Scanadu to shut down support for its Scout device and customers are mad | Hacker News Reader

Scanadu to shut down support for its Scout device and customers are mad

techcrunch.com

105 pointsmatco1159 comments

Property rights are pretty ambiguous when a device purchased by a consumer requires a backend service and/or is managed by an outside company. Yhe consumer has no control and while they might own the hardware they have no guarantee of it's continued utility. Companies claiming DCMA violations against those who choose to alter their devices also limits their effective ownership. I personally avoid buying devices like this in the first place.

I do however think this company should be put out of business as they clearly misled their backers. Is there any information on how many of these wer sold? The device certainly doesn't look like it cost anywhere near $199 to manufacture. And if there enough people interested in hacking the device to do a clean-room reverse-engineering project it looks like it would be a fun little project.

Is anyone who owns one of these willing to loan it out for a while? They might also become cheap on eBay next May.

EDIT: As I've thought about this a bit more, why not push a firmware update for your users that allows the device to collect the information as a fitness device? Surely if a machine in the local mall can take my blood pressure without a technician monitoring it, you can own a scout that provides reasonably accurate measurements?

Yeah, I have the same rule of thumb right now - if it's cloud-dependent, steer clear. Especially if the device should presumably be partially or fully functional off-line, but isn't, because "cloud powah!" (in reality, because of the shitty business model that's basically "Hardware as a Service" + possibly data selling).

> The device certainly doesn't look like it cost anywhere near $199 to manufacture.

I can imagine it costing that much in very small batches, if it uses somewhat decent sensors. I can imagine $199 would be the cost of making a somewhat decent DIY version. I couldn't find any teardown though, and I'm pretty interested what sensors are they using, and if they're any better than the usual crap installed in fitness devices.

I spent the first 20 years of my career as an embedded systems engineer and the most expensive parts were generally the case and power supply. For devices with specialized sensors, they generally dominated the costs. I think you're right - it's the sensors that matter. I don't think $199 will buy you a laboratory grade sensor though.

If we're going to make OSS firmware for this (and what's "firmware" now that everything is FLASH?), creating a tear-down will be the first step. Anyone willing to loan one of these to TeMPOraL or me? If you're capable of doing it yourself, could you create a tear-down?

Info about the hardware is somewhat difficult to track down..

"The device runs the 32-bit version of Micrium RTOS (real time operating system). Under the hood, CNX Software says the layout seems somewhat similar to the Laird Technology BL600 and RFDigital RFD51822 modules, both based on Nordic nRF51822, an SoC with Cortex M0 CPU, 256kB flash, 16kB RAM, and a 2.4GHz Bluetooth low energy transceiver."[1]

[1] http://www.tgdaily.com/hardware-features/72028-video-scanadu...

So a standard BLE platform. But the most interesting information - sensors used - is unfortunately missing.

Not what you were looking for, but indicative that sensors were a key issue.

https://www.scanadu.com/blog/making-headway-with-the-scanadu...

> I don't think $199 will buy you a laboratory grade sensor though.

I don't think so either. I was thinking more about "not the cheapest crap you can source for < $10 on Aliexpress". Some sensors that are even remotely capable of making reliable and accurate measurements in end-user conditions.

Looking at you Glowforge.

Yeah, this is my go-to example of a cloud-dependent product that really shouldn't be.

Meanwhile you can't play super Mario run (at all) without working internet. Go figure.

To people who funded this, why ? There are some established products already on that market, why buy into a crowdfunding campaign when you can buy the same thing from established vendors ? why take the risk ? reminds me all these drone campaigns (like Zano and co), you can go on the internet, buy a drone, and even send it back if it doesn't work properly, if you buy it from Amazon or Ebay. With crowd funding you'll pay then wait years to get your products if it ever comes and by the time it is there you can get the exact same product, even cheaper from honest vendors. I don't understand why people choose to bear all the risk without some form of compensation. But I'd like to understand. I'm not talking about indie video games or an music album that needs funding, but crowd funding tech products in general.

The best is when Chinese copies of your crowdfunded product are available for 1/5th the price before shipping date of crowdfunded product. It's happened numerous times, e.g. with http://get.pressybutton.com IIRC.

