Would you install and use an NHS Covid tracking app?
Poll: Would you install and use an NHS Covid tracking app?
Total Members Polled: 875
Discussion
fblm said:
Why not? I thought that was the whole point of using Bluetooth no?
Technically, limited by beacon length, which could be very short (seconds). Now, if you link to every beacon you ever see, you’re going to get a lot of connections.
If you limit to beacons you see for an hour solid, you’re going to get family and close colleagues (people who you may not need a trace app to know when they’re sick - as discussed above)
Somewhere between those two is a value that produces useful data. The trick is to find it.
I’ve worked with BLE RSSI in the past. Using for anything more than gross measurements is fraught with complexity.
fblm said:
Why not? I thought that was the whole point of using Bluetooth no?
Because they've inserted a timing threshold to limit the false positives. If the app literally worked on the basis of "you've walked near someone who is ill", you'd get a positive every time you left the house to go to work. In a typical day, I probably walk "near" several thousand people - one of them will be ill.
So apparently they are saying "unless you are next to someone for 10 minutes it doesn't count". So on a long train journey, it works. On a bus, where people get on and off every 5 minutes, it doesn't work. It won't work in a lift, unless it is a really, really tall building. It won't work in the queue at Prets, unless the queue is huge.
About the only user community I see this being useful for is salesmen. They are about the only people who really do talk to random strangers on a regular basis. Other than them, most people talk to people they know, or have sufficient contact with to be told that someone is ill.
What is the problem with this not being perfect?
I'll use a parachute analogy. If you were told that your parachute would work 50% of the time, you'd certainly take one with you if you thought your plane would be shot down. After all, 50% is better than 0% if you have to go into combat. But would you go recreational sky diving with the same parachute? Of course not.
If we're going to use this app to enable people to do things differently - then it needs to work. Otherwise people will be given an entirely false sense of security from it. "Look, I travelled across London, and no one was ill". If you are told there is no risk, then you will take bigger risks.
If this was 80 or 90% good, it would be useful. Without a false positive filter, and with a 10 minute time out, I suspect it is less than 20% good for most of the population. I suspect the false positive filter needs location data, which would be killed by the privacy issues. It would be useful to see "you were with someone at "this location" who is infected - oh, that's my office, I'll pay attention. That would be different to "you were somewhere near to someone as you crossed the road" - less interesting.
No on every score.
For civil liberties, it isn't about Bill Gates knowing (or caring) where I've been. It's about the erosion of privacy and the normalisation of government holding vast amounts of information. If we do it to prevent a virus why not to catch a murderer? Infiltrate terrorist networks and radical groups? A fraudster or drug dealer? Trace benefit cheats or untaxed cars?
Even if you think the current government are completely benign and will use this information only for the best of reasons it isn't hard to see the potential for abuse, and the possibility of its scope creeping into areas way beyond coronavirus.
Secondly the question of whether it will be effective. Effective for what exactly? It is already starting to appear that the biggest mistake we have made is trashing our economy and society in a misguided attempt to manage a relatively minor health problem. I don't see any way this will help us tackle that. I suspect it will just swamp authorities with meaningless data. Besides which unless there is 100% take up and the system is consistently used properly it will not be much help.
And lastly the practicalities of it. The cost of HS2 has already doubled before they have started. Smart motorways have been pretty disastrous. Most government schemes seem to triple in cost and end up counter productive. Especially when they're attached to some emotive crisis and politicians attempting to save face when they've messed up.
For civil liberties, it isn't about Bill Gates knowing (or caring) where I've been. It's about the erosion of privacy and the normalisation of government holding vast amounts of information. If we do it to prevent a virus why not to catch a murderer? Infiltrate terrorist networks and radical groups? A fraudster or drug dealer? Trace benefit cheats or untaxed cars?
Even if you think the current government are completely benign and will use this information only for the best of reasons it isn't hard to see the potential for abuse, and the possibility of its scope creeping into areas way beyond coronavirus.
Secondly the question of whether it will be effective. Effective for what exactly? It is already starting to appear that the biggest mistake we have made is trashing our economy and society in a misguided attempt to manage a relatively minor health problem. I don't see any way this will help us tackle that. I suspect it will just swamp authorities with meaningless data. Besides which unless there is 100% take up and the system is consistently used properly it will not be much help.
