Coronavirus outbreak

[Wobblers]

Where "data file sizes being too big" means "Excel spreadsheet reached maximum number of columns", if the Twitter rumours are to be believed. So far, the primary source seems to be the Daily Mail, so I'm not confident to call them that stupid yet. It's pretty stupid not to have any upload verification and error handling on something as important as this, though.

If this is true, it's not only world-beating, but also self-defeating:

View: https://twitter.com/standupmaths/status/1313055411285774336

I can well believe Excel being behind this. After all, it's used everywhere, being the business tool of choice, public and private sector alike. Even when it's not the appropriate tool, it's still the office equivalent of a hammer. Every organisation you care to name will have an immense number of processes shoved into umpteen Excel spreadsheets (though you no doubt are all too aware of this!).

Still, it could have been worse: they could have decided to spend eleventy squillion on some SAP implementation running on Oracle using an unsecured S3 bucket....

 

[Rezillo]

Keep an eye on this. El Reg is usually the first place for those involved in this kind of incident to leak to.

https://www.theregister.com/2020/10/05/test_and_trace/

 

[Ming the Merciless]

I can well believe Excel being behind this. After all, it's used everywhere, being the business tool of choice, public and private sector alike. Even when it's not the appropriate tool, it's still the office equivalent of a hammer. Every organisation you care to name will have an immense number of processes shoved into umpteen Excel spreadsheets (though you no doubt are all too aware of this!).

Still, it could have been worse: they could have decided to spend eleventy squillion on some SAP implementation running on Oracle using an unsecured S3 bucket....

They probably have the spreadsheet in an unsecured S3 bucket for “efficiently” sharing the data with.....everyone.

 

[Wobblers]

She has a strong background in managing biotechnology research, development and delivery. Yes, that is in private industry but she will not be actually developing the drugs.....there are experts for that. Her responsibility will be involvement in the development and delivery programme, under the control and management of the government.

n

None of which indicate that she's the appropriate choice to organise a very large vaccination programme. What she's saying already worry me. Smallpox wasn't eradicated by vaccinating everyone - there just wasn't enough of the vaccine for that. It was done by "barrier vaccination": vaccinating those who had been in contact with infected people. It worked.

We will likely end up in the same place with regards to CV19: there won't be enough vaccine to go around, at least initially. CV19 can be a serious illness, even in those under 50. Consider that it's now thought that 10-20% will suffer "long Covid" and experience months of debilitating fatigue. This includes those who are the most economically active - the economic consequences alone are likely to be significant. I would think that barrier vaccination would be a more successful strategy than limiting it to the most vulnerable - not least because it will probably still take many months to vaccinate even that group. (Of course, this assumes that any vaccine builds up resistance quickly.)

Put simply, I'd be happier to see an actual expert in the field to be in charge of the vaccination programme.

 

[stowie]

Suggestions elsewhere that it was the column limit that was reached, as the genius who designed the spreadsheet thought it would be a good idea to have a column for each case, rather than a row.

Hard to believe, if that's true.

That is just weird. Why would you do that?!

 

[RoadRider400]

Have there been any figures released for England today, or have I missed it?

The tombola machine has broken unfortunately.

 

[MarkF]

North of 12000 cases today, and apparently not inflated by the Excel fark up.

That's really bad, looks like doubling every 10 days or so still. I think the estimate for the peak of the first wave was 100,000 per day, which translated to 1000 deaths per day. We'll be back there by the end of the month at this rate (three doublings).

[Please, someone tell me I've got this wrong because this seems awfully grim news to me]

And beyond an expected seasonal rise (like flu) how do these "cases" correlate to hospitalisations, ICU admissions and deaths?

The PCR test inventor (Kary Mullis) said it was not to be used for medical diagnosis, only research as it was so unreliable. The false positives in China in March were 80%. Our government is using a deeply flawed system from which to produce spurious "cases" and then using those figures to determine policy/strategy to protect us from a pandemic that may have been over months ago. A huge % of these new and "missed" cases will be kids, college and uni kids, I'll stick my neck out and say not 1 will die and I'd be surprised if 1was hospitalised, yet they'll prove happy hunting grounds for "case" seekers. It's not logical, an example being the 3 UK students in Italy, stuck in quarantine for 2 months and with no end in sight because they keep testing positive, for 2 months!!! Or Spains sudden drop in "cases" when they lowered the PCR cycle threshhold and double tested. Cases fall, restrictions are working, cases rise, need more restrictions (= more cases), as one journalist said last week "It's like ducking witches".............

Be interesting to see what the government comes up with on Friday when it has to produce evidence on which it's restrictions were based, now that the crowd funded legal injunction must be heard next week.

Great Barrington Declaration.

 

[roubaixtuesday]

how do these "cases" correlate to hospitalisations, ICU admissions and deaths?

Deaths are following behind as would be expected, also doubling every ten days or so.

That will take us into the hundreds every day by the end of the month.

 

[tom73]

Other than political baggage Scotland's app is basically the same as ours. Look's to have the same issues with the blue tooth too.
https://www.bbc.co.uk/news/uk-scotland-54418278

 

[icowden]

That is just weird. Why would you do that?!

To me that sounds unlikely.
The more plausible issue is the row limit. As to "why excel" well...

NHS Labs use lab software to process the results of screening tests. In order to send those results back to the organisation that requested the screening, there are 2 main methods:-

1) HL7 / FIHR clinical messaging
2) A CSV file

The first option is obviously the best but you need to have quite a few things in place. Firstly you need to be a registered organisation and have an endpoint into the N3 network. This is a private Network used by the NHS to ensure that your health records remain confidential. It's not quick or easy to set up. Then you need to have software that can receive the HL7 / FIHR messaging and convert that back into relational data, which you can then use for analysis. Then you need a contractual agreement and data sharing agreement with the labs that are going to send the results to you.

The second option is what tends to get used when it has been determined that option one is too difficult / costly / time consuming / not yet available. So Serco Test and Trace is almost certainly receiving CSV files.

From what the Grauniad has reported it seems likely that instead of importing the received CSV into a suitable database system, results seem to have been copied and pasted into a master Excel doc, and then reports have been driven off that. So when the Excel doc ran out of rows, data was lost.

 

[mjr]

The PCR test inventor (Kary Mullis) said it was not to be used for medical diagnosis, only research as it was so unreliable. [...]

Stop spreading fakebook fakes, please! This one has been factchecked and is easy to find:
https://www.reuters.com/article/uk-factcheck-pcr-idUSKBN24420X

Your shocking conclusions cannot logically follow from this because it is false.

 

[fossyant]

Local news did a little digging into the super high infection rates for Manchester. Close to 500 per 100,000. Student population is more like 3,000 per 100,000 - yes we expected that. It is still only 3%

 

[oldwheels]

Other than political baggage Scotland's app is basically the same as ours. Look's to have the same issues with the blue tooth too.
https://www.bbc.co.uk/news/uk-scotland-54418278

So out of million people BBC Scotland manage to find one problem with someone who will probably turn out to be a political activist. They have form.

 

[Ming the Merciless]

That is just weird. Why would you do that?!

Because they couldn’t find any real IT people .

 

[tom73]

So out of million people BBC Scotland manage to find one problem with someone who will probably turn out to be a political activist. They have form.

I have know idea all I was pointing out is which ever app you use that is not centralised. They all run off the Apple/Google close contact program and are basically the same. So they will all have issues around false positives.