Bluetooth.

That's what the re-writes to the code that Apple and google are doing allows for.

There are flaws, though

No, what's being spoken of is self-reporting. It'll be down to the user to tell the app whether or not they have contracted the disease.

Here you go:

Tell me that's not open to abuse from trolls.

I'm not entirely sure how it's proposed to have been done, but I got this from an article I read a while ago. I've tried looking for it again, but haven't found it.

I'm sure it was a reputable source, but I'm definitely not as confident as some of the specifics as I was.

Well, that's a screw-up on the level only the British are capable of.

A contact tracing app can be very useful, but this is indeed BS.

McHrozni

Bluetooth works through all the perspex screens / shields that have been recently installed to protect cashiers in shops. It also works through most thin walls / floors / ceilings.

I don't see how any app can recognize that a Bluetooth phone signal coming from a phone on the other side of a wall doesn't represent a risk of virus transmission.

This is indeed one of the issues.

UK COVID-19 contact-tracing app data may be kept for 'research' after crisis ends MPs told and users will not be able to ask NHS admins to delete their COVID-19 contact-tracking data from government servers.

https://theregister.co.uk/2020/05/04...ts_parliament/

I won't be downloading Cummings data harvesting app.

Google Marc Werner and decide if it’s actually a COVIDー19 app or just another Cambridge Analytica ******** project

Will the UK pass Italy for the second highest COVID deaths today, or will that hold off until tomorrow?

Matt Hancock delivers his 100,000th daily coronavirus briefing*.

*Number includes appearances he has yet to make, but has the capacity to.

Wales has passed 1,000 Coronavirus deaths. :(

This article goes into why the UK's model for the contact tracing app won't work. The TL/DR version is that in order for it to work on an iPhone the phone has to be on, awake, and have the app running in the foreground. It can be in the background on an Android, but only for a few minutes.

It also goes in to the privacy issues a little.

When I said earlier that I would probably download it I did so under two assumptions that I now know to be false - 1) that it would actually work, even if only in a limited capacity, and 2) that it was using the API that Apple and google had developed, which seemed good when it came to privacy issues.

I'm generally not as bothered about privacy issues as some would no doubt think that I should be but, even if I can't see any particular harm for me personally in this data, I feel very differently about the government having this data and me having no control over it than I do about Apple having some of my data, given that there is documented evidence of Apple protecting its users privacy.

The data sits on your phone until you notify yourself as a case then it will report contacts. In principle this is no different from a face to face, or phone interview with a human contact tracer asking who you have met with and entering the data into a database.
https://www.nhsx.nhs.uk/blogs/digita...-saving-lives/

The Apple/Google app also requires voluntary alerting of the app that you have covid-19 to trigger contact tracing. So that is not a difference between the two models.

For security issues read here.

https://www.ncsc.gov.uk/blog-post/se...ct-tracing-app

The latest daily reported death total for the UK (29,427) is now higher than the total for Italy (29,315)
The UK has reached this figure faster in its epidemic than Italy.
Raab says there will be no "real verdict" until the pandemic is over

Of course not, and that'll be years away so no kickings until 2025 at the earliest....:rolleyes:

You are still probably better to down load it. You would get notifications about potential contacts from altruistic symptomatic people you had been in contact with, with potential access to testing that you might not otherwise have. If you are truly concerned about the NHS having access to the limited amount of data held on your phone, then you don't need to alert the system when you develop covid-19. (Remembering the NHS probably has you phone number, address, date of birth, details of next of kin, sexual history, psychiatric history, alcohol and smoking history, results of drug screens etc. So I am not sure that your phone is going to reveal much else?) If contact tracing happens to you then more intrusive data and less anonymously will be collected by contact tracers and entered into a database.

How long before they remove people without British passports and claims they belong to their home countries death stats? The way this government shirks its responsibilities is sickening.

Government scientist Neil Ferguson resigns after breaking lockdown rules to meet his married lover

Prof Ferguson allowed the woman to visit him at home during the lockdown while lecturing the public on the need for strict social distancing

https://www.telegraph.co.uk/news/202...igns-breaking/

U.S. officials believe China covered up the extent of the coronavirus outbreak — and how contagious the disease is — to stock up on medical supplies needed to respond to it, intelligence documents show.

Chinese leaders “intentionally concealed the severity” of the pandemic from the world in early January, according to a four-page Department of Homeland Security intelligence report dated May 1

https://apnews.com/bf685dcf52125be54e030834ab7062a8

:dl:

That is a classic worthy of the ages.

...anything that suits the agenda.