And lastly the practicalities of it. The cost of HS2 has already doubled before they have started. Smart motorways have been pretty disastrous. Most government schemes seem to triple in cost and end up counter productive. Especially when they're attached to some emotive crisis and politicians attempting to save face when they've messed up.
https://www.theguardian.com/commentisfree/2020/apr...
Interesting article in the Guardian about this. Be warned it's very much in the 'privacy over all else' vein. However it does bring up one nugget of information - that, according to the article, the NHSX app doesn't use the Apple/ Google system.
Which raises two points. If they want it to use Bluetooth, without the Apple/ Google API, it will stop working as soon as the app goes background, as certainly on Apple, and I believe on the latest couple of Android builds, third party apps are prevented from using Bluetooth in the background - hence the French government pressuring Apple to change this.
Secondly, this probably means it's using location data, which is an entirely different kettle of fish.
Interesting article in the Guardian about this. Be warned it's very much in the 'privacy over all else' vein. However it does bring up one nugget of information - that, according to the article, the NHSX app doesn't use the Apple/ Google system.
Which raises two points. If they want it to use Bluetooth, without the Apple/ Google API, it will stop working as soon as the app goes background, as certainly on Apple, and I believe on the latest couple of Android builds, third party apps are prevented from using Bluetooth in the background - hence the French government pressuring Apple to change this.
Secondly, this probably means it's using location data, which is an entirely different kettle of fish.
Yes, the NHSX app predates the Apple/Google API.
There was already talk of being able to de-anonymize the data
https://www.theguardian.com/world/2020/apr/13/nhs-...
Hell no. Not trustworthy.
There was already talk of being able to de-anonymize the data
https://www.theguardian.com/world/2020/apr/13/nhs-...
Hell no. Not trustworthy.
grumbledoak said:
Yes, the NHSX app predates the Apple/Google API.
There was already talk of being able to de-anonymize the data
https://www.theguardian.com/world/2020/apr/13/nhs-...
Hell no. Not trustworthy.
Yes - but there was talk when the Apple/ Google API launched that they were going to incorporate it. Clearly they've decided against it...! Probably too much of a re-write at too later stage in it's development to be viable without delays, but I suspect it'll bite them if the app is clearly a privacy risk.There was already talk of being able to de-anonymize the data
https://www.theguardian.com/world/2020/apr/13/nhs-...
Hell no. Not trustworthy.
Edit - just read that article, which I'd missed up to now. Yeah, that's not looking great. I suspect they may have an interesting time getting it into the App Store....!
Edited by pip t on Wednesday 22 April 11:57
Edited by pip t on Wednesday 22 April 12:10
I think we very much hold different opinions on this and the level of sensitivity required....I suspect we also may never agree.
t, or, in small percentage of cases you may suffer very serious illness, and you may die. So the 50% chance of catching it only leads to a very small risk of a serious outcome.
At the end of the day, all of the potential methods of mitigating the spread of CV19 are designed to help. Not to 100% solve, simply because we're not going to be able to 100% solve. Methods that help a bit, even if only a bit, all together make keeping it under control more viable.
Even the most privacy invading contact tracing app will only help. And it'll help less if very few people install it because of the invasion of privacy. For me at least, better to have a less effective app, installed by a bigger user base. In the end that'll probably help more than the alternative.
rxe said:
Because they've inserted a timing threshold to limit the false positives.
If the app literally worked on the basis of "you've walked near someone who is ill", you'd get a positive every time you left the house to go to work. In a typical day, I probably walk "near" several thousand people - one of them will be ill.
So apparently they are saying "unless you are next to someone for 10 minutes it doesn't count". So on a long train journey, it works. On a bus, where people get on and off every 5 minutes, it doesn't work. It won't work in a lift, unless it is a really, really tall building. It won't work in the queue at Prets, unless the queue is huge.
The thing is to help, it doesn't need to. Because if you're only with the infected person for a short amount of time, it would appear the transmission risk is low. So no need for an alert. Of course, in a few cases there may be transmission, but these will be a small amount of edge cases. If the app literally worked on the basis of "you've walked near someone who is ill", you'd get a positive every time you left the house to go to work. In a typical day, I probably walk "near" several thousand people - one of them will be ill.
So apparently they are saying "unless you are next to someone for 10 minutes it doesn't count". So on a long train journey, it works. On a bus, where people get on and off every 5 minutes, it doesn't work. It won't work in a lift, unless it is a really, really tall building. It won't work in the queue at Prets, unless the queue is huge.
rxe said:
About the only user community I see this being useful for is salesmen. They are about the only people who really do talk to random strangers on a regular basis. Other than them, most people talk to people they know, or have sufficient contact with to be told that someone is ill.
