Test And Trace.....

[ToraToraTora]

https://www.bbc.co.uk/news/uk-52829357
Why are they using a manual system rather than the software?

[Gromit]

TTT,
When we last discussed the app the Nationwide launch date was 1st June. That has now changed to
// Downing Street would only commit to saying it aims "to roll out the app across the country in the coming weeks" //

Apparently, they have been working on a second app which uses the Google/Apple APIs, in case the IoW trial fails.

It looks like the UK will be one of the last countries to introduce a Corona app. Most countries had theirs weeks ago.

[Gromit]

The whole point of a trialling a Beta version of an app, is to identify problems and fix them. It looks as though that has proven more difficult than the Government expected.
The App was supposed to be the centre piece of Track & Trace, with feet on the ground complementing it.
The fact that Track & Trace has now commenced without the app doesn’t bode well.
Government commissioned software that don’t work has blighted Conservative and Labour Governments for decades. It would be unfortunate if such a critical app for the health of the nation, ended up as a fiasco.

[ToraToraTora]

yes but if, as you said last week, those that had the app outside of the IOW only had the icon, how did that have any effect on the trial?

[Gromit]

The Government thought they had 60,000 Isle of Wighters testing their app, but instead they only had two thirds of that. The downloads from the rest of the UK were providing no data.
For the app to be effective they need 60% of the population to have it, they have nearer 30%. When the UK mainland downloaders were taken into account they had 40,000 testers in the IoW from a population of 140,000.

[New Judge]

//...those that had the app outside of the IOW only had the icon, how did that have any effect on the trial?//

Because one of the stats relied upon is the %age of the population that is using the app. The government said x% of the 140,000 on the IoW had downloaded it. If some of the downloads are not on the IoW and so not part of the 140k, then the %age is wrong. I don't know how critical the usage is, but I remember reading somewhere that it needs 60% usage to be effective. If they don't know what %age the usage is they cannot judge its effectiveness.

[woofgang]

puzzled I am not on the vulnerable list and can get sufficient delivery slots.

[ToraToraTora]

fair enough judge but it's a simple arithmetic task to exclude the non IOW ones from the whole and clean up the stats.

[Gromit]

Another snippet I have just found out.
The woman in charge of all that data the app gathers is Dido Harding. She was a University chum of David Cameron. She is now head of NHS Improvement, but you may remember her when she was CEO of TalkTalk when millions of their customers had their banking details hacked from TalkTalk.

https://www.campaignlive.co.uk/article/talk-talk-boss-dido-hardings-utter-ignorance-lesson-us/1370062?src_site=marketingmagazine

[Gromit]

TTT,

You obviously didn’t understand my earlier answer.

They needed 60% of 140,000.
But they only got 30%.

[Gromit]

Ignore that. ^

They wanted 84,000 downloads but got half of that.

[ToraToraTora]

yes but I was responding to the judge's post saying that they could not know the percentage because the non IOW ones were skewing it. The app knows the location so it should be able to exclude the non IOW ones and ring fence the trial. Yes I take your point that not enough downloaded it anyway.

[diddlydo]

Apparently ONS reported that 79% of those who tested positive had no symptoms - how/why did they get a test without symptoms?

[Gromit]

TTT,
They got no data from your phone, and the thousands of other mainlanders. So they know how many IoW testers they have. The number is not skewed, it is just not as many as they originally though, and about half of what they wanted.
National roll out has now been delayed for several weeks, we will have a better idea then if it works or not.

[ToraToraTora]

ok but according to the judge the figures were effected by the non IOW apps. So perhaps you and the judge can thrash this out. If the non IOW apps had no bearing on the data how can they be to blame for the failure as stated here in the Telegraph?
https://www.telegraph.co.uk/news/0/isle-wight-contact-tracing-trial-undermined-people-uk-mainland/
quoted earlier by danny?

[Gromit]

I’m sure NJ and I agree.
The data wasn’t affected. The amount of data was. They are getting less data than they originally expected.

[ToraToraTora]

judge said: "If some of the downloads are not on the IoW and so not part of the 140k, then the %age is wrong"!

[New Judge]

I think we do, but I'm not sure that they can isolate the non-IoW downloaders. They may get no data. But they'd get none from mine if I downloaded the app because my phone is rarely on. There may be people on the IoW who have downloaded it but behave similarly to me. Frankly it's all a bit hit and miss. When they devise these systems they make an assumption that everybody behave according to their game plan. But they don't.

[New Judge]

Just to be clear, initially I assumed that they simply counted the number of downloads. If that's the case and it is known that some are non-IoW, then unless some adjustments are made to the numbers the percentage use on the IoW (which is what's required) will be wrong. So the next question is how do they identify the non IoW downloads? If it's simply by counting the ones they get no data from that is not reliable for the reasons I stated.

[THECORBYLOON]

DIDDLY, folk have been invited to take a test. It's part of an ONS survey of the general public.

https://www.ons.gov.uk/surveys/informationforhouseholdsandindividuals/householdandindividualsurveys/covid19infectionsurveycis#about-the-study