Working From Home - Impacts of Coronavirus

[SiliconWizard]

You're very much in the minority, likely less than 0.001% of the population. The number of people who don't own a smartphone is statistically irrelevant, there's no need to develop a standalone device. If you really wanted to volunteer, I'm sure someone would give you an old smartphone that could be used strictly for the tracking app if that was the desire.

Uh, yeah. https://en.wikipedia.org/wiki/List_of_countries_by_smartphone_penetration

[SiliconWizard]

What kind of strong incentive are you talking about exactly?

Your guess is as good as mine. Could range from absolutely nothing to, say, access to a ventilator. Free beer, maybe. Lots of possibilities!

Access to any medical help - they said they would never condition that to the use of the app. Hmm, so far, that is. And I don't think you could, that would certainly be unconstitutional per se in most countries.

I'm not going to guess, I'm just saying it seems completely unrealistic, especially in our western democracies. Unless, of course, they lie about the "voluntary" part.

[PlainName]

Having Bluetooth on constantly is going to kill the battery fast too.

My phone has BT enabled permanently (talks to a smartwatch which samples my heart rate at least every 10 mins, amongst other things). Also has GSM on permanently (though not data). Battery lasts 3/4 days on the phone, 3+weeks on the watch.

OTOH, turn on 5GHz wifi and you can watch the battery drain before your eyes. Lucky if it lasts a day.

[rdl]

I thought I read recently that about 40% of people in the US lacked a smartphone, but a quick check now indicates it's probably only 25% or so.

[PlainName]

Access to any medical help - they said they would never condition that to the use of the app.

Fine. Use your imagination, then. I did include 'absolutely nothing' as a possibility so you don't have to pick the worst option

[james_s]

I thought I read recently that about 40% of people in the US lacked a smartphone, but a quick check now indicates it's probably only 25% or so.

That's actually a bit shocking. I've been trying to think of anyone I personally know who doesn't have one and I can think of one who didn't for a while but now I don't know whether he does or not. Even my 70 year old parents have them and my partner's nearly 90 year old grandparents. I'm amazed that more than 1% of the population over about the age of 8 doesn't have one.

The link someone else posted said it was 22% in 2018, so it's got to be under 20% now two years later. I'm still blown away by this but maybe some parts of the country have much lower penetration? Given I live only a few miles from Microsoft, Google, Amazon and numerous other tech companies it's likely this area has a higher penetration than most. I can't think of the last time I saw anyone using a non-smartphone, it's been years.

[SilverSolder]

Having Bluetooth on constantly is going to kill the battery fast too.

My phone has BT enabled permanently (talks to a smartwatch which samples my heart rate at least every 10 mins, amongst other things). Also has GSM on permanently (though not data). Battery lasts 3/4 days on the phone, 3+weeks on the watch.

OTOH, turn on 5GHz wifi and you can watch the battery drain before your eyes. Lucky if it lasts a day.

I get about 3 days life out of an old Galaxy S5 here (modded with a bigger battery).  Bluetooth does shorten that down a bit, but I haven't measured the effects.

[Bud]

This shit may quickly become weaponized... Say, airlines may deny you boarding the flight if you do not have the app installed.

[SiliconWizard]

I get about 3 days life out of an old Galaxy S5 here (modded with a bigger battery).  Bluetooth does shorten that down a bit, but I haven't measured the effects.

Yeah, any modern Bluetooth chipset will not draw too much power on 'idle'.

But in this case, if you're frequently in areas with many people, Bluetooth communication will happen frequently and draw significantly more, even though it's apparently using BLE, so that's just unfrequent and small packets.

But between BLE and the required OS, that suggests the app will require minimal hardware & OS version to run - which is likely to further limit the % of people with an adequate smartphone.

[SiliconWizard]

This shit may quickly become weaponized... Say, airlines may deny you boarding the flight if you do not have the app installed.

Yup.
It may not be made mandatory from your government, but private companies MAY make this a prerequisite for accessing their services...

[SilverSolder]

I get about 3 days life out of an old Galaxy S5 here (modded with a bigger battery).  Bluetooth does shorten that down a bit, but I haven't measured the effects.

Yeah, any modern Bluetooth chipset will not draw too much power on 'idle'.

But in this case, if you're frequently in areas with many people, Bluetooth communication will happen frequently and draw significantly more, even though it's apparently using BLE, so that's just unfrequent and small packets.

But between BLE and the required OS, that suggests the app will require minimal hardware & OS version to run - which is likely to further limit the % of people with an adequate smartphone.

Is it actually possible to enable BLE separately from classic Bluetooth?  -  if not, maybe it is the classic part that is draining the battery, always listening for new devices coming in range etc.

[SiliconWizard]

I get about 3 days life out of an old Galaxy S5 here (modded with a bigger battery).  Bluetooth does shorten that down a bit, but I haven't measured the effects.

Yeah, any modern Bluetooth chipset will not draw too much power on 'idle'.

But in this case, if you're frequently in areas with many people, Bluetooth communication will happen frequently and draw significantly more, even though it's apparently using BLE, so that's just unfrequent and small packets.