So if I want to have a device without doing all the work of getting it built, I can just put together a Kickstarter and then refund everyone once the knockoff is on the market?

Or buy the clones from the copier at the lower price, pass them on to your backers and profit!

What could possibly go wrong!?!

Or - product-ception - create a product, have it ready to sell, Kickstarter it at a higher price; be both sides of the market.

create a product, have it ready to sell, Kickstarter it at a higher price; be both sides of the market.

This is what successful research groups do: get an industrial customer, invent some tech in the process (but you won't have neither time nor desire to do the boring work of academic writing), and when the project is complete, get some funding for the results you already have. Everybody's happy, just don't forget to negotiate rights to retain the basic (not application-specific) technology.

I think one of the cases is when there aren't any equivalent products available from "established vendors". It sometimes makes great sense (e.g. Pebble, their recent shutdown notwithstanding). But to be honest, when I browse tech section on Kickstarter, 90% of products there scream "pointless" or "bullshit".

Many of those products - such as Zano - are too good to be true. People buy them because they promise features or a price-point that doesn't currently exist. Sometimes that works out, and the campaign delivers on its marketing promises. And sometimes, well, you get a Zano. If you're lucky.

I find it amazingly ridiculous that people would pay for things based on basically an ad. That's what a Kickstarter is. How many things are available on Amazon that we wait to see reviews for from legitimate companies? And yet for Kickstarters, people are willing to believe that these ads are completely the truth and it will be exactly what you think it is?

It's like we've gone backwards by 40 years in terms of trusting companies' ads.

There wasn't when it was funded.

What is the FDA's problem with it? I feel like with this sort of thing and the 23andme.com issue that the FDA is being too restrictive about things that really shouldn't even be under their jurisdiction in the first place. For drugs, stringent testing and regulations is great, but how is someone going to harm themselves by knowing things about their own genome or recording things like their own body temperature and heart rate?

The FDA thing is an excuse. The FDA didn't shut down their product, they are lying.

Oh yep you are right. That article seems to be completely misleading or lying about that.

https://twitter.com/walterdebrouwer/status/80910713363551436...

This appears to be the case. From one of the comments:

   Chuck Juhl 
   Contract Consultant/Analyst at C.W. Juhl Consulting

   I warned folks two years ago that Scanadu was a scamadu. 
   It's unlikely that the reason for the FDA action is 
   because Scanadu could not devise a secure method to 
   transmit data to a cell phone. Other consumer medical 
   devices with similar features have FDA approva already, 
   including the CheckMe device. It's more likely that 
   Scamadu was scamming the FDA as well with its faux 
   "usability" study. According to an FOIA request, the FDA 
   has no record of Scanadu submitting a 510k application 
   for the Scout or any other device.

   And as an aside, Scanadu did not make the final cut for 
   the X-Prize either.

Isn't the CheckMe just a vital signs monitor like you would find in any hospital? Ok, maybe it's a bit smaller and does a touch more (ECG). But I don't think it does as much the X-Prize "tricorders" are supposed to.

Point being, there's apparently no indication that the company even asked for FDA approval.

The GP's question is valid, though. Why does the FDA insist on shutting down or restricting services like 23AndMe when the worst case scenario for a service failure is that you get a false positive on a disease risk indicator (for which you're going to have to go to a real doctor to confirm anyway) or you think you're 1/8th Irish when you're actually 1/8th Welsh?

A false disease positive or negative can easily fuck up your life. Lots of follow-up tests are invasive or expensive and take time. The FDA is totally right to require high standards and premarket approval.

No one would give you invasive follow-up tests without getting a second genetic test. Doctors aren't stupid; they know that most positives are false positives.

The FDA's claim to jurisdiction (which I feel is BS) ... is that 23&me were providing "medical advice" or "diagnoses" by saying that you share genomes linked to x percent chance of Y ...

Am I missing something? 23&me were forced to validate their service with the FDA. They did and got passed. So they proved that their medical diagnosis are accurate.

I don't understand the hating on the FDA. It's better than having full-time quackery going on.

Edit: Here's an example. 23&Me tells you you're going to develop Alzheimer's. There's no test for that (before it develops), so you can't go to a doctor to validate it. Beyond the mental pain of living with this for years it's reasonable to assume some people would end their lives rather than be a burden to loved ones. But if they fucked up with their analysis and no one forced them to prove it?

