The contact-tracing app, accuracy?

How does this contact-tracing app work?

I mean GPS is typically accurate to around 5m. So it could easily put 2 people in the same location, even if they were up to 10m apart.

It also detects Bluetooth, but IME the range of that varies widely and I don’t think it would be capable of measuring distance from the signal?

Am I missing something?

It’s not GPS. It’s basically Bluetooth with the app announcing itself and listening for other devices. If two phones (call them Bob and Jeff) detected each other yesterday and Jeff discovers that he has Covid then Bob can be alerted. Distance isn’t relevant, if two bluetooth devices can communicate then they’re close enough.

https://www.theregister.co.uk/2020/05/05/uk_coronavirus_app/ explains the theory and why the UK’s implementation of a good idea is pretty terrible.

Distance isn’t relevant, if two bluetooth devices can communicate then they’re close enough.

My Bluetooth stuff works at a fair distance, it varies from device to device. However, it’s never less than about 5m if clear line of sight.

How can that be considered close contact?

The point is to identify people who have been in proximity to each other.

You can’t use GPS because you’d need to constantly track an individual’s location, which has horrific privacy issues, will consume all the battery life, and won’t work inside.

The Google/Apple approach is unique IDs that are swapped over Bluetooth. If two phones are near each other then the IDs are swapped and a proximity is logged. Yes this isn’t perfect: I can sneeze on a door handle and then you use the door an hour later, but it’s better than nothing because it does track everyone that you bumped into when walking around the supermarket.

First results of basic testing of the NHS app has shown that it’s failed on every level, which really doesn’t surprise me at all, taking into account the previous history of an interconnected system for all hospitals and doctors surgeries to be able to access a patient’s records regardless of where they were was dumped after years of effort and tens of billions squandered on it. I’ll possibly install the Apple/Google app, but I really don’t trust the Government-backed NHS app worth a damn.

I’ll possibly install the Apple/Google app,

My understanding is that Apple/Google have co-operated on the backend but you still need a regional app to use it. So, it’s the NHS app or no app.

My understanding is that Apple/Google have co-operated on the backend but you still need a regional app to use it.

Yep, they just wrote the BT API.

Good article on the underlying methodology and security principals in the API:

https://www.cnbc.com/2020/04/28/apple-iphone-contact-tracing-how-it-came-together.html

From a government that was won on the back of dodgy data practices, no thanks.
I’m all for a tracing app but not something that has cummings finger prints on it

From a government that was won on the back of dodgy data practices, no thanks.
I’m all for a tracing app but not something that has cummings finger prints on it

Agree with this, from what I have read there is no chance this app will be going anywhere near my phone/data.

Only slight caveat if I stick it on my old work phone as a burner phone with a pay as you go sim with dummy personal information.

My understanding is that Apple/Google have co-operated on the backend but you still need a regional app to use it. So, it’s the NHS app or no app.

I’m not even sure that the NHS app uses the A/G tracing API? (Because they don’t want to follow the privacy rules) The whole point of the A/G API is that it works in the background ALL THE TIME, whereas the NHS app has to be open/running on both phones in order to register a “contact”.

’m not even sure that the NHS app uses the A/G tracing API?

If it can run in the background then it probably has to as there wasn’t a way for a background app to run BT (other than play music) in iOS prior to the new API.

I wouldn’t be surprised if the NHS app doesn’t work technically. I understand that the new Google / Apple API is required for apps to do the bluetooth comms with other phones. These are generally blocked by Android / iOS because of privacy concerns. I can’t find anything written down on this, but on Radio 4 WatO today the interviewee said that the developers had ‘a workaround’ for this. That doesn’t sound robust enough to be rolling out as a national app.

My Bluetooth stuff works at a fair distance, it varies from device to device. However, it’s never less than about 5m if clear line of sight.

How can that be considered close contact?

Because the 2m thing isn’t a magic barrier.

