The NHS COVID-19 app

[stuart]

Nice article to put minds at rest. If only other COVID-19 issues could be sorted as quickly:
https://www.theregister.com/2020/09/28/ ... s_its_way/

Stuart

[JGD]

Reassuring words it would seem from Stuart's link to The Register.

NHSX and the Department of Health resolved this myopia. From Sunday, anyone who receives a positive test result administered by another government body will also receive an SMS code allowing them to register the result with the app and warn all their close contacts.

"Everyone who receives a positive test result can now log their result on the app. If you get a test in a NHS hospital, through a PHE lab – which carry out tests for NHS hospitals – or in a surveillance study, you can request a code from NHS Test and Trace to log a positive result," the Department of Health said in a statement yesterday.

However, it is unable to recognise negative test results performed outside of the app.

The Register piece has another link to a .GOV Statement:

https://www.gov.uk/government/news/nhs- ... lion-times

And it has further commentary about downloads and users scoring the apps from Apple and Google.

People in England and Wales have given an overwhelming response to calls for them to download the new NHS COVID-19 app, with over 10 million people downloading it so far, 6 million of whom did so on its first day (Thursday 24 September).

The app forms a central part of the NHS Test and Trace service in England and the NHS Wales Test, Trace, Protect programme – identifying contacts of those who have tested positive for coronavirus.

On Sunday 27 September at 12pm, there had been over 10 million downloads across compatible Google and Apple devices in England and Wales.

In addition, the app has received a warm reception from those downloading it with reviewers on the Apple app store giving it a 4.5 star review (out of 5) and the Google Play Store giving it 4.1 stars (out of 5).

Perhaps we are getting close to receiving something that works as advertised from HMG. It is to be hoped the tracking features actually work to assist us.

[Robin Orton]

I've now downloaded this myself as a matter of civic responsibility, but with fear and trepidation. Does it make some sort of horrible noise when you've been in contact with someone naughty?

[JGD]

Don't yet know - not encountered that situation yet.

But I will also say I will be disappointed if it causes my iPhone to deliver any form of electric shock- just by way of ensuring it got my attention.

[stuart]

But I will also say I will be disappointed if it causes my iPhone to deliver any form of electric shock- just by way of ensuring it got my attention.

A combined Covid/Defibrillator app? Cool, but would it need a better battery? Otherwise:
"I'm sorry but there is insufficient charge to perform your body update. Goodbye and thank you for dying with EE."

Stuart

[zara]

Tried to download but not able to on iPhone 6 tried iPad Air not compatible . So I cannot join in. Not buying a new phone just for this.

[JGD]

Hi Zara

Sorry to hear you can't load the app.

I wonder what version of IoS you have on the iPhone 6?

iPhone 6s/6s Plus can be updated to IoS Version 14 - which will work with the app.
iPhone 6/6 Plus can be updated only to IoS Version 12 - which will not work with the app.

The NHS support website says this:

The app is supported on Android and Apple (iOS) phones. The app currently supports Apple iOS versions 13.5 and higher, and Android version Marshmallow (v6.0) and higher. This is because the app needs the Exposure Notification framework developed by Apple and Google, which is only available in these versions.

I am really sorry if you only can get I0S 12.

Anyone out there know how to feed back to the development teams and ask why it can't be made to operate with IoS 12?

The issue is very important - so as an alternative, can Apple release a version of a usable IoS, ie 13.5 or higher, that will work on the iPhone 6/6 Plus?

[mosy]

Andy Burnham and others are still asking why test and trace is not handed to local authorities who have a better handle on outbreak hotspots plus better access to data which the government currently chooses to keep to itself or releases at the last minute. Given the app's shortcomings and uncertainty of people actually self isolating if so directed, surely the government should acknowledged the efficacy of local management.

[mosy]

A question on the test itself. The UK test, we are told, is pointless if no symptms are present (false negative) yet TV says that the White House personnel take daily tests. Do they use a different test?

[JGD]

From a report in the New York Times:

Even as the White House released new details about the president’s condition on Sunday, it continued to withhold others, including when Mr. Trump had his last negative test for the coronavirus and his first positive one. Two administration officials speaking on the condition of anonymity acknowledged that he had an undisclosed positive result from a rapid test on Thursday evening after returning from a fund-raiser at his golf club in Bedminster, N.J. But he did not reveal it when he subsequently called into Sean Hannity’s Fox News show and, in a raspy voice, said he was still waiting for results.

Only after the television show did the results of another, more sophisticated PCR test come back confirming the positive reading, according to the officials, an account previously reported by The Wall Street Journal.

https://www.nytimes.com/2020/10/04/us/p ... e=Homepage

So in response to Mosy's - do they use a different test - the answer would be, "they use many".

It is of great interest also that Rome airport's 30 minute test which means passengers can leave terminals knowing whether they have a negative or positive result is being rolled out to regional airports in Italy.

