Forgot your password?
typodupeerror
Technology

DARPA Has $3.2M to Sniff You Out 223

Posted by michael
from the eau-de-criminal dept.
quackking writes "The Army wants to sniff you out. This fedbizopps.gov link to a DARPA pre-RFRQ tells more. 'The Defense Advanced Research Projects Agency (DARPA) Advanced Technology Office (ATO), as part of the Odortype Detection Program, invites proposals to (1) determine whether genetically-determined odortypes can be used to identify specific individuals, and if so (2) to develop the science and enabling technology for detecting and identifying specific individuals by such odortypes. Total program funding for this effort will not exceed $3.2 million in FY 2003. Multiple awards are anticipated. Proposals are due by January 29, 2003.'"
This discussion has been archived. No new comments can be posted.

DARPA Has $3.2M to Sniff You Out

Comments Filter:
  • Intresting stuff (Score:3, Interesting)

    by blitzoid (618964) on Saturday December 14, 2002 @10:52PM (#4889921) Homepage
    But could wearing heavy perfume mask your scent enough to avoid detection? Bah, just stick to good ol' bloodhounds.
    • by Anonymous Coward
      ... the ripe smell of anchovy paste. Gak, cross your legs!
    • by Jeremiah Cornelius (137) on Saturday December 14, 2002 @11:10PM (#4890012) Homepage Journal
      I think the science behind this smells pretty fishy, and the whole idea stinks!
    • Re:Intresting stuff (Score:5, Interesting)

      by Seehund (86897) on Sunday December 15, 2002 @12:05AM (#4890208) Homepage Journal
      No, unless you actually altered or destroyed the detected "scent" molecules, then "masking" your scent with perfume or whatever wouldn't work. That works on organisms with olfactory organs, who can identify a scent as e.g. "banana", but can't tell the difference between minute structural differences between different banana-smelling molecules, if all "banana" molecules bind to the same receptors.

      OTOH I wonder just how useful this would be for identifying individuals with any great certainty. Unlike fingerprints, the genetic sequences of MHCs (major histocompatibility complexes) of two individuals can very well be partially or fully identical (organ transplants wouldn't work otherwise). This is more comparable to identifying -- or grouping -- people by blood typing, and its application would likely not be for e.g. forensic investigations needing certainties approaching 100%. I'm sure it still can have its uses though.

      For us damn foreigners, what's a "pre-RFRQ"?
      • Re:Intresting stuff (Score:4, Informative)

        by Seehund (86897) on Sunday December 15, 2002 @12:14AM (#4890236) Homepage Journal
        (For those who don't RTFA, it is theorised that the genes coding for our MHCs also determine what detectable scent molecules we spread around us.)
      • Re:Intresting stuff (Score:3, Informative)

        by Reziac (43301)
        All that would be needed is a deodorizer that reacts with a relevant scent molecule to turn it into some other chemical compound. Then your personal profile will no longer match your entry in the database.

        Scent masking (such as with ordinary perfume) doesn't work very well against dogs, because an experienced dog will pick up on the secondary or combined scent and follow that instead. Plus they aren't looking for just one particular scent molecule, but whatever combination the target happens to exude.

    • Not heavy perfume, but you could probably fool it with sweat from several individuals, applied to your skin or worn in a slow-release device.

      You could also synthesize a bunch of other substances (MHC scents, etc.) and wear them to overpower your own.

      -- If the government outlaws perfume, only terrorists will smell good --
    • So, DARPA thinks they can develop a way to track us all by sniffing our pheromones. Splashing on bottled pheromones would be more effective at confusing this technology than perfume alone, but there are drawbacks [uncoveror.com] to using bottled pheromones, and it's funny you should mention bloodhounds.
  • by Anonymous Coward
    The government spies on YOU!

    oh shit.. wait a minute...
    • by Anonymous Coward
      You smell government 20km away!!
    • The government spies on YOU!

      More like, in Soviet Russia the government might have spied on you.

      In the USA, post 9/11, the government will spy on you.

      So... what was the objective of this Cold War thing again?

  • by Anonymous Coward on Saturday December 14, 2002 @10:55PM (#4889942)
    And its called a dog
  • Sounds silly? (Score:2, Insightful)

    by Siguy (634325)
    Perhaps someone can enlighten me to where this will be very useful. I just can't even imagine how knowing what a terrorist smells like or whatever it is they're doing is going to be useful -Siguy [blogspot.com]
    • by Subcarrier (262294) on Saturday December 14, 2002 @11:12PM (#4890020)
      A targeted anti-personnel mine comes to mind. Could be useful for taking out enemy commanders. A retreating force could leave these scattered in the bushes. Of course, they'd have to acquire some samples beforehand. Who does Saddam's laundry, by the way?
      • by Cruciform (42896) on Sunday December 15, 2002 @12:34AM (#4890288) Homepage
        You could base it off the local food supply...
        If soldiers were under orders to consume only MRE packs while in a Middle Easter location, then mines could be programmed to detect strong odors like curry, or other oft-used spices from the region.