Edit2: Also, exempt 23&me from FDA regulation? Why should your lab be regulated either? This isn't just a slippery-slope argument. I don't understand the argument to exempt 23&me.

Because 23&me are not providing a medical diagnoses. They never were. They are giving you what they calculate as your chances to to carry or possibly have or develop health issues based on their database of other people's dna ... they are showing statistical similarities and never claim to make diagnosis.. What does the FDA have to do with statistical analysis?

To most people, there is no difference between "your chances to to carry or possibly have or develop health issues" and a diagnosis. Sure, I suppose I can see some small technicality that makes them different, but really, the way you react to either is about the same, which makes them about the same.

But on the other side of the coin what if the person made life changes that reduced their chance for Alzheimer's and never succumbed to it? Shouldn't a person have access to that information if they want it? Why should the FDA not allow the gathering of information about your own genome?

The FDA does. 23&Me were just forced to validate the accuracy of their testing with the FDA.

Edit: wording

I'm aware of the legal basis. I mean, morally, why do we put up with the FDA killing services that don't have a direct health risk?

The sheep implicitly trust their government. They do not understand that they are in bed with Big Industry -- in fact they ARE Big Industry b/c of the "revolving door".

So here's the real deal. The FDA has basically made up a rule that a medical device can only do one test (its a little more complicated than that but that's the gist of it). That's why if you look at the X-Prize finalists they aren't really "tricorders" at all. Each test has a separate component. Of course, even that probably won't be good enough and neither of these will see the light of day.

The same thing goes for what happened to 23&Me. The official statement from the FDA was "they were concerned that people would self-diagnose". Heaven forbid people should do such a thing in a free country! But what it really boils down to is the huge $$$ made on tests for individual genes -- each of which, I've read, can cost $1,000+ w/o insurance. If we could just do our whole genome in one fell swoop, think of all the money the industry would loose.

This is exactly why firms that build medical devices need to be fully transparent about what they're actually building in the eyes of the law, and what regulatory hurdles it must overcome before being being able to do the things people are likely to assume it can.

I think people would be fine with shutting down the study - but why remote shut down the devices? That's going too far.

The are passive devices, not active. They don't need much regulatory oversight if you don't use the data for anything medical.

Didn't Facebook get in trouble for violating research principles in their emotional-manipulation experiments, something to do with informed consent? Scanadu would appear to me to have done something similar.

I diligently read all the comments and besides all the good and superbly clear points made I have to say that the strongest is that Scanadu has, for a certain extend, mislead its backers to pay $199 to provide them with big health data, something that they would have paid for using the Mechanical Turk platform. In the end, they clearly scammed myself and other people out of their money and data.

My rule: no cloud service dependent devices >$50.

Exceptions are companies like Apple and maybe Microsoft that are capable of supporting the device for a lifecycle. Obviously Google is out.

$50 seems like an arbitrary number, not a logical chosen barrier. What about your data, and your privacy? Is that worth anything?

You say you have Apple (and maybe Microsoft) as exception but Microsoft was one of the first to join PRISM program. Apple's in it as well. Apple uses Siri. Do you entrust Apple with your voice data? With your Spotlight searches (or Microsoft, as I saw today that was being used, and I know I never set it up to use Bing cause I don't use Bing)? If you're not from the USA, it'd be stupid to depend on US cloud services ever since the Snowden revelations. Before that as well, but then it was just all theory.

Cloud is sometimes invisible and, well, murky (cloudy) as to . Therein lies a danger. A program like Signal does it right, but that is merely communication. I have no idea what the feasible alternative would be to cloud because something like a mobile phone just isn't powerful enough, and has its own set of privacy and security issues on top of that.

$50 is my personal threshold for disposable throwing of money away. $50 is decent dinner out.

I'm from the USA. I don't have any reason to implement an opsec level that would potentially defeat a government inquiry. If there were a targeted effort to get at my stuff, my home lacks the controls to even offer a moderate level of resistance.

I'm more worried about shelling out money for services the vanish.