No irony at all in DHS being vilified daily until they come up with something that meets that criterion.

Might even be true, but at a time of outrageous growth in racist attacks on South-East Asians, isn't going to reduce tensions at all.

Has Tubby decided that Covid might get him dumped as POTUS so starting a war with China is a good move?

maybes he has a two metre schlong

Gov. the app does not use your location.
App. first question, what is your postcode.
Gov. This app does not record your personal details.
App. This app requires access to your contacts and camera

Governments around the world. our tracing app does not store your details on a giant database, you don’t need to worry about that
Gov, our app stores all your details on a giant database

And you need to keep your phone unlocked and the app in the foreground for it to work.

Dominic Cummings and his Vote Leave mates will decide how best to use your data after the crisis is over.

And it's dying down in Italy. Daily increases are well under 1% of total infected, UK is at over 2% daily increase. UK is where Italy was three weeks ago or thereabout. Death rates in both follow the infection rates.

UK will be the worst affected country in Europe*, by quite a margin.

*excluding Russia

McHrozni

The question is. Why ?

• Is it because of the UK's population distribution and density
• Is it because of the London airports
• Is it because of the UK's demographics
• Is it because of the UK's healthcare system
• Is it because of the UK government's (mis)handling of the Coronavirus
• Is it just random distribution and/or bad luck
• Are there other factors I haven't considered
• Is it a combination of some or all of the above

There are plenty of facebook posts where people are claiming that, for our population density, the UK is actually doing very well indeed on our per-capita death rates. Nick Hancock, by asking no-one to judge until the Coronavirus epidemic is over, is suggesting that the UK isn't actually doing that badly and it may be down to how other countries are reporting their figures.

Then again, the official figures are double what was considered "a good result" a few weeks ago, the real number may be considerably higher and we may only have seen a fraction of the overall death toll depending on how long the "tail" of this wave of infection is and how many waves there will be :(

The BBC's Reality Check has done a good analysis here:

Coronavirus: Can you compare the UK with Italy?

Yes, I read that yesterday and really didn't come away any the wiser.

The article seemed to claim that the UK's numbers aren't so bad because our population is 10% greater, we are more rigorous about counting and we have proportionally fewer excess deaths than Italy. Hooray, on that basis we're only second or third worst in Europe.

Indeed! A great question. Let's do them in turn.

Italy and Germany have about the same population density as UK. Spain and France have far lesser population densities. Population density does not seem to be a major predictive factor.

Major airports are present all over the continent and there is no visible connection to outbreaks.

Italy is the oldest country on the continent. So, no.

Ah, now we're seeing the first point that may be relevant. Italy has about double ICU capacity compared to UK, for about the same population. Indeed a more robust healthcare system may help, but the key is in the next point.

Definitely a major if not the sole factor. Both Italy and UK recorded the first cases on the same day (January 31st), but Italy acted much sooner and more forcefully than UK did. Italian response was insufficient and caused a major uptick in death toll, British was outright criminally negligent.

Nah.

There probably are, but the main point was the (mis)handling of the epidemic - acting too late and not forcefully enough. This is the case literarily everywhere we look, it stands to reason it also works in the UK.

Except of course Wales is particularily hard hit and there is little to no correlation to population density and death toll. Sweden is one of the hardest hit countries per capita and Sweden makes Scotland seem like a metropolis.

What's the difference? Sweden has a different approach with very limited measures.

McHrozni

It could be argued that because of the first three items on your list, it makes a very much stronger case that the lockdown should have been implemented even earlier by our government. R3 level infections were taking place in the week commencing 17th March, when the Cheltenham Festival and a Liverpool football match were allowed to go ahead. How many individuals were infected during that week and how many did they infect subsequently?

Or it could just be the factor I saw recently neatly summarised:

No, it only depends on one of those. Vietnam has a greater population density than UK, but their population is not nearly as dense.

Vox named it one of the success stories.

https://www.vox.com/2020/5/5/2124783...iceland-greece

Vietnam health care system, well, leaves something to be desired. Their response however has been swift and effective and kept the outbreak under the lid largely using the old fashioned way.
Before you ask if this is due to inadequate testing, check the bar graph in the article. In Vietnam, one of a thousand tests came out positive. In UK the number is one in five.

McHrozni

I could also get notifications from trolls, causing needless stress.

It also seems to me that if it won't actually help with contact tracing (as it seems it won't due to it requiring phones to be awake with the app in the foreground, due to the limitations of bluetooth, and due to it being highly unlikely that 60% of the population will download the app and have it with them on a phone that is awake with the app in the foreground), then it's actively dangerous by giving people a false sense of security and thereby encouraging risky behaviour. I don't want to be complicit in that.