        It could be used in reverse as well...
        You feed all your soldiers rations with heavy amounts of garlic. Everyone stinks but they can safely run though a minefield of combined motion/odor detectors. The pursuing enemy attempts to follow them in and is blown apart because their lack of key odor doesn't disable the mines.
      • by Teach (29386) <graham AT grahammitchell DOT com> on Sunday December 15, 2002 @12:55AM (#4890341) Homepage

        A targeted anti-personnel mine comes to mind. Could be useful for taking out enemy commanders.

        Yet another example of something cyberpunk author William Gibson envisioned many years ago. Quoting the first four paragraphs of his second novel, Count Zero, published in 1986:

        They set a slamhound on Turner's trail in New Dehli, slotted it to his pheromones and the color of his hair. It caught up to him on a street called Chandni Chauk and came scrambling for his rented BMW through a forest of bare brown legs and pedicab tires. Its core was a kilogram of recrystallized hexogene and flaked TNT.

        He didn't see it coming. The last he saw of India was the pink stucco façade of a place called the Khush-Oil Hotel.

        Because he had a good agent, he had a good contract. Because he had a good contract, he was in Singapore an hour after the explosion. Most of him, anyway. The Dutch surgeon like to joke about that, how an unspecified percentage of Turner hadn't made it out of Palam International in that first flight and had to spend the night there in a shed, in a support vat.

        It took the Dutchman and his team three months to put Turner together again. They cloned a square meter of skin for him, grew it on slabs of collagen and shark-cartilage polysaccharides. They bought eyes and genitals on the open market. The eyes were green.

        About ten years ago, I had a band that was called Slamhound for a little while, until we found out that the name was already taken by an L.A. glam-rock band. Ouch!

      • Combine this with those robot butterflies. Then we can roll all our paranoid assassination devices into one package.
    • Re:Sounds silly? (Score:3, Interesting)

      by Lord Crc (151920)
      One application I saw on Discovery channel, was a machine which looked pretty much like a metal detector ala airport style. You stood sideways in it, a small puf of air was blown on you, and the sensors "snorted" in the air. They where very sensitive to all different sorts of explosives and similar chemicals. Can't recall the exact figures, but it was in the region of if you touched anything the last couple of days, it'll go off.
    • Aromatic Compounds (Score:4, Informative)

      by VoidEngineer (633446) on Sunday December 15, 2002 @12:08AM (#4890218)
      So, odors are generally caused by 'aromatic molecules'. The nose, actually, is a molecular shape sensing device. Knowing what a terrorist smells like is central to knowing what kind of compounds and chemicals they've been working with. Somebody who smells acidic, dusty, and metallic is doing very different kind of work than somebody who smells of of wood/bark, musty, or moldy. The first person may be working with metals and acid etching things, whereas the second person may be a mycologist, and growing fungus. Between the two, the former is more likely to be making a bomb; the later bioweapons.
  • by 403Forbidden (610018) on Saturday December 14, 2002 @10:55PM (#4889944)
    Developing the equipment to identify unique scents would be costly, bulky, and probably easily confused by purfumes and other forms of distraction.

    I say that nature does the best job, use some sort of animal to sniff a trail, or use a better means to identify a person.

    As it is, fingerprints, eye scans, and DNA are much better than smell, and how would you store the signature of a scent in a database?

    "subject has a old-man on crack smell about him."
    • "and how would you store the signature of a scent in a database?"

      Since scents are just chemicals, one could filter out the background scents, and then store a list of the remaining chemicals in a db.
      • Since scents are just chemicals, one could filter out the background scents, and then store a list of the remaining chemicals in a db.

        Dogs do this all the time -- but HOW they do it is beyond me. They're able to seperate different smells way beyond what a human being can. When I try and ID a person by smell I usually pick up their perfume or detergent instead of their BO, which is nice and all, but if they switch product lines I'm hosed. Dogs on the other hand will smell their BO, their perfume, the funk from their socks, and know the difference between them. At least that's what I hear on the Discovery channel. I've never telepathically communicated with a dog to ask them this first hand; so I'm not 100% sure here.

        So, now the trick is to come up with something that can not only measure smell but measure it in a way that it's seperates each of the signatures out into different signals and then IDs them in sme way, shape, or form.

        I really think something like this has been a long-time coming. Dogs have been used to track people and identify substances for a heck of a long time. I don't see any "big brother" issues here either, and I'm usually pretty iffy about that kind of thing. Becaues the actual method has been proven (dogs) so long as it's implemented right it sounds great to me. There -is- a bit of a problem with pin-pointing the source of the smell though, and it would even be possible for somebody to "rub off" their smell onto you and signal a false match, but I'd imagine the odds of the latter is pretty rare. Would make for some interesting check-in procedures at the airport...

        "Did you pack your own luggage? Has anybody asked you to carry anything on board for them? Have any strangers tried rubbing their stinky bodies up against in an attempt to make you smell like a terrorist?"

        Pretty sure if I had a naked middle-eastern man rubbing his body against mine out of nowhere I'd be worried more about getting the fsck away from him than getting ID'ed as a terrorist at the airport.
        • by Monkelectric (546685) <<slashdot> <at> <monkelectric.com>> on Sunday December 15, 2002 @03:16AM (#4890745)
          Dogs do this all the time -- but HOW they do it is beyond me.