> I'm from the USA. [...] > $50 is decent dinner out. [...] > I'm more worried about shelling out money for services the vanish.

Fair enough! How far do you go with this? Open source firmware and the like?

> I'm from the USA. I don't have any reason to implement an opsec level that would potentially defeat a government inquiry. If there were a targeted effort to get at my stuff, my home lacks the controls to even offer a moderate level of resistance.

Oh, I have no issue if the secret service of my country targets someone (including me) specifically. My problem is with mass surveillance, dragnets, and silly attempts to find needles in haystacks. That I would try to avoid, try to not support those by voting with my wallet.

Also, I'd like you to consider the following: it isn't only the secret service or the NSA. The Snowden revelations are just the tip of the iceberg, or the wake up call. There's also the insurance company, the headhunter, data leaks because $cloud_provider can't secure their shit, profiling based on your data, etc.

I don't disagree. But the highest risk factors for me, like sharing of health and financial data, are fully out of my control.

GP does not seem to be talking about privacy , rather support and availablity of the service

Why is Google out? Have any of their cloud service devices been put to pasture?

You could consider that Alphabet buying Nest and them shutting down all service was in a way Google stopping operations on that.

But I believe that is mostly related to the unending list of services they have stopped. Code, Reader, or even APIs and libraries they stop without a single warning (cough cough angular v1.x).

Seriously, the list is big. https://en.wikipedia.org/wiki/Category:Discontinued_Google_s...

I have no faith in Google's ability to deliver consumer services that aren't part of the core business and are free.

Facebook is eroding that core business, so I expected more belt-tightening in coming years.

Google Reader, Picasa and many more: https://en.wikipedia.org/wiki/Category:Discontinued_Google_s...

Nest remotely bricked Revolv devices a year and a bit after acquiring the company.

Yeah, at least with Microsoft, you know it Plays For Sure(tm).

I got the email from them. It said "cease to function." And so, I wrote them to ask what that means. They hit me with a generic:

Dear,

Thank you for supporting our crowdfunding campaign for the Scanadu Scout® investigational device on Indiegogo.

When we started developing the Scanadu Scout®, we knew there was a great deal of work ahead of us. The Scanadu Scout® was a first of its kind, and we remain committed to bringing innovative technologies to consumers.

Through this journey, we have collected an enormous amount of data over the past two years, which will inform the development of future products. We are now turning our attention towards future products. We look forward to sharing more on the new opportunities we’ve decided to pursue in the upcoming months.

Regards,

Scanadu Team

www.scanadu.com

Scanadu Inc, 1196 Borregas Avenue, Ste 200, Sunnyvale, CA 94089

And this, of course, is why you have to be very careful before investing in Indigogo/Kickstarter projects. Also, why I won't use a fitbit. Anyone know how to intercept fitbit data?

Couldn't tell they end the trial and release open-source sorftware for the device that people can use at their own risk?

> From the beginning of the campaign, this was an investigational device that was part of a study which has now reached its endpoint with data collection for the study ending in November 2016,” a Scanadu spokesperson told TechCrunch. “FDA regulations require that all investigational studies be brought to closure and their respective devices be deactivated. As a result, we will deactivate the Scanadu Scout® devices by May 15, 2017.

I'd guess it's considered a safety measure - even if it's something innocuous-seeming, any 'medical device' could end up with peoples' lives depending on it. It's better for it to stop working entirely than to appear to keep working if it's not reliable.

Well should it not have been clear in their campaign and communication from day ONE ? Why charge the user that much then? Can i get a refund now ? why con backers to pay for selling their data and $199 on top of that for just 18month "study"?

Basically I think that's dumb.

That sounds harsh, but I think this is over-reach on the part of the FDA and is limiting innovation.

A long as the device doesn't give a diagnostic output I don't see it as a big issue. The potential benefit from personal heath monitoring is huge.

It's the same with 23andme and other consumer genomic tests, let people experiment a bit when it's essentially harmless.

@temporal @smoyer I'll happily lend mine. PM me to coordinate or my username at Gmail. I did just google image search the scout and the version I have must be one of the earliest batches

This is why the right to be forgotten is important. It was a mistake to grant these people your personal data, but at the very least you can make them delete your personal data.

Reply on news.ycombinator.com