If I’m sat in a room with someone all day within bluetooth range, I’ll probably catch whatever they have. If I cycle past someone coming in the opposite direction on the towpath then the odds of transmitting anything are slim/none.

It’s also not meant to be 100% accurate, if someone has symptoms, they could get tested once there is capacity, then if positive they can alert all the people their phone thinks they’ve been near and they can get tested. They can probably also write out a list of family members and friends they’ve been to the pub with, and ask their employer to send out an e-mail saying someone in the building has it, please get tested, etc and then you’re most of the way to a complete list.

The principle being that if most people with the virus can know they have it, and can isolate themselves then transmission rates should be very low anyway. The hope is that the level of infection is controlled in the thousands and contact tracing is relatively straightforward, rather than hundreds of thousands and not having a clue who has it and who hasn’t.

On balance I probably would download it, it could save someones life. But the implementation and privacy issues are just bonkers when there is a better way.

Opps, UK cocks up yet again…

Comment Britain is sleepwalking into another coronavirus disaster by failing to listen to global consensus and expert analysis with the release of the NHS COVID-19 contact-tracking app.

On Monday, the UK government explained in depth and in clearly written language how its iOS and Android smartphone application – undergoing trials in the Isle of Wight – will work, and why it is a better solution to the one by Apple and Google that other nations have decided to adopt. It has also released a more technical explanation.

Unfortunately for folks in UK, while the explanation is coherent, calm, well-reasoned and plausible, it is likely to be a repeat of the disastrous “herd immunity” policy that the government initially backed as a way to explain why it didn’t need to go into a national lockdown. That policy was also well-reasoned and well-explained by a small number of very competent doctors and scientists who just happened to be wrong.

Here’s what happening: there are broadly two types of coronavirus contact-tracing apps; those that are centralized and those that are decentralized. The first takes data from people’s phones and saves it on a central system where experts are trusted to make the best possible use of the data, including providing advice to people as and when necessary.

The second, decentralized approach, as set out by Apple and Google, puts users in more control of their information, and alerts them automatically with no intervention from a third party. Apple and Google have also banned apps that use their decentralized and anonymized API from accessing location services to track and identify people, despite pressure to do so. And they have said they will only allow one app per country, or state in the US.

Both types use Bluetooth to detect other nearby phones also running the software. Thus, when someone catches the coronavirus, people can be warned if their phone was within 6ft of that patient’s phone for more than a few minutes.

https://www.theregister.co.uk/2020/05/05/uk_coronavirus_app/

TLDR: It doesn’t work.

If it can run in the background then it probably has to as there wasn’t a way for a background app to run BT (other than play music) in iOS prior to the new API.

@footflaps I thought that was the whole point of the criticism of the NHS app – it DOESN’T run in the background?

https://www.theregister.co.uk/2020/05/05/uk_coronavirus_app/

TLDR: It doesn’t work.

Great link thanks @footflaps.

Hey I posted that link in the first reply 🙂 The NHS app is DOA.

1) Absolute Bluetooth radio strength (RSSI) is a poor method of measuring distance, exponentially more so on Android due to vastly more diverse hardware platforms. Rather, you can only really tell if you’re getting closer / further away by looking at change data. The Bluetooth SIG itself advises you not to do it: https://www.bluetooth.com/blog/proximity-and-rssi/ [EDIT: I’ve just realised this article is 5 years old so this may no longer be current advice.]

2) Bluetooth doesn’t work like that. iOS outright forbids background apps from continually broadcasting Bluetooth IDs and (later versions of) Android restricts it to a few minutes. You’d either need to go through Apple’s / Google’s locked-down API or expect everyone to be walking round with their phones on and unlocked.

Their claimed workaround for this is that it can still listen in the background and respond to other device broadcasts, so that rather hangs on a sort of ‘herd immunity’ from people walking around with Android 7 or lower devices.