          There is a reason for this actually ... if I remember my freshman bio class, in the human brain the nose is "connected" to a center that controls emotion. This is why we are unable to apply any cognitive function to what we smell. Think about the other senses, we have developed complex language to describe them (blue, shiny, transluscent, iridescent... sharp, soft, prickly, sticky, coarse etc, there are also formalized words to describe sounds but I wont go into detail), but with smell there are really only a few words people used to describe a smell, "good" and "bad" most of the time.

          Dogs have alot more of their brain dedicated to their olfactory sensors (and I believe more sensors as well?). I wouldn't be surprised if dogs have abilities we don't like being able to recall a smell like we could recall sounds or images. I always thought it was weird that I could remember every note of "The great gig in the sky" and recall it at will as if I was listening to it, but ... I cant even remember what my last g/fs perfume smelled like unless I smell it.

          Maybe someone who knows more can give us more information

          • There is a reason for this actually ... if I remember my freshman bio class, in the human brain the nose is "connected" to a center that controls emotion. This is why we are unable to apply any cognitive function to what we smell. Think about the other senses, we have developed complex language to describe them (blue, shiny, transluscent, iridescent... sharp, soft, prickly, sticky, coarse etc, there are also formalized words to describe sounds but I wont go into detail), but with smell there are really only a few words people used to describe a smell, "good" and "bad" most of the time.

            I think you're right. I can remember two smells. One being that of blood (don't ask) and another being of a perfume I particularly like. I remeber them by remembering the emotions that come to mind and the events that transpired when I did smell them.

          • This is why we are unable to apply any cognitive function to what we smell.
            Not really true. It's quite possible to apply cognative power to what you smell. I don't think that people think as much about their sense of smell as they do about some other senses, but you can think about it with practice. A good example of this is in cooking; a lot of what we think of as our sense of taste is actually part of our sense of smell. So when you taste something and think "this would be better with a bit more sage in it" you're thinking analytically with your sense of smell.
            I wouldn't be surprised if dogs have abilities we don't like being able to recall a smell like we could recall sounds or images.

            I can do that. Again, I think that it's just something that takes practice and attention. I get that practice partly because I'm interested in smell and partly because it comes in handy in my job; I'm a chemist. I can't remember smells as well as I can remember sights, but I can remember them.

            If you want to learn to recognize smells, you have to put as much effort into smelling them as you do into listening to music. Don't just notice a smell and then ignore it. Stop, close your eyes, and focus on the smell for a while. I find that it's frequently helpful to open my mouth a little bit and inhale through both my mouth and nose. Try to come up with words to describe it. This is hard because, as you point out, we lack good words for describing smells, but putting things into words helps you to remember them. I find myself describing smells in terms of how they're the same or different from other smells. If you start doing that regularly any time you smell something interesting, you'll be very surprised at how much more accute your sense of smell becomes.

    • by Frymaster (171343) on Saturday December 14, 2002 @11:08PM (#4889998) Homepage Journal
      As it is, fingerprints, eye scans, and DNA are much better than smell

      not really. all the abovementioned methods require the participation of the identified person (well, you can lift someone's fingerprints from that wine glass... but to compare them you need to have good ink sheet ones).

      odour can be detected surreptitiously... say when passing through an airplane security gate, and the person can be identified without being aware of it. if someone scans your retina, you'll notice. if they pick up your smell with a hidden sniffer you won't.

      very insidious idea.

      • by Quixote (154172) on Sunday December 15, 2002 @12:18AM (#4890246) Homepage Journal
        not really. all the above mentioned methods require the participation of the identified person Not for eye (iris) scans. Here's an anecdote. 17 years ago, National Geographic published an eye-catching (no pun intended) picture of an Aghan girl in a Pakistani refugee camp. This year, after the fall of the Taliban, the original photographer (Steve McCurry) went back to that region to try and locate her. Well, to make a long story [nationalgeographic.com] short, he found her; but the verification was done using iris scanning technology (story here [webdesk.com]). Interestingly, the company (Iridian) scanned the negative of the original, 17-year-old photo and used that to do the iris matching with the current photo of the girl (woman now). But the point is: the iris was captured from a 17-year-old photo.
        • Well, to be fair it was a 17-year-old medium-format, professsionally archived head shot photo. That is why there was enough information preserved in the negative to make an iris match. That won't work with just any old 35mm pic.

      • by The Tyro (247333) on Sunday December 15, 2002 @02:38AM (#4890649)

        IANAL, etc, etc... not legal advice, blah, blah, blah

        Courts have long held that using odors is not necessarily a violation of your fourth amendment right to be free of unreaonable search and seizure (with a few exceptions).

        The air around your vehicle, luggage, or other "object" is free to sniff, so drug sniffing dogs and explosive jiffy-sniffers are usable without a warrant. Vehicle "stop and sniff" random checkpoints have run into some trouble, but if you've been stopped by a police officer for some reason (traffic offense), and he suspects the presence of drugs, he can call for a dog, no problem. If said K-9 alerts on your car, probable cause to search is established. I believe the case was United States v. Place in the early 1980s.

        Air around your person has been treated a bit differently, since random, agressive sniffing by a dog, without some articulable suspicion, has been considered a "search" by some courts.