But between BLE and the required OS, that suggests the app will require minimal hardware & OS version to run - which is likely to further limit the % of people with an adequate smartphone.

Is it actually possible to enable BLE separately from classic Bluetooth?

I don't think so. Never seen a smartphone on which you could.

[coppice]

You're very much in the minority, likely less than 0.001% of the population. The number of people who don't own a smartphone is statistically irrelevant, there's no need to develop a standalone device. If you really wanted to volunteer, I'm sure someone would give you an old smartphone that could be used strictly for the tracking app if that was the desire.

Have you considered the large number of mostly older people who don't use a smartphone, because they struggle to read them? There are still phone models with big chunky keys and large digits on the display for people with poor sight that sell in considerable quantities. Those older people are a key demographic that most needs monitoring.

[Someone]

Another question: how can the app know who is infected? I suppose it entirely relies on each invidual voluntarily declaring this through the app? How reliable would that be really, knowing we are definitely not nearly as obedient/compliant as korean or chinese people?

This is one of the reasons local governments/health authorities want more control over any tracing system, so they can have some control over what is marked for tracing.

But that level of control/inspection is only needed for the database of "confirmed" cases. The google/apple model provides the other 90% of the work, but they require the users to have control over when/if they release any data, and what they do with the information. The big picture view of most of these systems:

a) People walk around and share randomly generated identifier numbers with others near them (how often each persons random identifier number changes is a big privacy question)

b) Each device keeps its personal database of which identifiers it has seen (if you include when/where metadata as well then privacy people get unhappy)

the next step is where this all diverges:

c) Government mandates all data is delivered to them for offical uses, and they have the de-identifying data to link any record back to specific people. They promise this is better as they will actively contact and follow up with people who might be infected.
or
c) Users decide if/when they announce they were infected, and upload the history of their recent identifiers to a central store (could be region/locale specific). Other users periodically pull the official list of confirmed identifiers and check if they have recoded contacts in their local data. This relies on the users deciding on what they want to do with the information, some people might not announce they were infected, and others might take no action even if they had a contact match.

[james_s]

Have you considered the large number of mostly older people who don't use a smartphone, because they struggle to read them? There are still phone models with big chunky keys and large digits on the display for people with poor sight that sell in considerable quantities. Those older people are a key demographic that most needs monitoring.

I don't know these older people.

As I already said, my parents are in their 70s, I have older relatives in their 80s, my partner's grandparents are late 80s-early 90s, a group of their friends we get together with now and then, all of them have smartphones. I genuinely cannot think of anyone I I know, of any age over about 8 who doesn't have one. Smartphones have high contrast settings, zoom and other features specifically for people with poor eyesight and other disabilities.

Obviously there are a lot more out there than I realized, I just don't personally know any of them and have never seen them so I have to guess it's more common somewhere else.

[nctnico]

I understand this as Bluetooth tracking compared with GPS tracking. Converting a Bluetooth UUID to an actual person's ID or phone number requires access to private information and your contact list, and who trusts these cheezy app developers to not harvest it all and sell it to third parties ala zuckerburg style and claim it was a hack.

But Corona tracking apps don't work that way. The Bluetooth ID gets hashed (which means converted to a unique fingerprint which cannot be traced back to the original; see SHA256 for example). On your phone you'll have a list with those fingerprints for a couple of weeks. A central server receives messages from phones of which the user says he/she is infected and you get a message saying a fingerprint is infected.

In short: there is no absolute location tracking involved and no personal c.q. traceable information is shared. Also many Corona tracking apps are open source so anyone can see how it works. As bd139 wrote: you should really do some investigation into how Corona tracking apps actually work instead of writing utter nonsense.

How does the hashed Bluetooth ID or beacon get resolved back to the original person for notification?

That is not necessary. Your phone can query a server for infected hashes and if one of these hashes matches your list with contacted hashes then you where near an infected person.

c) Users decide if/when they announce they were infected, and upload the history of their recent identifiers to a central store (could be region/locale specific). Other users periodically pull the official list of confirmed identifiers and check if they have recoded contacts in their local data. This relies on the users deciding on what they want to do with the information, some people might not announce they were infected, and others might take no action even if they had a contact match.

Probably. But the less people use a tracking app properly, the more strict lock-down measures are needed. So in the end it depends on the people whether they prefer their freedom or prefer acting like a kid.

[SiliconWizard]

I understand this as Bluetooth tracking compared with GPS tracking. Converting a Bluetooth UUID to an actual person's ID or phone number requires access to private information and your contact list, and who trusts these cheezy app developers to not harvest it all and sell it to third parties ala zuckerburg style and claim it was a hack.

But Corona tracking apps don't work that way. The Bluetooth ID gets hashed (which means converted to a unique fingerprint which cannot be traced back to the original; see SHA256 for example). On your phone you'll have a list with those fingerprints for a couple of weeks. A central server receives messages from phones of which the user says he/she is infected and you get a message saying a fingerprint is infected.