3) Bluetooth doesn’t understand walls. Is it going to flag up that I’ve been within 2 metres of my next-door neighbours when I haven’t even left the house? False positives aplenty.

4) Unless they’re only allowing post-test reporting, the risks of abuse out of either malice or devilment are high. Little Johnny fancies a couple of weeks off school; a corporate employee takes a stroll around a competitor; someone attaches a phone to their cat or dog and lets them out to play. This will absolutely happen without proper control, I saw a video the other day about a guy with a wheelbarrow full of mobile phones causing traffic jams on Google Maps for the lols.

5) It’s being developed by Faculty – formerly ASI Data Science. They’re heavily into AI and machine learning and whilst they might not be inherently evil they’ve been working for Cummings and his Machiavellian schemes since at least his Vote Leave days.

I’m not one for conspiracy theories but it does rather beg the question as to why you’d go to data mining specialists in order to ask them to write a Bluetooth tracking and recording application. For this reason I’m out.

My understanding is that Apple/Google have co-operated on the backend but you still need a regional app to use it.

My understanding is that the UK isn’t using the Apple / Google APIs at all.

Hey I posted that link in the first reply

you expect people to actually read the thread before commenting? Are you new here? 😉

I’m not one for conspiracy theories but it does rather beg the question as to why you’d go to data mining specialists in order to ask them to write a Bluetooth tracking and recording application. For this reason I’m out.

Exactly this. Google and Apple co-operated to make both the privacy (it’s actually anonymous) and technical (the OS can listen constantly on behalf of the app)) issues disappear. There is no good reason why the NHS would instead build their own.

whats to stop someone saying they have the virus and symptoms on their app and go around spreading false positives?

Seems the testing is going well.

https://www.hsj.co.uk/story.aspx?storyCode=7027564

Oh.

Hey I posted that link in the first reply 🙂 The NHS app is DOA.

Cripes I’m sorry! Thanks for your other posts on this thread too. It’s a fascinating subject to me because I’m a developer and would never (nowadays) consider trying to make a workaround for something that the app shouldn’t be allowed to do – so I’m intrigued to see how this plays out.

Senior figures told HSJ that it had been hard to assess the app because the government was “going about it in a kind of a hamfisted way. They haven’t got clear versions, so it’s been impossible to get fixed code base from them for NHS Digital to test. They keep changing it all over the place”.

With an application as important as this, and with such major privacy implications, your project delivery has to be state of the art, from start to finish. Comments like this just confirm it’s government amateur hour again.

It’s going to be a PITA when people get the here, (large national power station) and there’s the first confirmed case. Because we’re basically all at work, (a combination of local managers who don’t trust us to work at home plus genuine reasons to be here at least some of the time) everyone is going to get the message to isolate until they can get tested all at the same time, leading to 90%(?) absence for a couple of because we all routinely mingle about.

Agree with this, from what I have read there is no chance this app will be going anywhere near my phone/data.

your data being exactly how much more important than peoples lives?

Agree with this, from what I have read there is no chance this app will be going anywhere near my phone/data.

Won’t go on my phone, battery only lasts a day if I stick it on flight mode when I’m out the house! Keeping it out of sleep it goes flat in under an hour.

your data being exactly how much more important than peoples lives?

Things don’t stop being bad because there’s something potentially worse. I’d rather have a kick in the bollocks than cancer but that wouldn’t make me pro-bollocksbooting.

In any case, I have serious doubts that this is going to ‘save lives’ as it currently stands.

your data being exactly how much more important than peoples lives?

Unfortunately totally irrelevant as the UK app is simple not going to work at all.

No doubt Apple/Google will be painted as the baddies in this even though they’ve been saying “work with us if you want to make something that actually functions” from day one.

Some interestImg detail here based on facts. I know right..

NCSC inside knowledge

It’s a long read but goes into the right amount of detail.

TL/DR it’s fine, I’ll be using it.