        There was a case of high school students being personally sniffed, and found unconstitutional... it was B.C. versus Plumas Unified School District. Here's a link with some info: Newspaper article [gctelegram.com] you can probably google for the whole text of the decision.

        Based on some of the above cases, this might actually BE unconstitutional, since it's a direct sniff of a person, not an object. You can sniff people without individualized suspicion, but the state has to seriously justify it... minimal privacy invasion, and compelling state interest. However, with the current terrorism problem, and simply having to walk through a gate of some kind to be sniffed (minimal invasion of privacy), this might pass constitutional muster. The lawyers will have their work cut out for them with this one.

    • It would be interesting to see how they would implement this.

      The theory on how animals tell people apart has to do with combinations of long-chain carboxylic acids. They have long fatty chains, and the long carbon chains keep the volatility low, but dogs can supposedly detect them. Their noses are quite sensitive for these compounds; much more so than human noses. On the other hand, humans noses are more sensitive for some other compounds... the mercaptans, for instance.

      I just wonder if they can make equipment that will detect such low, low levels of these compounds, and whether there is enough variation (with the limited number of long chain fatty acids) to produce a unique signature among billions of people.

      Unless I'm way off-base and they are going in a totally different direction, I don't see how this is possible.
    • Developing the equipment to identify unique scents would be costly, bulky, and probably easily confused by purfumes and other forms of distraction.

      Given that it doesn't currently exist I don't really see wher eyou get that from.


      I say that nature does the best job, use some sort of animal to sniff a trail, or use a better means to identify a person.


      I think that's the idea here.. re-implement the dog. Dog's can't have RJ-45 jacked into their head to make a peer-to-peer database of smells that they've learned over the years. Computers can.

      Plus, this might be the first OO system in the world that actually uses crap from those silly college course examples.

      class Animal {};

      class Dog inherits Animal {};

      class GermanShephard inherits Dog {};


      As it is, fingerprints, eye scans, and DNA are much better than smell, and how would you store the signature of a scent in a database?


      If that was easily answered DARPA wouldn't be tossing $3.2 million at the problem.
      • I agree with you except for this:

        "If that was easily answered DARPA wouldn't be tossing $3.2 million at the problem."

        From my understanding DARPA will toss $3.2M at most any problem you can come up with as long as they haven't tossed $3.2M at it before... or at least if they have already tossed the $3.2M you need to talk to someone else to get it authorized.

        Once just for kicks I sent an email to them speaking of an amazing new technology I could develop for a low level i/o interface iobs or descrambled "BIOS". That got me refered to another department, I sent the mail to that department (I can't remember what it was, that was about 3yrs ago, but another computer related department). The head of that department sent me a mail back saying he was very interested in this potential breakthrough technology but needed more information about possible uses and some more details of potential implementation. Yes that's right, I was doing nothing more than describing the general function of the BIOS on every IBM PC Clone out there... I didn't send him his more information and let it drop at this point but it was a real eye opener ;)
      • Dog's can't have RJ-45 jacked into their head to make a peer-to-peer database of smells that they've learned over the years.

        Hmmm, that sounds like a dare to me.

        Umm, imagine a Beowulf cluster of these??? Or maybe a 'wulf cluster would be considered a 'pack'.
    • It's called exploration. Advancing the fields of science. We spent billions to get into space, and does that leave us any better off than we already are? Just because you can't think of any practical applications of this, or you don't think it will work, doesn't mean that it shouldn't be tried.

      You did hit on a great use for this though -- a handheld smell detector, that could latch onto a criminal's scent, and track him down. The cool thing about it would be that once it gets the chemical composition of the criminal's smell, it doesn't forget, and it doesn't get distracted by dog poo [imdb.com].
    • No, such equipment would not be bulky at all. In fact, it would probably rather inconspicuous and it could be located in public places. The 'smell sensor' itself would most likely be some type of fabricated microchip. They have already developed such smell sensing chips that can for example, detect the difference between fresh and rotten fish. If it is possible for an animal to be able to track scents, then it is just as possible to develop some sort of automated software to do that.
    • great sig. actually, this sort of thing may be doable. the DARPA web solicitation refers to sampling the environment for a chemical signature derived from a person's Major Histocompatibility Complex (MHC) proteins. these have a genetic basis and when sampling blood can be used to discriminate individuals without examining the DNA which encodes them. obviously, the technology DARPA wants developed will be to sample small amounts of protein from a defined environment to look for an immunological signature of an individual.

      from a distance and without making direct contact with an individual, this type of technology would obviously be more useful to remote monitoring devices than fingerprints, DNA, etc. who knows, possibly in a 2nd or 3rd gen incarnations something like this could be used in a predator type aircraft scanning an area. flying bloodhounds in a sense.

    • Don't forget about its possible use in forensics, like at crime scenes. Even if they don't keep this stuff in a permanent record, they could still use it to contradict or support other evidence. More evidence is always better, unless you're guilty.

      Of course, this would be of varying use depending on the accuracy of the process and the sensitivity. Make it too sensitive, and you'll get the smell of the cop who handled whatever it is you're looking at... and then they can use that as an excuse to explain planted evidence, or evidence when a cop commits a crime.

