I saw someone point out that Sweden, with coronavirus, is in fact two stories: - Stockholm - Rest of Sweden
... which marches well with the suggestion that the local population density really has to be taken into account.
As of today, Stockholm passed the figure of 500 deaths per million. (1,192 deaths in a population of 2.377 million).
The rest of Sweden has a figure of 122 deaths per million (960 deaths out of 7.85 million)
Quite a difference.
If you applied that to the UK, and exluded London, then that would be interesting.
Looking at mortality per 100 000 population, all of London is at the top end, but very top is the Black Country and Birmingham. In England at least it seems more evenly spread. Leics continues at about half the national rate, a quarter of the West Midlands.
There are some anomalies. Bristol area looks quite low for its degree of population density, as does Stoke and Staffs.
A lot of the Home Counties has high rates for its population density. Commuters I suppose.
The very lowest areas are places with very little commuting, Devon, Lincs, Dorset Humber etc. I note Hartlepool is low, but Newcastle and Sunderland high.
Trains, trams and buses must have been teeming with the virus. Take care Sunil!
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago the idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available. But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago I idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available.
But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
Sure, but they've come about because the capabilities are now there, and in areas where there is a lot of data to train the model on. The issue is whether there is sufficient available (relevant, complete and accurate) data on previous pandemics, that you could use to train the model on.
This isn't my field, so very happy to be educated, but it's not immediately obvious to me what the past dataset looks like.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
My son's school did an interesting questionnaire recently about fees for this term where there is some teaching online. From a range of options 40 odd per cent were content to pay the existing fees. I suspect that they were a combination of the rich and those who are not financially affected (there are a lot of doctors kids at the school). The next largest was those seeking a 20% discount which included me. Some went for smaller increases and about 70 are seeking bursaries on the basis that they can no longer afford the fees at all.
Which shows that the economic impact of this is currently very uneven and it is very likely to become even more so going forward.
If you don't mind me asking will your son's school survive? I've heard predictions that private schools and universities are due to take a thumping because of collapse in numbers of international students. Scottish Uni sector particularly vulnerable because domestic students don't pay fees - and little wriggle room since Salmond caused the words of his pledge to be carved into a boulder.
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago I idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available.
But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
Sure, but they've come about because the capabilities are now there, and in areas where there is a lot of data to train the model on. The issue is whether there is sufficient available (relevant, complete and accurate) data on previous pandemics, that you could use to train the model on.
This isn't my field, so very happy to be educated, but it's not immediately obvious to me what the past dataset looks like.
Ok, so Gaussian Processes. Again they are something that was first suggested in the 70s. The whole points of these methods is to make predictions with limited data sets...so no throw a billion images at a computer and he shocked it does a pretty good job of recognising cats.
But the limiting factor wasn't really just a computational power, there were some fundamental mathematical issues about how to actually take what seems like a nice idea, with some very attractive properties, and turn it into something useful.
That work has only really occurred in the past 15 years or so. Sure having bigger computational power has certainly driven interest, but a number of significant advances have occurred e.g GPLVM for probabilistic dimensionality reduction, which in turn sparks lots of new avenues / techniques.
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago the idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available. But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
What would the application of these techniques to this problem have achieved? I have no idea what the point of the IHME model is - it's clearly not very useful to produce such wide ranges of deaths.
But the LSHTM/Imperial ones showed lots of people dying and by all accounts pushed the govt to action. The reason they didn't show this earlier seems much more to be we didn't have enough data on key elements such as fatality rate, transmission rate, how many cases there actually are etc. I don't think fancy techniques can get around that.
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago the idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available. But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
What would the application of these techniques to this problem have achieved? I have no idea what the point of the IHME model is - it's clearly not very useful to produce such wide ranges of deaths.
But the LSHTM/Imperial ones showed lots of people dying and by all accounts pushed the govt to action. The reason they didn't show this earlier seems much more to be we didn't have enough data on key elements such as fatality rate, transmission rate, how many cases there actually are etc. I don't think fancy techniques can get around that.
I wasn't saying deep neural networks were the answer, rather giving an example of large scale advancements in ML.
As suggested down thread, Gaussian Processes are made for these kind of problems. Gaussian processes are often built on the infrastructure of something like Tensorflow, which is the open source library Google uses for deep learning.
I think there is a fairly good reason why the Royal Society have brought together a load of new people from fields well outside the narrow range that was being used and who are now working these problems.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
There's going to be talk of almost nothing else but this target, for the next seven days.
In the eyes of the media and opposition, meeting (or missing) that target will define the whole government response to the pandemic.
