Perhaps two decades ago I would've thought this was interesting and a good idea, but now stuff like this is scary as it just brings us closer to the ultra-surveillance dystopia.
"Everything we measure about you can and will be used against you."
In response to the COVID-19 pandemic, CDC launched the National Wastewater Surveillance System (NWSS) in September 2020. CDC developed NWSS to coordinate and build the nation’s capacity to track the presence of SARS-CoV-2, the virus that causes COVID-19, in wastewater samples collected across the country.
Which is not particularly personally invasive and is a great tool.
The thing with the cryptic infection in Ohio was maybe personally invasive (but as far as I know, no PII has been revealed): https://www.darkdaily.com/?p=43174
I make no judgement about its invasiveness or usefulness. A possible future is for this monitoring to go further upstream to become more personalized and cross boundaries into other monitoring domains, as other comments noted many pharmaceutical products or metabolites can be detected in human waste.
Possibly something holding practical implementations back is all the other waste put into sewage, which may stop things up in the manner of “fatbergs.” [0]
A related idea I have percolating is to change municipal composting programs from bins and trucks to grind and pipe by deploying more sink fixtures that grind food materials into small pieces so they go thru the drain. Then focus extracting/separating biological materials from other effluvium.
> Wait until we've got smart toilets analyzing our poop and pee everywhere we go.
I'm hacking on a local-only, slurp-every-data-point-about-myself program, and I missed adding a analyzer for feces and urine! Thanks for sharing your sarcasm :)
I work in microbiology lab automation. Setting up automated culturing shouldn't be out of reach with the tech available to ordinary people today, but it will smell pretty bad.
In 2003 I installed a scale in our dorm toilet. When you sat down you push a button and when done you push it again. If you were in the top five you could enter your (three letter max) name using a two button system. Top five were displayed using a super low resolution screen similar to a pinball high score screen.
That may already be a thing (garbage weighing) somewhere. There was a trial where residents were be billed by the weight of their garbage. The trial involved adding QR codes to the bins and an adapted garbage truck that weighed the bin and read the QR code upon disposal. I don't know what happened after the trial (or if they had an incognito mode).
There's places where your tax amount depends on how much trash by weight you produce. I know people in some of those cities that drive their trash to the next town over and avoid paying.
It's currently operated in Dublin. Barcode on back of each of 3 bins (recyclables, non-recyclables and organic). Recyclables and organic are free, and non-recyclables includes a quota of x kg per month included in your monthly fee. You pay extra per kg if you go over the quota for non-recyclables in any month. The system encourages recycling, but at the risk of people contaminating the (free) recyclables bin with non recyclable material.
I distinctly remember seeing a TV personality in France (Jérome Bonaldi, for fellow French people on HN) presenting, in the 90s, Japanese toilets that did that.
20+ years later, I live in Japan and they're nowhere to be found (thankfully)
When consumer-grade EEG devices started popping up a few years back, I was thinking the next step would be that they'd be integrated into headphones. I was assuming it would be a selling point, but never thought it might be a hidden feature for the benefit of the manufacturer/state.
The EEG space hasn't done anything interesting in more than a decade. I almost focused my career in the space when emotiv was first making moves. Out of curiosity, I've popped into a few conferences and they've all been pathetic. Similar to how blockchain was going to be incorporated into everything. A lot of hopeful possibilities wound up as dead ends.
Fifth generation warfare probably doesn't need a mostly useless and high noise-to-signal surveillance method like EEGs.
There's unfortunately only so much you can achieve with scalp-based electrodes. The applications everyone dreams about - like fine BCI or controlling robotic arms - are hard to achieve with signals once they've had to pass through the cranium.
If you see "Google" anywhere, replace with "advertising giant and leading surveillance capitalist" and you can gauge how happy you should be with the new tech development that is presented.
Yes, how soon until your smart/spy/corporate door refuses to let you out, because your smart/spy/corporate phone has determined you're ill, even if you are asymptomatic?
I'd suggest looking up 'heart rate variability'. If the headphones are sensitive enough to measure that, it would be quite possible to build a stress-response profile of you with regards to different content.
> Imagine how much an advertiser would pay for knowing what their advert did to viewers heart rates.
It might be a rhetorical question but do someone in the industry has any idea? It has some value, like most metrics but my suspicion is that it is relatively low.
It is a proxy for a secondary metric. The reason people put up ads is to drive up sales, so that's the important metric. A secondary metric is engagement it can be directly measured by click through, time watching, etc... And then you have heart rate, which may indicate an emotional response, but it can be anything else. For example I am pretty sure that on average the heart rate goes up when TV ads show, but it may just be because of the physical effort of standing up and walking to the bathroom. Which is probably more significant than the emotional response caused by ads most people are desensitized about.
>And then you have heart rate, which may indicate an emotional response, but it can be anything else.
It could be useful for A/B testing, with the heart rate metric being the average value of your overall data.
I wonder if you could also compare WHEN the heart rate increases with the timing of the ad - for example if your ad ends with a funny punchline and your logo, seeing the heart rate spike is a good sign people are laughing at it.
Totally, the trust was broken long ago and solidified when they removed "don't be evil". Same for most tech companies. Mozilla still has its soul, craigslist too, a few others...
You must have gotten a dud because my qc45s last much longer than my 35s. I’ve been using them 8 hours a day for years too. So much in fact I replaced the “unreplaceable” battery in my 35’s.