  • I propose.. (Score:5, Funny)

    by dj28 (212815) on Saturday December 14, 2002 @10:58PM (#4889958)
    That they use RMS as a test subject. Given his potent odor, their prototype equipment will have an easier time functioning.
  • by Anonymous Coward
    who could do this. He could pick out ethnic types with almost nearly complete accuracy. Then Howard got bored with him and brought out a guy who could fart a whole lot.
    • I saw a challenge show in the UK- they had challenges for people who claimed to be able to do amazing stuff.

      There was a guy who claimed he could identify stamps by tasting them. He really could! He had difficulty with one stamp, but that was identified by actually chewing and _eating_ it!

      Stamps! It was mostly guys.

      I suppose when women have these abilities/obsessions/compulsions/hobbies/interests they go see a psychiatrist. Whereas guys go on TV and brag about it. Or form a Special Interest Group with other likeminded guys. ;).
  • Hold on! (Score:5, Funny)

    by jaymzter (452402) on Saturday December 14, 2002 @11:00PM (#4889963) Homepage
    Ok, you have my name, social security number, IP address, you want to decide how I use MY computer, you take pictures of me when I go to sporting events, you want to cache my surfing habits, sniff my e-mail, and NOW you want to know what my ass smells like??

    Two Letters: FO!

    Oh, and by the way, All your funk belong to us!
  • STASI (Score:4, Insightful)

    by micaiah (593598) on Saturday December 14, 2002 @11:00PM (#4889965)
    This kind of reminds me of the East German's intelligence program of keeping people's scents on file. Maybe that will be next?
  • Okay, this is getting rediculous...

    *Insert ATM card* *place armpit next to machine* *make cash withdrawl*

    We already have Retinal Scans, Voice Scans, DNA Scans, Fingerprint Scans, and Heat Signature scans, who knows what else they've come up with, why the hell would someone invest this much money into something virtually useless. Instead of lineups we just going to brush body odor laiden cotton swabs in front of witnesses face?

    A HA! that's the stinky perp!! I wouldn't forget that smell anywhere!!

    Plus wouldn't this be extremely easy to fake, the nose is one of the weakest sensory organs alone for a reason, it's a additional sensor that aides other senses, mainly taste. Anyone who wants to argue with me, fine argue, but I know that compared to the sense of touch, sight, and hearing, smell is one of the more non-essential senses. And I know this isn't a nose, but the nose being non-important might be a clue to not spend 3.2 million dollars.

    I got an idea, lets quit saying how much social security and federal aid are hurting and divert funds for researching CRAP (pun intended) to the people!

    • I know that compared to the sense of touch, sight, and hearing, smell is one of the more non-essential senses.

      Tell that to a bloodhound.

  • It's a PLOT (Score:5, Funny)

    by K8Fan (37875) on Saturday December 14, 2002 @11:03PM (#4889976) Journal

    This is just another sneaky government plot, this one to get geeks to bathe!

    • by SuperDuG (134989) <be@NOSPam.eclec.tk> on Saturday December 14, 2002 @11:09PM (#4890008) Homepage Journal
      This is just another sneaky government plot, this one to get geeks to bathe!

      You're wrong, and here's why: This is not a plot to get geeks to bathe it's a plot to encouorage geeks NOT to bathe.

      Take into example, the government knows it can spot a geek rather easily on the streets (reference thinkgeek/linux/sci-fi attire, no real hairstyle, and complete lack of self-control), however an average bum holds these same qualities. If you were able to have one deciding factor to divide the geeks from the bums it would be the shower factor. So geeks, protect yourselves, damn the man, and DO NOT SHOWER.

    • Eureka! (Score:2, Funny)

      Maybe we can use this technology for good, and we'll finally be able to tell the difference between the Trekkers from the Trekkies.

      Dr. Who fans, you're next.
      • Shit! All these years, living as a Trek fan, hiding the fact I'm really a Doctor Who fan... well, I may as well just come out now!

        No toasters, please!
  • Uh oh. (Score:5, Funny)

    by MisterFancypants (615129) on Saturday December 14, 2002 @11:04PM (#4889983)
    Could the robotic hounds be far behind? Run, Montag, run!
  • Taco Bell will become illegal!
  • Not possible (Score:2, Informative)


    Body odours are, as the proposal points out, the result of carboxylic acids.

    Although the term carboxylic acid covers very many molecules - basically anything with a HO-C=O somewhere on it, the molecule has to be volatile to have a smell. The problem is that not many acids are volatile - the very composition of the molecule means it makes hydrogen bonds with others easily, and even light acids are involatile liquids or solids.

    This means there is a small pool of molecules to pick from so the chances of an individual having a unique blend is very small.
    • yet, dogs can(proven well enough) smell who of dozen guys touched a gun that was then presented to the dog.(it takes a little training though, i think the point is to create a machine to do this fast, automatically & etc)

      so there are differences, it might be ratios of those substances or you might be just trolling.
      • Coz I'm sure the dog smells a LOT more on the gun than just those dozen guys.

        After all, dogs can detect cancer in people, and detect mines just from shipped air samples stored in test tubes. Trained dogs can even detect lung cancer from air samples.

        So my guess it's more a communication problem.