What is the problem with this not being perfect?
I'll use a parachute analogy. If you were told that your parachute would work 50% of the time, you'd certainly take one with you if you thought your plane would be shot down. After all, 50% is better than 0% if you have to go into combat. But would you go recreational sky diving with the same parachute? Of course not.
Not a great analogy. If your parachute fails to open, it's pretty much certain death or very serious injury. So a 50% chance of your parachute not working is a huge risk. If you contract CV19, you may not notice, you may have very mild symptoms, you may feel a bit poorly, you may feel thoroughly sWhat is the problem with this not being perfect?
I'll use a parachute analogy. If you were told that your parachute would work 50% of the time, you'd certainly take one with you if you thought your plane would be shot down. After all, 50% is better than 0% if you have to go into combat. But would you go recreational sky diving with the same parachute? Of course not.
t, or, in small percentage of cases you may suffer very serious illness, and you may die. So the 50% chance of catching it only leads to a very small risk of a serious outcome.At the end of the day, all of the potential methods of mitigating the spread of CV19 are designed to help. Not to 100% solve, simply because we're not going to be able to 100% solve. Methods that help a bit, even if only a bit, all together make keeping it under control more viable.
Even the most privacy invading contact tracing app will only help. And it'll help less if very few people install it because of the invasion of privacy. For me at least, better to have a less effective app, installed by a bigger user base. In the end that'll probably help more than the alternative.
I am afraid this smacks of state monitoring of my movements. How long before a police officer stops you and demands to see your smartphone, because the law says you have to have a "reasonable excuse" not to be using the app in public?
And then the government will say "it's a vital part of our fight against terrorism, so track and trace stays with us forever".
It'll either not work in any useful way, so we needn't bother with it.
Or it'll give the state greater ability to monitor beyond the virus crisis, so we shouldn't bother with it.
And then the government will say "it's a vital part of our fight against terrorism, so track and trace stays with us forever".
It'll either not work in any useful way, so we needn't bother with it.
Or it'll give the state greater ability to monitor beyond the virus crisis, so we shouldn't bother with it.
BLE beacons will run in the background for any app; that’s kind of the point of them.
Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Mr E said:
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Sorry, I don't quite follow you. Do you mean that citizens will be legally obliged to have the app if they leave their home? And who is "we"?Brave Fart said:
Sorry, I don't quite follow you. Do you mean that citizens will be legally obliged to have the app if they leave their home? And who is "we"?
‘We’ is us as a wider society. I have no idea what the government may or may not decide to do. It very much depends on our approach to risk.
Mr E said:
BLE beacons will run in the background for any app; that’s kind of the point of them.
Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Yes, apologies - both iOS 13 & the later Android OS's can allow background bluetooth, but only with the users explicit permission.Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Which leads to the question, why is France kicking up such a fuss with Apple over the privacy protections.....the people who download the app simply select 'allow' when asked by the OS if the app should have access to Bluetooth, and everything's hunky dory?
Were I to go down a speculation rabbit hole with this (Forgive me, don your tin foil for one moment), the only scenario where this would cause a problem is if downloading the app is mandated by law. In that case it becomes an issue as you could download the app (Yes officer, here it is on my home screen) but then revoke the bluetooth permissions to render it essentially inactive....hmmmmm.
Jonesy23 said:
I don't install random apps from untrusted sources on my phone. So no, I won't be doing this.
Anyway it's a bit worthless if not backed by actual tests for infection and acquired immunity. It's just more junk data to confuse things at this point.
Apple/Play store count as untrusted?Anyway it's a bit worthless if not backed by actual tests for infection and acquired immunity. It's just more junk data to confuse things at this point.
Jonesy23 said:
I don't install random apps from untrusted sources on my phone. So no, I won't be doing this.
Anyway it's a bit worthless if not backed by actual tests for infection and acquired immunity. It's just more junk data to confuse things at this point.
If it's open source and is using the Apple/Google functionality rather than trying to do its own thing I wouldn't put it in the "random apps from untrusted sources" bucket.Anyway it's a bit worthless if not backed by actual tests for infection and acquired immunity. It's just more junk data to confuse things at this point.
pip t said:
Because if you're only with the infected person for a short amount of time, it would appear the transmission risk is low. So no need for an alert. Of course, in a few cases there may be transmission, but these will be a small amount of edge cases.
