Oh oh.

[cromwell]

NHS and IT aren't initials that sit comfortably together. The NHS Connecting for Health IT project was cancelled in 2011 having produced virtually the square root of nothing and having cost £10 Billion pounds.
This project is now used as a case study in poor Project Management!
(I was actually interviewed for a job on the project; it was a jaw dropping experience, we'll say).
Part of NHS IT called NHS Digital made headlines when they advised 10,000 patients to "shield", only to then realise that they had already died.
However the NHS firmly believes that it can create a contact tracing application.
Now, Apple and Google have already formed a partnership to create tools for contact tracing apps that would be available to third parties, like the NHS.
But no, the NHS want to do it themselves.
In the Daily Telegraph yesterday they said they were only "two or three weeks" away from rollout.
Well, I sincerely doubt that.
However quickly you develop the software there are a million and one things that can can wrong. Firstly you have to test the software and three weeks is no time at all to write it, let alone test it properly.
The NHS want to store all the data on a central server. Do they have a spare one lying around big enough to hold all the data? How do they propose to link all the smartphones to the server? Will the network be resilient? How are they going to manage backups and restores if necessary?

In my opinion they have precisely zero chance of getting all that up and running reliably in three months, let alone three weeks.

You don't re-invent the wheel and Apple and Google have a head start on the NHS in this field.

I do hope that I am wrong, but experience says that I won't be.

[Kaz]

Mick would heartily agree with you on this!

[TheOstrich]

And apart from the technology issues, are we going to be "forced" to download this app on our smart phones? Will it be an offence to leave home without it switched on?

Our society is nothing like as disciplined as South Korea where they seem to have had some "success" with it. I really cannot see how this contact tracing move can be anything remotely near effective without compulsion.

[Suff]

Cromwell, as I understand the situation, Apple and Google want the thing anonymised and informational to the app user, I. E. Telling the user to get themselves tested because they are at risk.

France, the UK and a few other countries wand the data held centrally so their health service can enforce both testing and isolation in the way SK does it.

There is a problem with creating your own app and not with Apple, especially, because it won't run as part of the system in the background and will, as a consequence, suck the battery flat if constantly on. This is a bit easier with Android but not that much.

Apparently the NHS app is supposed to check, offline, daily, then report.

The NHS have more chance of making a chimps tea party look like dinner with the queen than they do of making a success of this. The NHS has a lot of experience making IT chimps tea parties though..

They have no chance of doing this. I am reminded of VW's current ongoing problems with the ID.3 software. Every test iteration they add at least 250 new bugs... They only have 35,000 of these vehicles right now.

They might surprise but their track record. Is 100% failure in this kind of situation.

[Workingman]

I really cannot see how this contact tracing move can be anything remotely near effective without compulsion.

Nor me, and I am one of many who is following the rules but will not comply forever. For the present I will comply, but there might come a time when i say "Sod it, I am off to do my own thing." I won't be alone.

[cromwell]

The only chance they would have of doing it is if they had been working on it for months and months already.
But as Os and WM have pointed out, a tracing app only works if you take your phone with you everywhere!
Done in three weeks? No. Done in three months? No.
"Moving to rollout" is one of those phrases which sound good, but what does it mean? We will finish writing the code in three weeks?
Anyone who has written software, tested software or run an internal development project knows that the NHS timescale is rubbish, and blatant rubbish.
I suppose it comes back to our politicians having no real world of anything much apart from politics. They don't realise that they are listening to rubbish.

The front line staff of the NHS are doing great things. But the big NHS quangos are determined that the solutions to the Covid crisis have to come from them. It has to be them that does the testing, provides the tracking app, etc.

Because if they don't people will ask (entirely correctly) what their purpose is and do we actually need them?

Whether this is the best way of handing the crisis is debatable.

[Suff]

As I said earlier WM, don't like the app, don't take your phone or leave it switched off or in flight mode.

What bothers me is BT is on and broadcasting the whole time. I do Not allow BT on unless I am actively using it.

Crommers, they could be moving to Alpha test rollout, but even that would be hopelessly optimistic.

As you say, anyone who has actually done this kind of work knows it is total fantasy land. Even if you don't ask it is not hard to ask someone who does know. Even employing an independent consultant for a week to review is hardly ground breaking....

[Workingman]

I'll try a different tack.

At Uni I hated coding with a passion but fortunately there was a repository of coding blocks that we could splice into our own work. They were for the more mundane stuff like turning conveyors on or off, that sort of thing. They were time savers to allow us to do our own specific things. So that's one thing.

The other is that tracking apps have been around for ages. Some are cell / mast based where others us GPS. They come with a whole range of functions and many are available through the app stores. Not all phones have the GPS function so those apps can probably be left to one side for now.

I cannot see any problem with the NHS taking an already working app, tweaking it a bit, and then putting it an NHS wrapper. It would still need testing to the nth, but it would be a way of cutting the time down by many months. However, it has to be done right and not half-cocked and my confidence in that is low.

My big concern, apart from the personal intrusion, still remains that such a thing will be obsolete before it hits the streets and could of itself be a danger. If R=>1 the numbers being tracked and traced will become so big not everyone could be contacted and tested. All it will be doing is confirming what we already know through other means, that the virus is wild. That in itself could create a panic on a scale not seen before. Just imagine hundreds of thousands of us getting a text every day: "Records show that in the past 24 hrs you have been in contact with x coronavirus carriers - please isolate for 14 days. Thank You, HMG."

[Suff]

My reading says that most of these apps intend to use Bluetooth to verify the proximity to another app user. When one app user is diagnosed, all app users which came into proximity will be notified.

My BT is always off and will remain that way. Also just because you were in BT range of an infected user does not mean you got it. BT range is 10m and in building location is impossible.

There is no tracking app, currently in existence, designed for this job. Apple ang Google and tracking specialists have the best chance of making a go of it, I can't see the NHS having something which is better than nothing any time soon.

[Workingman]

Yes, Bluetooth between devices, but then that info has to be sent to err, wherever, so that the tracking and tracing and testing can be done. I also have it switched off, I'm more a cable guy when it comes to transfers.

I know that there is currently no app for the job, but one of the existing ones could be tweaked in extremis. Not that I'll be buying in as I think it could be more of a danger than it's worth, in more ways than one. See my last post.