NHS Covid 19 App

[TopCat]

It couldn't tell you when your actually close to them anyway. It'll tell you a week later (when they've had a test) and you'll have no idea who it was.

So nothing will happen via the app for at least a week?

 

[Pickman's model]

Has anyone downloaded it? Anyone using it?
I'm a bit reticent.

Yeh I'm not keen either

 

[TopCat]

I did not get any code or even entered my details? But its scanning?

 

[maomao]

So nothing will happen via the app for at least a week?

It could be quicker, a couple of days or something, but the whole testing system is fucked atm.

 

[Winot]

Your understanding based on what? Do you have a link?

Imogen Parker, the head of policy at the tech thinktank Ada Lovelace Institute, quoted here:

"Parker also raised alarm at the prospect of large numbers of people being advised to self-isolate based on “false positive” results. “The best data I’ve seen suggests 45% false positives and 33% false negatives,” she said, “but phone proximity isn’t everything. The growing body of evidence about things like the substantially limited risk outside versus inside really matters. We need to make sure the app can identify risk, not just identify phones.”

 

[cupid_stunt]

I did not get any code or even entered my details? But its scanning?

You don't need a code now it's officially launched across England & Wales.

It doesn't ask for you details, as it's anonymous, apart from the phone number, which is required for a text alert, if you have been within 2m of someone that gets a positive test, for 15 minutes or longer.

 

[Cid]

Is this the one that was privately developed and meant to be ready in June? They can stick it up their arse.

No, afaik this is based on the Apple-google thing.

 

[TopCat]

Imogen Parker, the head of policy at the tech thinktank Ada Lovelace Institute, quoted here:

"Parker also raised alarm at the prospect of large numbers of people being advised to self-isolate based on “false positive” results. “The best data I’ve seen suggests 45% false positives and 33% false negatives,” she said, “but phone proximity isn’t everything. The growing body of evidence about things like the substantially limited risk outside versus inside really matters. We need to make sure the app can identify risk, not just identify phones.”

I'm due to go back to work shortly which is a drag. 2 weeks more lazing about might work for me.

 

[Cid]

I’ve installed it on the principle that I don’t go near people that often, and will be something of an impetus to keep distance.

 

[platinumsage]

Imogen Parker, the head of policy at the tech thinktank Ada Lovelace Institute, quoted here:

"Parker also raised alarm at the prospect of large numbers of people being advised to self-isolate based on “false positive” results. “The best data I’ve seen suggests 45% false positives and 33% false negatives,” she said, “but phone proximity isn’t everything. The growing body of evidence about things like the substantially limited risk outside versus inside really matters. We need to make sure the app can identify risk, not just identify phones.”

Thanks, I traced it to an article in the Telegraph in August:

Testing so far has found that, for every 10 cases that should be detected, seven are while three are missed. But calculations from 100,000 simulations also found a "false positive" rate of 45 per cent - meaning almost half of all the cases being identified as close contacts do not actually meet the criteria.

Health officials said they hoped to improve on the accuracy on this in the next update to the technology, which happens next month. They added that some of those counted as "false positives" could involve people who been in fairly close contact with Covid cases but further away than two metres.

Professor Christophe Fraser, of Oxford University's Big Data Institute, an adviser to the NHS on the app, said: "At the moment, contact tracing is based on getting people to recall who they have been close to for 15 minutes. This [level of performance] is way better than that."

New NHS contact tracing app has 'false positive' rate of almost 50 per cent

Data suggests almost half all the cases being identified as close contacts do not actually meet the criteria

www.telegraph.co.uk

 

[maomao]

I'm only close to people that long if they're family or in my class at uni anyway. There's procedures in place for the latter and while my kids are possible vectors (my wife isn't really, she's not going out much at the moment) they don't have phones so the app wouldn't work there.

 

[platinumsage]

My neighbour accidentally connected to my bluetooth speakers once and started playing an audiobook. I wonder how Google/Apple's bluetooth tech judges distance and whether it will cope with walls...

 

[Winot]

Thanks, I traced it to an article in the Telegraph in August:

Testing so far has found that, for every 10 cases that should be detected, seven are while three are missed. But calculations from 100,000 simulations also found a "false positive" rate of 45 per cent - meaning almost half of all the cases being identified as close contacts do not actually meet the criteria.

Health officials said they hoped to improve on the accuracy on this in the next update to the technology, which happens next month. They added that some of those counted as "false positives" could involve people who been in fairly close contact with Covid cases but further away than two metres.

Professor Christophe Fraser, of Oxford University's Big Data Institute, an adviser to the NHS on the app, said: "At the moment, contact tracing is based on getting people to recall who they have been close to for 15 minutes. This [level of performance] is way better than that."