For simple audio devices, maybe with a hardware revision, but it's unlikely the driver circuitry would be routed back to an analog sensing pin unless you were doing some closed loop feedback stuff
I mean you can literally plug a regular pair of headphones into a microphone port instead of a speaker port, then yell into them, and it'll record your voice.
But yes the Google buds are Bluetooth and use separate microphones to send recorded audio back to the device for voice calls, etc.
"With no extra hardware" seems to be a loose interpretation of the paper. Their solution, according to the paper, involves either a COTS earbud that has been taken to pieces and wired into their signal processing box, or a full custom preproduction earbud with 6 extra microphones and an unreleased signal pipeline.
I tried to disable the cookies through the consent form on this site. But that form is very, very long... Just impossible to disallow cookies in a standard way.
There was a startup years ago, not sure what happened to it, that was using earlobes for realtime blood pressure monitoring which is a "holy grail"
Ultimate accessory for a sports watch would be to have earbuds that also monitor HRV and blood pressure which is pretty much garbage from the risk at high intensity.
If I'm reading this correctly, "audioplethysmography" is more than heart rate. It's a plot of the actual functioning of the heart, and can detect problems with valves, irregular heartbeat, clogged arteries, etc, too.
It might be clearer for them to say that Google could make another device using the same BOM as the earbud. The title made me think that it only required custom software.
It seems a lot like photoplethysmograph (PPG), in that you get a time domain signal of something related to instantaneous capillary blood pressure, and you process that signal to derive heart rate, and sometimes HRV and other things. The APG transducer is a microphone rather than a PPG LED+photodiode optical pickup.
I don't know how far you could take it though. Irregular heartbeat in a resting patient, possibly. Clogged artery not in the measurement path in an active patient - probably not. The further from the heart the less and less it looks like an ABP blood pressure wave. I bet activities like jumping up and down would show some amount of interference from blood sloshing around in APG just like in PPG.
78 comments
[ 4.2 ms ] story [ 132 ms ] thread"Everything we measure about you can and will be used against you."
Then there will truly be nowhere to hide.
https://www.cdc.gov/nwss/wastewater-surveillance.html
The thing with the cryptic infection in Ohio was maybe personally invasive (but as far as I know, no PII has been revealed): https://www.darkdaily.com/?p=43174
Possibly something holding practical implementations back is all the other waste put into sewage, which may stop things up in the manner of “fatbergs.” [0]
A related idea I have percolating is to change municipal composting programs from bins and trucks to grind and pipe by deploying more sink fixtures that grind food materials into small pieces so they go thru the drain. Then focus extracting/separating biological materials from other effluvium.
https://en.wikipedia.org/wiki/Fatberg
https://www.bath.ac.uk/announcements/study-tracks-illicit-dr...
On a slightly less dystopian note, I can't wait for the DefCon talk on spoofing anus prints
I'm hacking on a local-only, slurp-every-data-point-about-myself program, and I missed adding a analyzer for feces and urine! Thanks for sharing your sarcasm :)
Hmm. May I recommend switching to a less...abrasive fiber supplement? You might be able to relax your protocol a smidge if you do.
https://youtu.be/DJklHwoYgBQ
https://supercombodeluxe.com/gmen/
20+ years later, I live in Japan and they're nowhere to be found (thankfully)
Fifth generation warfare probably doesn't need a mostly useless and high noise-to-signal surveillance method like EEGs.
‘Dystopian’ is exactly the right word.
It might be a rhetorical question but do someone in the industry has any idea? It has some value, like most metrics but my suspicion is that it is relatively low.
It is a proxy for a secondary metric. The reason people put up ads is to drive up sales, so that's the important metric. A secondary metric is engagement it can be directly measured by click through, time watching, etc... And then you have heart rate, which may indicate an emotional response, but it can be anything else. For example I am pretty sure that on average the heart rate goes up when TV ads show, but it may just be because of the physical effort of standing up and walking to the bathroom. Which is probably more significant than the emotional response caused by ads most people are desensitized about.
It could be useful for A/B testing, with the heart rate metric being the average value of your overall data.
I wonder if you could also compare WHEN the heart rate increases with the timing of the ad - for example if your ad ends with a funny punchline and your logo, seeing the heart rate spike is a good sign people are laughing at it.
[0]: https://en.wikipedia.org/wiki/All_your_base_are_belong_to_us
...as in it has a microphone.
But yes the Google buds are Bluetooth and use separate microphones to send recorded audio back to the device for voice calls, etc.
I'm not aware of anyone doing that with a single electromagnet and 2 wires.
They are using the ANC speaker mic pair as the source of _acoustic_ information to infer HR from within the earbud.
As in there are no electrode pads or light sensors or some other known method for HR monitor.
I'm concerned that protracted probing may contribute to hearing loss.
Ultimate accessory for a sports watch would be to have earbuds that also monitor HRV and blood pressure which is pretty much garbage from the risk at high intensity.
I don't know how far you could take it though. Irregular heartbeat in a resting patient, possibly. Clogged artery not in the measurement path in an active patient - probably not. The further from the heart the less and less it looks like an ABP blood pressure wave. I bet activities like jumping up and down would show some amount of interference from blood sloshing around in APG just like in PPG.
If you want to look at some data, https://physionet.org/ has some recordings of PPG and ABP and you can see how they differ. (Edit) like this one: https://physionet.org/content/mimic3wdb-matched/1.0/