        It's probably like calling a wine expert over the phone and asking them to identify which of 12 unlabeled identical bottles of wine you sent to him come from the same vineyard, when 1) you don't speak the same language. 2) you don't even know the names of any regions. 3) Your only knowledge of wine is it's fermented grape juice. 4) You were born with no sense of smell.

        Sure it can be done. Easier if you had done similar work together before. But it will still take time, even for slightly different scenarios.

        If the DARPA finds a way to improve interspecies communication that could help.

        Most dogs probably never have to ask "How are you?", they can smell the answer just a few feet away.

        Of course there are dumb ones who might be able to smell all that, but never remember/figure out what it all means. Still, they probably never needed to learn about their sense of smell - just need to know "walkies, food, sit and stay".
    • Will take a fair bit of doing for machines. Not sure if they'd be able to do it with 3.2 million without dogs.

      Dogs can identify individuals. And they can even distinguish twins from each other, if they are exposed to both trails simultaneously, if it's one at a time, it's about the same as us having difficulty identifying twins by sight on different occasions.

      And they can even detect _cancer_ in individuals.

      http://www.canoe.ca/HealthNews/980910_jones.html
      http://abcnews.go.com/sections/scitech/DailyNews /d ogs020611.html

      For more do a search on: dogs smell cancer
  • East Germany (Score:5, Interesting)

    by hrieke (126185) on Saturday December 14, 2002 @11:08PM (#4890006) Homepage
    The Statiz (sp? E. German Secret Police) did something like this once. They would take samples of everything and place it in sealed jars so if they needed to track you with the hounds later, they could in theory open the jar with a sample of your sofa in it and let the dogs loose.

    Funny thing was that it didn't work.
    • Re:East Germany (Score:2, Interesting)

      by ToteAdler (631239)
      The Stasi [cnn.com] actually just had to swipe something you touched with a special cloth and then put it in a jar. If I remeber correctly for some reason (maybe a chemical added) being in the jar amplified the smell so even a human could distinctly tell the difference between them. Actually the Stasi's main problem was that they collected too much information. They had data on almost every citizen and they weren't able to process it all to determine who was and who wasn't doing things they wern't suppose to. It seems like our government maybe headed down the same path but with the help of computers and centralized DB maybe they'll get it right...
  • The premise was used in a scifi story... sorry I can't recall the book but basically a smartbomb was let loose to track the target by genetically-determined odortype... I don't think perfume can adequately cover the primary body odour.
  • Why (Score:4, Insightful)

    by Z0mb1eman (629653) on Saturday December 14, 2002 @11:20PM (#4890051) Homepage
    My guess would be that this would be useful because scent, unlike appearance, is harder to alter. A wanted criminal can just put on some different clothes, maybe grow a beard, etc, and he won't be easily recognized - wear shades, a hat, and he won't be very easy to recognize with any sort of automated system. Other methods of identification - fingerprints or retinal scan - are difficult to apply without the target noticing (and cooperating). I could see machines at airports or bus terminals that "sniff out" anyone who passes by, and if the smell matches with any in its database, bingo... IF the technology works, it could be far more reliable than current methods.

    Of course, all this hinges on the idea that slapping on some cheap cologne won't confuse the machine. And I won't go into the privacy/1984/control etc. arguments...
    • Ok, then all I need to do as a fugitive is lay low and get new clothes at the thrift store every week. As an additional confound, I can donate all my clothes every few months just to make me harder to pin down.
    • by Parsec (1702)

      cheap cologne

      Or hunter's deer urine, or playing with your friend's pets, or walking through a cow pasture, or having sex (with another person), or any medication or other injected/swallowed chemical which may disturb your hormonal balance...

    • Re:Why (Score:4, Informative)

      by Reziac (43301) on Sunday December 15, 2002 @12:28AM (#4890264) Homepage Journal
      Actually, what you eat/drink/smoke alters your body chemistry, thus the waste products exuded by your skin. Which in turn alters your scent. Hell, even my human nose can pick up that much. And dogs can readily pick out people who eat certain foods (such as Mexican food).

      Tho if this becomes practical, I foresee a thriving market in whole-body deodorant washes. It won't fool a really good dog (probably because they're sensitive to a whole spectrum of scents) but I'd bet it would fool a sensor-and-database arrangement, which perforce would be more limited in its sensory range.

      BTW, German Shepherd Dogs have poor noses compared to Labrador Retrievers, and Labs train up easier for this sort of work.

      • True, tho my nose is currently screwed due to eating/drinking/smoking, as you mentioned. For warlike purposes as needed by the Army/DARPA I would actually prefer Boxers (muscle) or Great Danes (size), in the belief that dogs noses are so much superior to ours in general that the specific breed is almost a non-issue. IMHO, technological measures are screwed beyond a certain point, compared to biological measures. Including dogs.

        And yes, I have a soft spot in my heart for Boxers and Danes, so this post is entirely non-objective.
        • Re:Why (Score:4, Interesting)

          by Reziac (43301) on Sunday December 15, 2002 @01:28AM (#4890458) Homepage Journal
          Actually, not all dogs have a nose superior to a human. Some toy breeds have a nose that is demonstrably inferior to an average human nose!! In my observation (as a professional dog trainer) Boxers and Danes are about on a par with most pet breeds in the nose dept., but seriously sub-average compared to hunting breeds that have been selected for scenting ability.