Why is it our main stream media does not press HMG on why these rapid and virtually immediate turnaround tests are not available in the UK?

The tests themselves are not a barrier to the spread of Covid-19 but help greatly in any containment phase.

[stuart]

I think the answer is these rapid tests have, to date, been less accurate. There is a difference between someone passing through an airport who can be presumed negative until shown positive - and someone who is showing symptoms or been in close contact and should be assumed to be positive unless shown negative. The false negative problem.

This will change of course and we can presume the White House gate rapid tests are the same as airport tests and 'good enough' for those passing through who can to be stopped almost immediately. NB Trump's proper test took several hours to process so was of the more reliable type.

Stuart

[JGD]

Just watched HMG's Health Secretary Matt Hancock blurb on the floor of the House of Commons to MPs that the latest error "has not substantially changed" the government's assessment of Covid-19 and its impact, and "has not impacted the basis on which decisions about local action were taken last week".

The technical issue meant the daily case totals reported on the government's coronavirus dashboard over the past week have been lower than the true number.

It was caused by some data files exceeding the maximum size, and issues in transfer of data between NHS Test and Trace and PHE.

He said care homes, schools and hospitals have not been directly affected because outbreak control in these settings "does not primarily rely on this particular PHE system".

He acknowledged it as a "serious issue" that is being "investigated fully".

"This incident should never have happened. But the team have acted swiftly to minimise its impact," he said.

Microsoft's webpage help information describes the physical limitations of any modern version of an Excel workbook thus:

32-bit environment is subject to 2 gigabytes (GB) of virtual address space, shared by Excel, the workbook, and add-ins that run in the same process. A data model’s share of the address space might run up to 500 – 700 megabytes (MB), but could be less if other data models and add-ins are loaded.

64-bit environment imposes no hard limits on file size. Workbook size is limited only by available memory and system resources.

with

Total number of rows and columns on a worksheet 1,048,576 rows by 16,384 columns

So in a linear model each workbook in a spreadsheet could handle 1,048,576 records maximum capacity in the available rows.

There are many recorded instances where large institutions used legacy spreadsheets set up in 32 bit environments who also fell foul of the 2Gb file size limitation. These lessons were learned more than 10 years ago. Many were hit hard when the models being used simply chopped off all data that the spreadsheets could not process.

Everyone who worked with large scale SQL data extractions and transfers was subsequently trained to be alert to these limitations.

I think we are observing the replication of a fundamental problem that was well recognised and indeed Hancock might actually be telling us that HMG and its agencies "...knew about it in advance"".

But not those in charge of data management on this project it seems. The problem is compounded by the PHE developers choice of an outdated file format to manage the data storage and manipulation tasks - and thus lost tracts of data.

It is reported that daily figures for the end of the week were actually nearer 11,000, rather than the roughly 7,000 reported.

Keir Starmer describes this as a fiasco. He goes on to say "The Conservatives’ serial incompetence means we are losing control of the virus."

Hancock's shambolic denials that the error has made any significant difference to decisions increases the sense that PM Johnson and his cohorts would choose to magnify their attacks on those who dare conduct scrutiny upon HMG's mis-managed activities. Not to forget the PM's assertions that attacked the populace for not obeying rules.

Integrity there is none. Political, technical and managerial competence is thin on the ground. We are bereft of effective government.

[stuart]

It's worse that that JGD. It appears probable they were using XLS files (64K rows was as big as you would ever need in 1987), not the modern XLSX which can take a million. Spreadsheets, on a good day, are the IT equivalent of the back of an envelope. They are not databases. When mis-used as databases this sort of stuff happens. That's why NHSX says they mustn't be. Somebody didn't get the memo. Maybe it was sent via Excel.

More here: https://www.theregister.com/2020/10/05/ ... act_error/

As some wag noted on Hancock-ups: There's a sliding scale from a Hancock, through a Dido up to a full on Grayling. Because there's failing and then there's Grayling.

We were at least spared that.

Stuart

[JGD]

Simply unbelievable.

We thought "Yes, Prime Minister" had taught us all we had to know about HMG cock-ups.

How naïve. How little did we but know.

From Stuart's same link at The Register:

It added: "Of these, over 75 per cent (11,968) relate to cases that should have been reported between 30 September and 2 October" – meaning all the contacts of 3,873 people were left outside of the track-and-trace system for up to seven full days.

This comes just days after multiple users were erroneously told to "self-isolate" during a "test" of the much-delayed contact tracing app for England and Wales, NHS COVID-19, which was meant to launch on 1 June but instead launched on 24 September (Scotland and Northern Ireland already had their own COVID-19 awareness apps by this point).

Just two months back, Baroness Dido Harding, [Queen of Carnage sic text in strikethrough] head of the UK's COVID-19 Test and Trace organisation and former CEO of hacker-magnet ISP TalkTalk, was announced as head of the new National Institute for Health Protection, the agency that replaces Public Health England.