You seem to overlooking the rather important fact that we are all talking about that figure because it was the the government that announced that target in the first place. Can't really blame anyone for criticising the government if it fails to meet a target it set for itself.
...The emergence of a pandemic affecting the respiratory system can result in a significant demand for face masks. This includes the use of cloth masks by large sections of the public, as can be seen during the current global spread of COVID-19. However, there is limited knowledge available on the performance of various commonly available fabrics used in cloth masks. Importantly, there is a need to evaluate filtration efficiencies as a function of aerosol particulate sizes in the 10 nm to 10 μm range, which is particularly relevant for respiratory virus transmission. We have carried out these studies for several common fabrics including cotton, silk, chiffon, flannel, various synthetics, and their combinations. Although the filtration efficiencies for various fabrics when a single layer was used ranged from 5 to 80% and 5 to 95% for particle sizes of <300 nm and >300 nm, respectively, the efficiencies improved when multiple layers were used and when using a specific combination of different fabrics. Filtration efficiencies of the hybrids (such as cotton–silk, cotton–chiffon, cotton–flannel) was >80% (for particles <300 nm) and >90% (for particles >300 nm). We speculate that the enhanced performance of the hybrids is likely due to the combined effect of mechanical and electrostatic-based filtration. Cotton, the most widely used material for cloth masks performs better at higher weave densities (i.e., thread count) and can make a significant difference in filtration efficiencies. Our studies also imply that gaps (as caused by an improper fit of the mask) can result in over a 60% decrease in the filtration efficiency, implying the need for future cloth mask design studies to take into account issues of “fit” and leakage, while allowing the exhaled air to vent efficiently. Overall, we find that combinations of various commonly available fabrics used in cloth masks can potentially provide significant protection against the transmission of aerosol particles...
I saw someone point out that Sweden, with coronavirus, is in fact two stories: - Stockholm - Rest of Sweden
... which marches well with the suggestion that the local population density really has to be taken into account.
As of today, Stockholm passed the figure of 500 deaths per million. (1,192 deaths in a population of 2.377 million).
The rest of Sweden has a figure of 122 deaths per million (960 deaths out of 7.85 million)
Quite a difference.
If you applied that to the UK, and exluded London, then that would be interesting.
Looking at mortality per 100 000 population, all of London is at the top end, but very top is the Black Country and Birmingham. In England at least it seems more evenly spread. Leics continues at about half the national rate, a quarter of the West Midlands.
There are some anomalies. Bristol area looks quite low for its degree of population density, as does Stoke and Staffs.
A lot of the Home Counties has high rates for its population density. Commuters I suppose.
The very lowest areas are places with very little commuting, Devon, Lincs, Dorset Humber etc. I note Hartlepool is low, but Newcastle and Sunderland high.
Trains, trams and buses must have been teeming with the virus. Take care Sunil!
FPT: 8 weeks and a day since I visited Central London (Liverpool Street station - um, to trainspot a new Class 745 intercity unit ) 8 weeks since I rode the Tube anywhere (Gants Hill to Epping via Hainault and Woodford and back) 7 weeks and 6 days since I flew from Luton to Aberdeen (to see my brother, his wife and their newborn son!) 7 weeks since I did the train from Aberdeen to Inverness and back 6 weeks and a day since I rode any plane, train or bus at all (plane from Aberdeen to Southend, two trains from Southend Airport to Romford via Shenfield, bus from Romford to Gants Hill) 5 weeks and 6 days since I visited a really big Sainsbury's 5 weeks and 2 days since I did any shopping at all (local shops/pharmacy) - though I also visited Valentines Park
2:55PM Is there now some evidence that this thing peaks after 60 days and burns out after 120?
Hmm.
*****
Could that be right or is it too much to ask?
Ok I’m off for my legal daily
Well 120 days is around 4 months
Do the viruses have little clocks in them? Surely it's more likely that the falling off is more down to government and people's raction to it?
Or is it possible that enough of us are exposed sufficiently to get our immune systems working on it even if the dose is not sufficient to actually give us it? The way SARS and H1N1 largely fizzled out was hard to explain.
It's possible but we don't have the testing data to confirm the percentage infected. I suspect that it's in single figures. Sars - "twenty-first century science played a relatively small role in controlling SARS; nineteenth-century techniques continued to prove their value”. People were isolated and contact traced. It was less infectious and showed up sooner when people were which helped enormously. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636331/
I wasn't saying deep neural networks were the answer, rather giving an example of large scale advancements in ML.
As suggested down thread, Gaussian Processes are made for these kind of problems. Gaussian processes are often built on the infrastructure of something like Tensorflow, which is the open source library Google uses for deep learning.