          A fairly consistent clue to how good a dog's nose is, is to watch how long it takes the dog to identify a scent. The better the nose, the faster it happens -- the best noses ID a scent in passing and don't even slow down to do it. A dog that sniffs and sniffs before finally deciding what it's smelling, is a dog that has a poor enough nose that it can't quite make out the scent. Kinda like a nearsighted human trying to read a sign that's just a little too far away. :)

  • I see a future where there are flying insect nanobots that attack and inject Vx nerve agents based on genetic odor patterns. It sounds very racist to me.
  • But seriously, people do smell differently, and I don't think it just from the food they eat. I have a poor sense of smell but I seem to be sensitive to certain types of odors, by chance all of them kinda gross-maybe it's because I grew up on a farm, who knows?

    People who are related genetically seem to have have similar body ordors, so it potentially could be used to distinguish individuals.
  • by Anonymous Coward
    Crazy Muslims never shower and STINK!!!
  • some of my daughter's stinky flatulence to aid in the research. she drinks soy milk and eats soy cheese (milk allergies are fun to plan around) and she *definitely* has an odor all her own. anyone got any ideas on the best way to pass along some samples to DARPA? i'd hate to choose a container that failed to retain the potency :-D
  • by Anonymous Coward
    YOU smell the government. And boy, does it ever stink...
  • Imagine if you will... individuals trained to so control their body odour that they can produce on demand odours triggering fear, love, hate, submission, domination... the boardroom would never be the same and perfume would be antiquated. Join the Army become a B.O. warrior.
  • So what? (Score:2, Interesting)

    by Tuxinatorium (463682)
    As long as this technology is only used to enforce good laws (i.e., against murder) then there's no problem. They're not going to start getting you for victimless crimes (except drugs and software piracy, maybe). So you have nothing to fear.
  • by Ilan Volow (539597) on Saturday December 14, 2002 @11:46PM (#4890147) Homepage
    And suddenly the large stockpiles of Old Spice found in Afghan caves seemed a little less ridiculous.
  • by Critical_ (25211) on Saturday December 14, 2002 @11:51PM (#4890163) Homepage
    ...of science articles. It is very unlikely that humans have a genetically determined smell. In 1992, Blaustein & Waldman did an experiment on tadpoles to see if they could recognize their kin based on scent. The reasoning behind it was to see if they could in fact be breeding collectively to increase indirect fitness. Out of the 12 species tested, they found that 8 showed a kin bias while 4 didn't. Three species favored full siblings over half siblings, three favored half sibliings over non-siblings, and one favored maternal over paternal siblings. Was it Kin recognition? No. Why? Well there was variable expression of this favoring within species and satistically it wasn't favored at all. In other words, it was an experimetanl artifact.

    In 1990, Pfennig et al. repeated the experiment but fed different groups different diets. So non-kin got the same diet and kin recieved different diets. The result? Tadpoles stayed around those that ate the same material because they smelled the same. So it depended on diet rather than a genetic signature. However, further experiments showed that outside of nature, if the environment was completely identical then they could do some rough recognition but this condition never exists in the real world.

    I have huge doubts the government will find a connection here. Before someone says that babies recognize the smell of their mothers, I want to say that is a common myth. Babies recognize the heart beat of their mothers and nothing more. What a waste of time and money.
    • Well, that's tadpoles.. but here's my experience as a dog breeder (33 years and an average of around 30 individuals in my kennel):

      For 12 years I had two essentially unrelated bloodlines. And I found that I could not keep dogs from the two different lines together or they'd fight (even when they didn't fight with their own relatives). Even those of the different lines that were raised together as pups would take a dislike to one another as adults. When I got some dogs from a third unrelated line, BOTH the other lines tried their best to kill them!! OTOH, they tended not to care one way or the other about newly-introduced dogs of other *breeds*. So it wasn't just a "new dog in town" problem.

      Second, I've observed that in general, given a choice, dogs will mate with a closely-related dog in preference to one that's not related (whether they know the individual dog or not). With some stud dogs, it's such a marked preference that they aren't at all interested in breeding unrelated bitches -- and act like they don't "smell right".

      BTW, contrary to common perception, newborn puppies recognise their dam (or other source of food, if bottle-fed) by touch first, scent somewhat later, and heartbeat *never*.

  • Jeez, get a dog!

    Seriously, when I read their specification for a device to enhance soldier performance (Silent engine, can run for hours without refuelin/recharging, will let a soldier carry extra gear, run faster and longer, jump higher and longer), I thought "its called a horse!".

    I bet their final product won't even go fetch...

  • by Anonymous Coward
    Considering that the article is partially humorous, and the icon for humor is a "stinky" foot, I would consider it more appropriate. ;-)
  • just not in such a high tech way.
    When the Berlin wall fell and the STASI archives were opened they found zillions of sealed jars of odor samples taken from "suspects" IE citizens..

