. guess i can not tell all my friends the next time i download a song illegally. This is stupid. It seems to me that the mosern court system has decided that it is guilty until proven innocent, not the other way around. Sorry my comment is so offf the wall.

     

    reply to this | link to this | view in thread ]

  11.  
    identicon
    RiskyMethodz, Oct 9th, 2006 @ 11:24am

    Re: Who am I

    DigitalBomb?


    I think it's hit on the head to say that comparing a short txt message to the content of one's thesis paper is not the same. In most txt messages, people tend to say the same things, and show very little differentiation from each other.

    wat r u doin

    r u comin home

    im in class

    wen do u want food

    ....are samples of text messages really significant enough to really rule anything out? Not to mention, just as they're saying now that people know their phones can be trackable so they do not have them with them when commiting a crime, do you think it's that difficult to knowingly spell a few small words differently? spelling of 4 letter words is not as significant as comparing the actualy characteristics of one's handwriting. I don't believe there is any plausible way to use text "style" to determine much anything...

     

    reply to this | link to this | view in thread ]

  12.  
    identicon
    Dosquatch, Oct 9th, 2006 @ 11:26am

    Re: no

    no results for the one uncommon word there [...]

    What, no mention of "rediculous", which isn't even a word?


    I sincerely apologize for turning language nazi, and humbly submit myself to a proper flailing with a dictionary.

     

    reply to this | link to this | view in thread ]

  13.  
    identicon
    Dosquatch, Oct 9th, 2006 @ 11:27am

    Re: Who am I

    Can you tell me what name I normally post under?

    Oh, shut up, Dorpus.

     

    reply to this | link to this | view in thread ]

  14.  
    identicon
    TriZz, Oct 9th, 2006 @ 11:43am

    Re:

    Where most people abbreviate text messages, I write texts like they're professional e-mails. It'd be really easy for someone to track wether or not it was me. Even my mother texts like a 14 y/o girl...

     

    reply to this | link to this | view in thread ]

  15.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 11:57am

    well, then any crime commited with formatted text will come back to trizz

     

    reply to this | link to this | view in thread ]

  16.  
    identicon
    Anonymous Coward, Oct 9th, 2006 @ 12:07pm

    Re: Data Retention now a good thing?

    the growing push for data-retention laws that include them, as well as advances in forensic analysis, will likely make them a much more valuable tool for investigators in the future.

    So data retention wouldn't obscure the problem by throwing more data at it?


    I think you are confusing "data harvesting" (where the government simply aggregates large amounts of data using automated processes in hopes of finding a pattern that could indicate a crime has occurred, is occurring or may occur in the future) and "evidence" (where a police officer with a warrant has "probable cause" to believe a crime has been committed and now investigating to confirm his suspicion).

    Data harvesting is a problem because the idea of simply storing hundreds of terabytes of data in hopes that one day you may need to aggregate that data would be a serious problem. For example.. a federal agent showing up at an ISP and asking them to provide all emails that include the word “terrorist” during the past 5 years. If all emails were retained for 10 years.. then it becomes a technological burden for any ISP to simply search through hundreds of terabytes of email in hopes of finding a single email that would be useful. In this case.. yes.. the needed email would be "obscured by throwing more data at it."

    The easier solution would be for the governement to develop technology that helps computers identify possible child porn or terrorist related chatter then store that data until a warrant is secured. I am still not sure if this is even technologically feasable right now but it might be in the future.

    Either solution still leaves the probelm of privacy since data is almost ALWAYS abused. (aka: AOL's recent "unintentional" release of personal information through search queries).

     

    reply to this | link to this | view in thread ]

  17.  
    icon
    chris (profile), Oct 9th, 2006 @ 12:26pm

    hopefully dumb criminals are easier to catch

    wat u doin?

    shot marvin in face

    he ok?

    no. dead.

    WTF?

    chek ur myspace i sent u pix

    kewl

    can i crash at ur place 2 nite?

     

    reply to this | link to this | view in thread ]

  18.  
    identicon
    Okita Soji, Oct 9th, 2006 @ 12:36pm

    Re: hopefully dumb criminals are easier to catch

    Sounds like someone was catching the edited version of Pulp Fiction on AMC's "Sunday Night Movie". Hence....shot "Marvin" in the face....

    Yeah I'm just creeped out in general with the complete ambiguity regarding the "limits" of secure communications via SMS. I'm starting to become of the opinion that the only "secure" information is what's between your ears. Farenhiet 451 anyone??

     

    reply to this | link to this | view in thread ]

  19.  
    identicon
    Justin, Oct 9th, 2006 @ 12:53pm

    I'd be easy to spot

    But I might get confused with TriZz. I will also long hand my text messages.

     

    reply to this | link to this | view in thread ]

  20.  
    identicon
    John Duncan Yoyo, Oct 9th, 2006 @ 1:11pm

    Re: Re:

    I tend to use T9 which guesses what you are trying to spell from the buttons pushed and gives you a list of alternatives. My texts are cryptic and short.

    My wife sends messages that are fairly detailed, use pretty acceptable grammar and spell things out with just the numbers.

     

    reply to this | link to this | view in thread ]

  21.  
    identicon
    westcliffer, Oct 9th, 2006 @ 2:13pm

    Mrs. Peacock

    Poppycock!!!! Possibly could narrow down the suspect list using a long message or interchange, but no way for brief ones...

     

    reply to this | link to this | view in thread ]

  22.  
    identicon
    Sanguine Dream, Oct 10th, 2006 @ 5:32am

    I don't think so...

    I honestly don't think trying this method should be adopted as a legitimate forensic technique. Short of sig at the end of the message how can you really tell who typed up said message? And even then someone could have just copied it to set them up.

    Wasn't it School Ties where someone cheated and they typed the notes instead of handwriting to avoid giving themselves away?

     

    reply to this | link to this | view in thread ]

  23.  
    identicon
    jonny, Nov 2nd, 2006 @ 4:07pm

    jonny

    jonny

     

    reply to this | link to this | view in thread ]

  24.  
    identicon
    JPillband, Nov 8th, 2006 @ 10:48pm

    Hi all, I am new here

    Hello to all. I am John Pillband, 22 y.o. This is my first day on the Internet. What should I do?

     

    reply to this | link to this | view in thread ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>
Follow Techdirt
A word from our sponsors...
Essential Reading
Techdirt Reading List
Techdirt Insider Chat
A word from our sponsors...
Recent Stories
A word from our sponsors...

Close

Email This