Hunh. It’s easier to sniff out thumb drives than drugs, I guess?
Lex is a drug-sniffing police dog. His owner trained Lex by giving him a treat every time he alerted, whether or not Lex was right. Is that a good way to train drug-sniffing dogs? Maybe not for innocent people who get stripped searched when they are falsely identified as drug carriers, but it’s great for police departments that use the dogs to enrich themselves with civil asset forfeitures.
(https://boingboing.net/2015/08/07/drug-sniffing-dogs-are-barely.html)
Weird.
so… was there anything on it? Seems like an incomplete story…
Any idea what the probable chemical being detected is?
“Stuff with the perp’s horrible human-skin-oil-crud on it” would be the obvious choice; but also too broad to be of much use as a search aid.
I imagine that both FR4 board and the basically-invincible-black-epoxy-of-encapsulation probably outgas a small amount(probably of something horrid); but that would cover basically any and all electronic devices, storage and otherwise; the packaging of NAND isn’t notably different from that of any other IC. Plus, if there is an active PC in the area(and where isn’t there?) that’s a big box full of ICs and circuit boards, being gently cooked to release any volatiles, with fans blowing them out of the case and all over the place, which one would think would drown out the emissions of a tiny, room temperature, electronic device.
Something I’m missing? Any ideas?
Nope, it’s just more profitable to have drug sniffing dogs that don’t actually sniff out drugs (whereas you can’t get yummy civil asset forfeitures simply by sniffing out electronics - they aren’t illegal on their own). As per the article that you linked:
My bad for not using /s. Thought sarcasm was obvious.
Of course there was something on the drive! Obviously this dog is trained to only smell drives with illegal data on them. There’s wouldn’t be any point in sniffing out regular drives with good data, would there?
My guess is that the dog is looking for a combination of certain plastics and not others. Possibly combined with traces of human scent. We’d have to check the dog’s training protocol to prune the hypotheses.
I wonder if some sort of conformal coating would be helpful…
We should get somebody to train their dog the same way and have a test rig for countermeasures.
It may have been. I’m on about 3 hours of sleep, so my brain isn’t doing very good braining today.
Would have been a far more interesting headline if you left several words off the end. 
(Underage) man’s best friend.
To crack computer crimes, the 26-year forensics-lab veteran based in Connecticut had to first identify the chemicals associated with electronic-storage devices. Hubball took circuit boards, hard disks and flash drives of computers and tested each component. He narrowed the analysis down to a single common chemical, which police declined to specify or describe.
If you believe what the police decide to report.
This is an obvious cover for parallel construction. Dogs are actually not much better than humans at smelling scents … their heads are just closer to the source.
Assuming there’s some truth in it, I bet the secret is already out there. Most likely hiding in the treatises about outgassing from electronics materials. The chemical will be most likely something of low-ish molecular weight, used as an additive. Most likely from the packaging of chips (not many options there, mostly epoxies filled with glass) or the circuitboards (again, few options, given that the higher-end electronics does not use the pressurized crap cardboard but a real glass fiber in epoxy matrix).
That will of course not be specific to digital storage media. But that can be coupled with other factors, like physical location, and manual weeding out of the false positives.
Thought: store the media inside the cases of existing common-and-“harmless” electronics, e.g. a radio or a TV.
This is what I was thinking, too.
Sorry to hear that - off the computer with you, and get to snoozing already.
In theory, the police would follow best practices to retrieve what is on the drive without any chance of altering the data on the drive and there by contaminating the evidence. That means not inserting it into any old computer, but rather a device that does not mount or read the device, but instead just makes a bit or hardware unit level duplicate of the storage device locally for investigation.
My guess is there are well trained computer forensics investigators the device would be shipped to for analysis, at least I hope that is the way it works.
Fun fact: scaled for total brain size, the region of a dog’s brain that processes smells is 30 times larger than our own!
Well, if it’s anything like what CSI Miami has taught me about forensics, all they have to do is set it down on their fancy translucent computer table/display/thing, and it’ll just suck all the data right off it wirelessly.
Countermeasure thought. Program the controller to detect sequential block reads (usual filesystem reads are way more quasi-random). Serve garbage or omit certain data.
…but classical encryption may be more useful. Especially if coupled with a way to hide the data out of sight. With today’s drive sizes, including the flash ones, smaller amounts of data can be hidden by steganography in harmless-looking containers.