See, you're saying NHS, when I said the government.

Bear in mind that a leaked NHS memo flagged this up as a concern: https://www.theguardian.com/world/20...nonymise-users

I may be misunderstanding you. But the data shows that the UK is over the peak and deaths are falling (which will represent infections 1 - 2 weeks ago). The peak of hospital deaths and deaths in those with a positive test peaked about 8 April, ONS data showing total deaths and deaths attributed to covid-19, for week ending 24 April registrations are lower than week ending 17 April.
https://www.ons.gov.uk/peoplepopulat...isional/latest
Given registrations are later than date of death this would appear to suggest that mortality has peaked and is currently falling. Though whether falls are happening as fast as elsewhere is hard to tell at present.
Looking at deaths with a linear rather than log axis on this site which uses a non-date based graph,
https://aatishb.com/covidtrends/
The rate of fall seems to be the same between UK and Italy at present (the lines are parallel). What is interesting is that the UK seems to have had a longer plateau. The reason for this is unclear. Perhaps people have been cheating on the lockdown so transmission persisted for longer? I would favour the fact that the UK may have a series of outbreaks that are staggered, there is some support for this in that London appears to have been ahead of the rest of the country so we are probably looking at two superimposed curves one for London and one for elsewhere.

If you look at France the two Northern regions have over 50% deaths
https://www.statista.com/statistics/...region-france/
comparing the UK
https://www.statista.com/statistics/...ion-in-the-uk/
There is a more even distribution, London and the South East are only 30% of cases despite the relatively high population.
Italy appears to have an even more localised pattern of disease
https://www.statista.com/statistics/...gion-in-italy/
(Unfortunately these figures are for total deaths and it would be helpful if there were population corrected figures but I cannot find them and don't have time to calculate them.)

It is an interesting question as to why the spread of covid-19 appears to be more widespread and less localised in the UK as compared to other countries. Italy did do a local lock down early which may have limited spread to elsewhere in Italy. However, looking at routes of entry into the UK
https://nextstrain.org/ncov/europe?b...=radial&p=full
There were multiple localised entries, e.g. from Russia into Scotland.
(To be clear these are sequence derived origins of virus and do not necessarily indicate an individual's travels, so a 'Russian" strain could have arrived from Eastern Europe or the Netherlands.) Certainly spread from Italy was widespread in the UK, and a number of localised outbreaks were attributable to people returning from Skiing holidays in Italy. (Which just reinforces my view that skiing is unnecessarily dangerous to be enjoyable.)

He said it was okay because he thought he was immune.

Well, that's all right then. I had symptoms back when this all started. I'm probably okay to host a party. That'd be fine, right?

It depends on whether you view NHS as separate from the UK government (or Scottish, Welsh and NI governments). Compared with the direct identifiable data the NHS holds which is not anonymised, the information from your phone seems of minimal concern. This is an NHS app, notifications will be to and from the NHS, it is not a government app, except in as much as you view the NHS as being part of the government. I note the memo which was a draft was discussing possible options for the development of the app, it does not say this is possible with the current app. One needs to be careful that discussions of options in development are not over-interpreted as representing the final shape of the app.

I am also curious whether you would refuse to interact with human contact tracers on privacy grounds? The data they would collect would be more extensive than an app would supply, and end up with the same people (NHS).

The issue about trolling applies to all these apps including the apple / google app. The NHS version is likely to be less vulnerable as it could filter people who repeatedly trigger an alert, whilst the apple / google app has no option to filter trolls.

The issue about whether the app runs in the background is due to apple / google refusing to allow the NHS app to run in the background. This problem could be solved with the co-operation of apple / google. Perhaps you could criticise them for preventing the development of an optimum app.

So different options have strengths and weaknesses. The NHS version is probably less vulnerable to trolling and could be integrated with testing, as the NHS would send out an alert to contacts that might include how to be tested, and perhaps direct contact with a contact tracer, or symptom questionnaire. An apple / google app notifies, would be more vulnerable to troll notifications and includes no follow up unless you voluntarily inform the health service you are a contact.

The locality data requested by the NHS app is only the first part of post code of place of residence, this localises to thousands of households. It does not monitor location by e.g. GPS. So they could not know that for instance your contact was through visiting your married lover (to be topical).