What scientific basis is there for this threshold? I'll help out: there is none. If there was, we wouldn't need social distancing when walking in the street. No issue with loads of people walking in parks, because you can't catch it. Unfortunately that simply isn't true. If it was true, people would struggle to catch colds on public transport. You catch this thing simply by touching a surface that has had virus shed on it. Not "holding it for 10 minutes".pip t said:
Not a great analogy. If your parachute fails to open, it's pretty much certain death or very serious injury. So a 50% chance of your parachute not working is a huge risk. If you contract CV19, you may not notice, you may have very mild symptoms, you may feel a bit poorly, you may feel thoroughly st, or, in small percentage of cases you may suffer very serious illness, and you may die. So the 50% chance of catching it only leads to a very small risk of a serious outcome.
[quote]
You're missing the point of the analogy. A fully working parachute enables recreational skydiving. A properly working app is (in theory) going to allow people to go back to a more normal life, confident that the app will protect society by highlighting potential infections quickly. If the app doesn't work, people will confidently go back to normal, but the protection the app affords will be a mirage.
To put it really brutally: the app assumes that you can't catch this on a 5 minute tube journey. Do you really believe this to be true?
[quote]
You're missing the point of the analogy. A fully working parachute enables recreational skydiving. A properly working app is (in theory) going to allow people to go back to a more normal life, confident that the app will protect society by highlighting potential infections quickly. If the app doesn't work, people will confidently go back to normal, but the protection the app affords will be a mirage.
To put it really brutally: the app assumes that you can't catch this on a 5 minute tube journey. Do you really believe this to be true?
pip t said:
Mr E said:
BLE beacons will run in the background for any app; that’s kind of the point of them.
Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Yes, apologies - both iOS 13 & the later Android OS's can allow background bluetooth, but only with the users explicit permission.Both Apple and ‘Droid OS will ask if the user wants to allow the app access to location data. Saying “no” prevents the beacons.
Regarding privacy, if we don’t do this I suspect the government will know exactly where you are almost all the time, mainly because you’ll be stuck at home.
Which leads to the question, why is France kicking up such a fuss with Apple over the privacy protections.....the people who download the app simply select 'allow' when asked by the OS if the app should have access to Bluetooth, and everything's hunky dory?
Were I to go down a speculation rabbit hole with this (Forgive me, don your tin foil for one moment), the only scenario where this would cause a problem is if downloading the app is mandated by law. In that case it becomes an issue as you could download the app (Yes officer, here it is on my home screen) but then revoke the bluetooth permissions to render it essentially inactive....hmmmmm.
https://www.gizmodo.co.uk/2020/04/france-wants-app...
The dangers of speed reading and assuming what an issue is about!
rxe said:
pip t said:
Because if you're only with the infected person for a short amount of time, it would appear the transmission risk is low. So no need for an alert. Of course, in a few cases there may be transmission, but these will be a small amount of edge cases.
What scientific basis is there for this threshold? I'll help out: there is none. If there was, we wouldn't need social distancing when walking in the street. No issue with loads of people walking in parks, because you can't catch it. Unfortunately that simply isn't true. If it was true, people would struggle to catch colds on public transport. You catch this thing simply by touching a surface that has had virus shed on it. Not "holding it for 10 minutes".pip t said:
Not a great analogy. If your parachute fails to open, it's pretty much certain death or very serious injury. So a 50% chance of your parachute not working is a huge risk. If you contract CV19, you may not notice, you may have very mild symptoms, you may feel a bit poorly, you may feel thoroughly st, or, in small percentage of cases you may suffer very serious illness, and you may die. So the 50% chance of catching it only leads to a very small risk of a serious outcome.
You're missing the point of the analogy. A fully working parachute enables recreational skydiving. A properly working app is (in theory) going to allow people to go back to a more normal life, confident that the app will protect society by highlighting potential infections quickly. If the app doesn't work, people will confidently go back to normal, but the protection the app affords will be a mirage. To put it really brutally: the app assumes that you can't catch this on a 5 minute tube journey. Do you really believe this to be true?
Never really seen anything this could do as a single saviour of the world but part of a plan that can work with this in the tool box, but also part of a plan that will work without this app. Just that this app could add more functionality to your life so to speak if you chose to use it.
Gassing Station | News, Politics & Economics | Top of Page | What's New | My Stuff