The test-and-trace programme in England was outsourced to Serco for £45m in June. The outsourcer is not responsible for the design or overall management of NHS Test and Trace, nor its IT systems. Nonetheless the outsourcer managed to leak the email addresses of around 300 human contact tracers, the people paid to find out who an infected person was in touch with, back in May.

Tories selected Mr Johnson as a brazen, self-serving election-winner who they could envisage as being unable to do anything wrong despite being dazzled - or blinded - by a long history of worrying deceits, lies and half-truths about which they could smile and assure themselves "He's a lad isn't he", "There's a bit of Churchill in him".

It has become evident he is unable to do anything right.

And as the journalist Tim Walker has tweeted this evening "There is an obvious difference between Churchill and Johnson. Churchill defeated Germany. Johnson defeated Britain."

https://twitter.com/ThatTimWalker/statu ... 77632?s=20

[JGD]

On two iPhone 7Plus devices located in our home my BH and I received a notification at around 22:00 on our Home screens.

The notification was fleeting and seemed to be informing or requesting of us a confirmation of a contact and seeking further info on duration and location. After a few seconds being displayed "on screen", both notifications vanished.

There is nothing within the app that makes the notification retrievable nor is there any specific warning of a "contact" incident having occurred. Nor is there any instruction(s) to follow.

It smacks of an error.

Did/has any one else encountered this Home screen notification? Validly or in error ?

You can review Exposure Notifications and Exposure Checks contained in the Exposure Logs that contain the secret matching key data and the number of transmissions that have been made to the NHS centre with a display of the Key Counts (presumably the number of other devices with which data has been exchanged) and a Matched Key Count (which is assumed to be the number of contacts made with devices where a positive test result for CV-19 has been recorded).

The number for Matched Key Counts is set to Zero in all my records.

My assumptions may seem broad - but I'm not sure how to respond.

Any other experiences or views?

[JGD]

Hmmm -further research reveals what is happening.

These are default messages from Apple and Google. There is no need to be concerned if you miss or overlook them. You're currently not able to turn off these default Apple and Google notifications. Important messages from the NHS COVID-19 app will always be visible to you from inside the app.

https://faq.covid19.nhs.uk/article/KA-01252/en-us

Fuller quote:

I have received a notification about “COVID-19 exposure logging”, “COVID-19 exposure notifications” or “Possible COVID-19 exposure” - what does this mean?
The NHS COVID-19 app uses Apple and Google Exposure Notifications. When exposure logging is switched on, Apple or Google may send you notifications. These may be called COVID-19 EXPOSURE LOGGING or COVID-19 Exposure Notifications. Sometimes it may seem as though these messages disappear, or you cannot click into them. These are default messages from Apple and Google. There is no need to be concerned if you miss or overlook them. You're currently not able to turn off these default Apple and Google notifications. Important messages from the NHS COVID-19 app will always be visible to you from inside the app.

Most of the notifications are just to remind you that the functionality is on and working. Others are labelled Possible COVID-19 Exposure. These messages do not come from the NHS COVID-19 app. The app uses a risk-scoring algorithm to filter out ‘false alarms’ based on distance and time. You only need to take action if you get a notification directly from the NHS COVID-19 app. Messages from the NHS COVID-19 app will not ‘disappear’ when you click them, you will be able to see the advice for you within the app when you open it.

So in short - no need to be concerned about the authenticity - if the message disappears and there is no corresponding Notification "inside" the app - the message should be disregarded.

And how many worried people will lose sleep on this one? Especially as no warning is issued in advance of the possibility of this happening.

[angela53]

Some employers, including providers of NHS services, are telling their staff to switch the app off when at work

[mosy]

Re NHS and Excel, according to several computer people, It's been well known for years (as JGD said) that Excel drops off data it can't handle once full, so Serco should have known it was tagging onto, i.e. accepting data from, a system with limitations, and that they should have made provision for this and that Dido Harding should have been informed by both sides of the issue and how it would be managed, which apparently didn't happen.

Seemingly, the parties did not talk to one another, hence numbers fell off the Excel screen unnoticed for days with no-one even knowing or thinking to check and pick them up. One said that it was no different from doing a simple tally when using any system, similar to cash in a till being the same total as that of sales entries. Most of us would know what check totals are, unless working for or contractually being paid for by the government it seems.

[JGD]

More incompetency of the worst kind.

Dido Harding as the person responsible for the management of this system, has had to acknowledge that its parameters were set incorrectly.

This error went undetected until the last week or so.

Consequently too few people were issued with notifications of contact when they should have been. Track, Test and Isolate has been rendered ineffective by this failure.

How much more insouciance from the Prime Minister must be tolerated before he acknowledges that he and his appointees are not up to leading the country during the pandemic.

[Robin Orton]

I agree. It's difficult to see how a Prime Minister and cabinet of such proved incompetence can survive. The Tory party must do its patriotic duty and boot them out forthwith.