A useful shorthand to get grips on how far you've come is to compare the number of newly diagnosed infections in the past day to the number of total diagnosed infections. In Italy that number is about 1% or lower and has been below 2% for weeks. UK is still at 2-3%, the drop is slow. Epidemic is said to be dying out when the number is consistently below 1%, which has been the case in Italy for just over a week. UK is at least two weeks away from that point and probably closer to a month.

The most likely reason for a longer plateau is greater availability of testing in the country with the shorter plateau. The reason is simple, as the control over the epidemic improved, a greater proportion of people got diagnosed and we see that as a plateau. The drop began when the epidemic began to wane.

There is one plausible explanation for that. Virus arrived in Europe far earlier than we knew, France found it in a sample from late December 2019, before China even notified the WHO. That person had no contact with China, which means there was a human to human tranmission in France in December 2019 already.

However - this is important - it turns out the viral load is crucial in determining the severity of the illness and infectivity of the patient. Those who recieved a significantly greater number of viral particles get sicker and spread the virus better. Until there was a sufficient number of people already caughing and sneezing the R₀ remained low. It wasn't until weeks later that enough slightly ill people accumulated that the outbreak truly began to spread.

This would fully explain why we're seeing these small, localized outbursts that don't amount to anything and a few much deadlier bouts of disease where the critical mass does accumulate. There's no denying this happens, more people died in Bergamo in March 2020 than in March 2019, 2018 and 2017 combined. However for this to happen you need to already have conditions very favorable for the disease to spread, with many hosts all over the place. You're quite safe until then but once you do have that critical mass, tens of thousands will die and there's little we can do to stop it.

A nastly little bug this SARS CoV-2.

McHrozni

No it probably wouldn't. A party would have many more possible interactions, than two individuals. One of whom was (potentially) immune, and the other had already been exposed.

I do think this is the lowest form of 'gotcha' journalism. Hypocrisy is not a major sin. If a murderer says killing people is wrong it does not mean killing people is right. I note the heavy use of 'married lover'. What is the relevance as to whether the individual is married? Is this in the public interest? I think not. All it will mean is that individuals who are not 'public figures' but are merely carrying out their job - giving technical advice to the government, will be less willing to be identified, will be less willing to discuss the technical details with the press, in case they are subject to the same sort of 'gotcha' story for some peccadillo. We are arguing for more transparency, to know who is on SAGE, I certainly would not be in a hurry to be identified or to communicate directly with the press after this. In addition the government has lost an expert advisor. So I do not think there is any public interest defence in this article indeed the consequence will be actually harmful to the public interest. There was no public health risk in the particular circumstances.

In contrast if a politician was arranging regular parties or indeed you, I would certainly view publicising this as in the public interest since unlike this case there is a very real public risk.

Comparing current deaths to current tested positive is a bad measure.
1) Policy and availability of testing differs widely between countries.
2) The deaths relate to infection rate around two weeks ago. So if you are over the peak and infection rate is falling, current infections are lower than infections two weeks ago, so as one goes over the peak the expectation is that the ratio of current deaths to current infections should rise. Conversely during the initial rising phase current deaths to current infections will be lower.

So one could argue the higher current deaths to current infections is actually a positive sign.

ETA I see I misread your post as you referred to total deaths not current deaths apologies

My comment about differing testing availability and policy still applies as to comparability. I'll need to think about how changes in the phase of epidemic can reflect total deaths to current infections.

So assuming you had perfect knowledge you would have two identical curves but with one being 1% of the magnitude of the other assuming IFR 1%, but displaced by 2 weeks. Initially the total deaths to current infections would be below 1% as we approach the peak and for 2 weeks afterwards the total deaths to current infection rate would rise even as current infections and current deaths fell. So I am not convinced this is a useful measure of the state of the epidemic. Is there an academic reference to use of this parameter?

It's not current deaths to current tested. It's positive results in the last 24 hour period compared to the total number of positive results throughout the testing, in the same jurdistiction.

If the change is not due to major changes in testing regimen (which you can exclude) then the epidemic is dying out when that number is consistently below 1%. It's one of the parameters closely followed by epidemiologists.

Yeah, I noticed :)
Total deaths have another couple of confounding variables - dead with Covid-19 or dead from Covid-19? Those two numbers will differ. Furthermore, recent studies have shown that even the best of countries only detect ~2% of Covid-19 patients, but the death tolls tend to be far larger than those attributed to Covid-19. Comparison is difficult until we know more.

McHrozni

He ****** up but he's resigned for it. Give him credit for that.

They'll be plenty of others who've done as bad or worse who won't.

The irony of our age is that those who have the integrity to admit, and maybe learn, from their mistakes resign while those that won't remain to repeat them. Integrity has negative survival value in politics.