New NHS contact tracing app has 'false positive' rate of almost 50 per cent

Data suggests almost half all the cases being identified as close contacts do not actually meet the criteria

www.telegraph.co.uk

45% false positives is a terrible figure for a system that has serious implications for users and is enforced with substantial fines. Let’s hope the actual statistics are better than that.

 

[prunus]

I'm only close to people that long if they're family or in my class at uni anyway. There's procedures in place for the latter and while my kids are possible vectors (my wife isn't really, she's not going out much at the moment) they don't have phones so the app wouldn't work there.

Why not download it anyway (genuine question)? It can’t harm, and can only help.

 

[Orang Utan]

It doesn't work on older iPhones (iOS 13.5 or under)
I downloaded a few days ago but uninstalled it yesterday. Not comfortable sharing info with Deloitte/Serco/G4. “NHS” app, my arse

 

[tommers]

All the reviews seem to be from before today, the official launch date, and many are just moaning it doesn't work because they don't have a code, due to them not being an NHS worker or living in the areas being trialled.

I wouldn't take any notice of the rating ATM.

Yeah i haven't but lots of people will. I'm sure there are ways of hiding apps in the play store or relaunching it once it's on general release so that you avoid this problem.

 

[platinumsage]

45% false positives is a terrible figure for a system that has serious implications for users and is enforced with substantial fines. Let’s hope the actual statistics are better than that.

But if most of those are prolonged contacts slightly further apart than 2 metres then it's not so bad is it? Surely they must have published more data about this by now?

 

[Badgers]

Downloaded it. Likely it will be fucking shit like everything these days but am interested and do hope it will work.

I am not working in an office, using pubs or public transport so my exposure is limited to the odd supermarket and very occasionally a charity shop.

Have been reporting via the Zoe App daily since it started.

 

[cupid_stunt]

45% false positives is a terrible figure for a system that has serious implications for users and is enforced with substantial fines. Let’s hope the actual statistics are better than that.

According to the tech expert on BBC News, you can't be fined for ignoring a text alert from the app, because they will not know as it's all anonymous.

 

[Cid]

Doesn’t seem to have too much effect on my battery (iPhone 7 plus).

 

[Winot]

But if most of those are prolonged contacts slightly further apart than 2 metres then it's not so bad is it? Surely they must have published more data about this by now?

I hope the data exists, yes, and I am willing to be persuaded. I do want this to work. I am not however going to risk my children being unnecessarily removed from school on the basis of an app that at the moment looks flawed.

 

[platinumsage]

It doesn't work on older iPhones (iOS 13.5 or under)
I downloaded a few days ago but uninstalled it yesterday. Not comfortable sharing info with Deloitte/Serco/G4. “NHS” app, my arse

Personal info remains on your phone, and the anonymized data is kept on the Dept of Health secure cloud Microsoft and Amazon servers.

[Withdrawn] NHS COVID-19 app: privacy information

Privacy notice and data protection impact assessment for the NHS COVID-19 mobile app.

www.gov.uk

 

[salem]

I'm not super keen on the idea but I think it's important to society so will give it a go - being careful to monitor what permissions it wants etc.

So far it's not asked for anything beyond the first half of my post code and Bluetooth access. I'm OK with that. This will be watched by privacy types very closely so I'm fairly confident it won't be trying to do anything beyond its remit.

The 'in-house' app they had originally planned to build I wouldn't have touched with a barge pole. This one so far seems harmless enough.

 

[Badgers]

Unless a critical mass download it then it is almost useless.

 

[salem]

Just an FYI it's currently the second result on Google play for NHS track and trace!

NHS COVID-19 - Apps on Google Play

Protect your loved ones with the NHS contact tracing app for England and Wales.

play.google.com

 

[souljacker]

Personal info remains on your phone, and the anonymized data is kept on the Dept of Health secure cloud Microsoft and Amazon servers.

So freely accessible to the NSA and GCHQ.

 

[Badgers]

So freely accessible to the NSA and GCHQ.

Like all our other data and info then?

 

[souljacker]

Like all our other data and info then?

Yep.

 

[gentlegreen]

Since from the week after next, and for the forseseeable future I will never be within two meters of another human being for more than a second or two, I imagine I can forgo this pleasure and leave my Bluetooth turned off.
I doubt it would even run on the phone I use most often.

 

[cybershot]

Eurgh, doesn't work on iPhone 6 my Mum has, and I'm in no position to upgrade mine to hand it down to her like I usually do. A) can't afford and b) really dumb time to buy an iPhone with new ones next month.

Can see her kicking off with me, lol.