In short: there is no absolute location tracking involved and no personal c.q. traceable information is shared. Also many Corona tracking apps are open source so anyone can see how it works. As bd139 wrote: you should really do some investigation into how Corona tracking apps actually work instead of writing utter nonsense.

How does the hashed Bluetooth ID or beacon get resolved back to the original person for notification?

That is not necessary. Your phone can query a server for infected hashes and if one of these hashes matches your list with contacted hashes then you where near an infected person.

The only thing is... as long as the phone queries a server, the server will get your phone's IP address... right? And can associate it with the queried hashes. And from your IP it should be possible to know who's the phone owner (but of course that would require the phone provider to cooperate.) Just a thought about it being completely anonymous.

[PlainName]

And can associate it with the queried hashes.

No. The phone sends nothing. The server sends the hash of the infected phone, your phone just uses that to see if it matches any hash stored only on the phone. All the server can tell is that you have the app installed, nothing else (until or unless you get infected and elect to let people know that).

[Electro Detective]

Please pardon this brief intrusion folks     

I can see why everyone is just rolling with this on faith and trust, it's a lot easier on the brain isn't it?
hoping it will blow away soon and convinced I'm full of it

I think you've made your point sufficiently clear now. Looking at your recent posts I think you have not brought up a new argument or fact in at least a week if not two (I didn't bother looking back further).

What about keeping your mouth shut for a while instead of constantly trying to derail yet another thread?

Dear thinkfat, I missed that you are the OP, and apologize if I've derailed your thread,

and for not creating a new argument and bin the old one, to appease you
 
stay tuned mate, working on something new now.. 

[EEVblog]

This shit may quickly become weaponized... Say, airlines may deny you boarding the flight if you do not have the app installed.

The free market will ultimately take care of that. But they already know who sat next to who, so no need for that.
And government and businesses are slowly losing the ability to capitalise on people's fear, all this stuff isn't going to last too much longer, people's patience is already starting to wear thin. Imagine what it'll be like in 6-12 months time time trying to push this stuff.

[Someone]

The only thing is... as long as the phone queries a server, the server will get your phone's IP address... right? And can associate it with the queried hashes. And from your IP it should be possible to know who's the phone owner (but of course that would require the phone provider to cooperate.) Just a thought about it being completely anonymous.

No. The phone sends nothing. The server sends the hash of the infected phone, your phone just uses that to see if it matches any hash stored only on the phone. All the server can tell is that you have the app installed, nothing else (until or unless you get infected and elect to let people know that).

SiliconWizard,
In the most private option being suggested (please read the explanation I provided above) the centralised store is sending out the lists of identifiers associated with infections to everyone equally, an excellent privacy standpoint as there is no difference between any of the users doing this part. The only identification would be choosing to "announce" as infected and uploading your recent identifiers, which indeed could be linked to the user/location back through the network connections. It would be possible to improve that step but I've not seen any proposals for an air gap/sneaker net in that part.

But the key difference in ideologies here is the being forced to do announce or not, particularly being forced to announce links constantly when you aren't infected.

[nctnico]

Actually the list with infected hashes can also be distributed using a decentralised system like in a peer-to-peer network. That way it is very hard to trace an 'I'm infected' hash back to it's origin.

[nctnico]

This shit may quickly become weaponized... Say, airlines may deny you boarding the flight if you do not have the app installed.

The free market will ultimately take care of that. But they already know who sat next to who, so no need for that.

At that point it is too late. There is a delay between getting infected and becoming contagious. One of the points of using an app is to get an announcement you have been in touch with an infected person a couple of days ago. Maybe the level of contacts has to be more than one level deep though to get an early warning (but that is something to figure out by virologists). Anyway, with an app we can literally get ahead of the virus.

[Someone]

Actually the list with infected hashes can also be distributed using a decentralised system like in a peer-to-peer network. That way it is very hard to trace an 'I'm infected' hash back to it's origin.

The verification still poses a bottleneck, anonymising test results etc.

The main point is to separate out identification and/or tracking of someone who is known to be infected and choosing to co-operate, vs just tracking everyone all the time.

[PlainName]

And government and businesses are slowly losing the ability to capitalise on people's fear, all this stuff isn't going to last too much longer, people's patience is already starting to wear thin.

Maybe. I mean, you're right that people are getting restless and if it all blew over right now we'd be back to where we were pretty soon. But that assumes it will blow over. Suppose we all start going back to work and it blows up again, like it's doing in Singapore - people will start to get the message that it's not an easy fix and that social distancing is actually pretty important. It wouldn't take too long for there to be new norms if we can just stick it out long enough to get over the 'strangeness' speed bump. If you look at, say, Hong Kong where masks are a normal part of life, that essentially came about through SARS, which was only 7 years ago.

There are other examples, both pretty significant (like sewers and flush loos) and not so much. Bidets are apparently all the rage in the US now, which they previously definitely weren't, and where they're being installed I reckon they'll be still be used when the novelty has worn off, and gradually permeate the rest of the country. It could also be a while before one allows one's stock of toilet paper to dwindle to one roll before seeking more.

So there are two new minor norms that could easily continue. It will no doubt massively depend on what happens next, and no-one really knows that at this point.