Given the governments record on IT systems, data security and deals with the likes of Cambridge Analytica etc then there is no way this is going on my phone.

The fact that it wont work makes me even more suspicious of the whole thing as it means its being done for 1 of 2 purposes

1. The most likely, its another government cock up or PR stunt designed to keep the masses happy and think they have a competent government in charge

2. The more foil hat version, it works but not for the publicly stated purpose

Option 1 is far more likely so there is no point installing it

Some interestImg detail here based on facts. I know right..

NCSC inside knowledge (https://www.ncsc.gov.uk/blog-post/security-behind-nhs-contact-tracing-app)

That’s the paper that the Register article from the 2nd post dissects, including pointing out that

“Despite what the NCSC has continued to imply, the app will not, as it stands, work all the time on iOS nor Android since version 8. The operating systems won’t allow the tracing application to broadcast its ID via Bluetooth to surrounding devices when it’s running in the background and not in active use. Apple’s iOS forbids it, and newer Google Android versions limit it to a few minutes after the app falls into the background.”

This quote sums it up for me 🙂

—
Senior figures told HSJ that it had been hard to assess the app because the government was “going about it in a kind of a hamfisted way. They haven’t got clear versions, so it’s been impossible to get fixed code base from them for NHS Digital to test. They keep changing it all over the place”.
—

The Bluetooth foreground issue seems to be the only point of any substance in the Register article. Don’t get me wrong I really like them and trust their judgement most of the time, but this just stinks of click bait.

Interesting times.

Some more details from Wired:
https://www.wired.co.uk/article/nhs-contact-tracing-app-data-privacy

This quote sums it up for me 🙂

—
Senior figures told HSJ that it had been hard to assess the app because the government was “going about it in a kind of a hamfisted way. They haven’t got clear versions, so it’s been impossible to get fixed code base from them for NHS Digital to test. They keep changing it all over the place”.

That’s what I was referring to earlier, but I couldn’t find the article anywhere, I’d forgotten to save it.

Track and trace isn’t really going to help these poor elderly people in care homes is it, and how would they spot a breakout of the virus in a prison? or any other establishments that require you to hand your mobile in at reception when you visit or go to work.

As many have already mentioned, NHS have often been handed shoddy IT solutions over the years and with this gang of second rate circus jugglers running the show I don’t see anything changing.

Although it might be very successful on Aer Lingus flights from Belfast.

Some interestImg detail here based on facts. I know right..

NCSC inside knowledge

It’s a long read but goes into the right amount of detail.

TL/DR it’s fine, I’ll be using it.

I’m more concerned about the tone of the NCSC article than the Government surreptitiously harvesting my data. I expect the NCSC to write balanced content not, “well the medical benefit is potential so high it outweighs the disadvantages – but everyone has promised to play fair so it will be ok.” The article is so in favour of the adopted solution I conclude it was written drafted by the NHSX team themselves!

I’ll not be using it (in part because Scotland doesn’t seem to be going that way) but also because without >50% of people I interact with using it, it will be pointless and there’s no way there is that much trust in government harvesting the data; IF that means people will die it’s the fault of either government for destroying trust or the app development team for living in a bubble where this crisis overruled public perception.

FWIW I don’t accept that not having a contact tracing app means more people will die. It might slow down an exit from lockdown. It might mean that social bubbles are harder to create or manage (or fall back on some much simpler technology), and it might even mean we end up in a second lockdown – but a smartphone app telling you you were close to someone unknown who might have got the virus doesn’t stop you getting it, not indeed you passing it to others (you think the third time someone gets one of those they are still going to be responding enthusiastically to quarantine pending test results?).

It does seem darkly hilarious that the contact tracing app is reliant on herd i-phone-ity. The background running issue just seems insurmountable tbh.

Ffs, I was going to put aside my concerns with the government and Cummings etc and install this to be honest.

If it ends up not even working?!…

Can’t this bloody shower do anything right?

Say wut bluetooth coronavirus