    How quickly people forget.....
  • Cant you just use allot of cheap cologne or some other kind of odor masker to disguise your stench?
  • Only me (Score:3, Insightful)

    by karmawarrior (311177) on Sunday December 15, 2002 @12:53AM (#4890333) Journal
    If you're concerned about the potential consequences to privacy and freedom this type of technology might entail, there's really only one thing you can do: Make your government aware of your misgivings. It's YOUR government damn it. You may have decided to let it run itself these last few years, but ultimately the founding fathers made sure that the government would be, in some way, answerable to you - be that, arguably as originally intended, on a State by State level, or, as it is now, on a more pluralist democratic level (yes, as long as the legislature is answerable to the populace, it's a democracy. You don't need more than that, all this BS about rule by plebicite is just that: BS)

    Your government throws money at all types of security "solutions" right now because it believes that is what you want it to do. It believes that, given the events of the last 14 months, you are frightened enough to break Franklin's famous principle about trading freedoms for security. It will do anything to make you feel safer, not only by making you safer, but by throwing tax payer dollars at pointless and socially dangerous projects such as "odor identification systems", as well as more infamous projects such as the face scanning technologies used in Tampa that were found to misidentify a large percentage of the population.

    This quagmire of government spending to make you feel safer regardless of the consequences will not disappear by itself. Unless people are prepared to actually act, not just talk about it on Slashdot, nothing will ever get done. Apathy is not an option.

    You can help by getting off your rear and writing to your congressman [house.gov] or senator [senate.gov]. Tell them not to do anything. Tell them that you appreciate the work being done to protect your safety, but if money keeps being thrown at more and more invasive and ultimately pointless security measures you will be forced to use less and less secure and intelligently designed alternatives. Let them know that SMP may make or break whether you can efficiently deploy OpenBSD on your workstations and servers. Explain the concerns you have about freedom, openness, and choice, and how them doing stuff all the time just for the sake of being popular harms all three. Let them know that this is an issue that effects YOU directly, that YOU vote, and that your vote will be influenced, indeed dependent, on whether or not they can summon up the political courage to spend an entire term getting nothing done.

    You CAN make a difference. Don't treat voting as a right, treat it as a duty. Keep informed, keep your political representatives informed on how you feel. And, most importantly of all, vote.

  • everyone's hygiene will no longer be kept a secret and all those hackers who never bathe will have to "come clean"!
  • didn't my dog do this a long time ago, via his genome? For free (as in beer), no less, since his parents evidently did the Wild Thing (TM) for free, also....
  • by Hubert_Shrump (256081) <cobranet@@@gmail...com> on Sunday December 15, 2002 @01:10AM (#4890401) Journal
    Loud noise immediately following is DARPAs collective forehead-slap.

  • by AgentTim3 (447311) on Sunday December 15, 2002 @01:39AM (#4890493) Journal
    Yeah, go for it.

    Smell my breath, you'll ID me as the Olde English, 0.20 percent. Bring it, suckas!

  • similar research (Score:3, Informative)

    by Doppler00 (534739) on Sunday December 15, 2002 @01:39AM (#4890494) Homepage Journal
    Has been done before: [uconn.edu]

    It's probably just a matter of adapting the existing technology. It will be interesting to see if they can distinguish one scent out of say a bunch of people in an airport looking for a known criminal.

  • We are advancing so far in technology we've decided we wish to compete with the dog. We need to spend 3.2M to determine if humans can be identified by scent? I'm fairly sure dogs identify humans by scent already.
  • Only terrorists... (Score:2, Insightful)

    by gmhowell (26755)
    Only terrorists use deodorant.
    Only terrorists use perfume/cologne.
    Only terrorists keep their mouth minty fresh.

    See? ESR was a patriot all along.
  • I can smell those unix guys a mile away - Mt. Dew and Taco Bell...
  • Cross-dressing? Orgies?

    This is going to be a bitch.
  • since they were doing this in vietnam. though then it was just people detection by sensing the chemicals humans emitt.

    i know the technology wasnt so hot back then because they were in a jungle (where everything is alive), but in almost everyother environment it should work just fine.
  • How DARPA operates (Score:3, Interesting)

    by Infonaut (96956) <infonaut@gmail.com> on Sunday December 15, 2002 @02:25PM (#4892544) Homepage Journal

    For a time I worked as a contractor on a program in the DARPA ISO (Information Systems Office). A common misperception about DARPA is that they're bumbling DoD idiots who are always running off chasing impossible goals.

    DARPA was established specifically to go after high-risk, high-payoff technologies. They know that many of their projects will not result in immediate payoff in terms of useable technologies, but they figure that those technologies which do make it will leapfrog two generations ahead of any competing technology.

    That being said, the program methodology at DARPA is oriented toward specific uses of technology. They're not generally interested in creating something just to see if the technology will work.

    People also have the impression that the research and development takes place *at* DARPA (the infamous "clones in the basement" episode of the X-Files springs to mind ;-) . The truth is that the project managers work out of DARPA, but university labs, defense contractors, and other organizations do the actual development work. In many cases, "failed" DARPA projects later lead to working technologies, based on the expertise gained during the original project.

  • Am I the only one who thinks they came up with this idea after watching Homer become a food critic?

    Next they'll fund the revolutionary technology that can "hear pudding".

UNIX was half a billion (500000000) seconds old on Tue Nov 5 00:53:20 1985 GMT (measuring since the time(2) epoch). -- Andy Tannenbaum

Working...