Right but whichever technique you choose - I don't see how that helps in a scenario when you don't know basic information. If you don't know how many cases there are, fatality rate, force of transmission - then it's unclear to me what improvement you'd get from new modelling techniques.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
I don't know Gordon Brown but can imagine him doing that. Reflects rather well on him. Makes me smile anyway.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
It would be nice if all the Malcolm Tucker stuff could stop just for the duration of the pandemic.
2:55PM Is there now some evidence that this thing peaks after 60 days and burns out after 120?
Hmm.
*****
Could that be right or is it too much to ask?
Ok I’m off for my legal daily
Well 120 days is around 4 months
Do the viruses have little clocks in them? Surely it's more likely that the falling off is more down to government and people's raction to it?
Or is it possible that enough of us are exposed sufficiently to get our immune systems working on it even if the dose is not sufficient to actually give us it? The way SARS and H1N1 largely fizzled out was hard to explain.
It's possible but we don't have the testing data to confirm the percentage infected. I suspect that it's in single figures. Sars - "twenty-first century science played a relatively small role in controlling SARS; nineteenth-century techniques continued to prove their value”. People were isolated and contact traced. It was less infectious and showed up sooner when people were which helped enormously. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636331/
I think a lot of the discussion around recent advances in computer modelling may be a little misguided. The key advance in the last decade or so (well, in my field, anyway) has been around processing speed and efficiency, which allows us to work on bigger datasets and do more things with them.
I beg to disagree.
For example, yes, about 10 years ago I idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available.
But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
Sure, but they've come about because the capabilities are now there, and in areas where there is a lot of data to train the model on. The issue is whether there is sufficient available (relevant, complete and accurate) data on previous pandemics, that you could use to train the model on.
This isn't my field, so very happy to be educated, but it's not immediately obvious to me what the past dataset looks like.
Ok, so Gaussian Processes. Again they are something that was first suggested in the 70s. The whole points of these methods is to make predictions with limited data sets...so no throw a billion images at a computer and he shocked it does a pretty good job of recognising cats.
But the limiting factor wasn't really just a computational power, there were some fundamental mathematical issues about how to actually take what seems like a nice idea, with some very attractive properties, and turn it into something useful.
That work has only really occurred in the past 15 years or so. Sure having bigger computational power has certainly driven interest, but a number of significant advances have occurred e.g GPLVM for probabilistic dimensionality reduction, which in turn sparks lots of new avenues / techniques.
Thank you. Like I said, this sin't my field, so I can't immediately see why this is a step change from traditional PCA approaches, but it's given me some more lockdown reading at least.
"More than 2.6million people in London may have already caught the coronavirus and recovered from it, data suggests.
Early results from antibody surveys - which reveal how many people have had the illness already - suggest the true death rate for COVID-19 may be anywhere between 0.1 and 0.6 per cent."
While I do think Mike's right that public support remains with the lockdown- despite the frothing of certain online elements- I don't think you should read these polls like election polls. Majorities don't matter here- if 30% of people decide to ignore the lockdown that's a huge problem.
FPT anabobazina posted this interesting remark 2:55PM Is there now some evidence that this thing peaks after 60 days and burns out after 120?
Hmm.
*****
Could that be right or is it too much to ask?
Ok I’m off for my legal daily
Well 120 days is around 4 months
Do the viruses have little clocks in them? Surely it's more likely that the falling off is more down to government and people's raction to it?
Or is it possible that enough of us are exposed sufficiently to get our immune systems working on it even if the dose is not sufficient to actually give us it? The way SARS and H1N1 largely fizzled out was hard to explain.
It's possible but we don't have the testing data to confirm the percentage infected. I suspect that it's in single figures. Sars - "twenty-first century science played a relatively small role in controlling SARS; nineteenth-century techniques continued to prove their value”. People were isolated and contact traced. It was less infectious and showed up sooner when people were which helped enormously. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636331/
Defoe understood the Covid problem back in the seventeenth century.
Here also I ought to leave a further remark for the use of posterity, concerning the manner of people's infecting one another; namely, that it was not the sick people only from whom the plague was immediately received by others that were sound, but the well. To explain myself: by the sick people I mean those who were known to be sick, had taken their beds, had been under cure, or had swellings and tumours upon them, and the like; these everybody could beware of; they were either in their beds or in such condition as could not be concealed.
By the well I mean such as had received the contagion, and had it really upon them, and in their blood, yet did not show the consequences of it in their countenances: nay, even were not sensible of it themselves, as many were not for several days. These breathed death in every place, and upon everybody who came near them; nay, their very clothes retained the infection, their hands would infect the things they touched, especially if they were warm and sweaty, and they were generally apt to sweat too...
While I do think Mike's right that public support remains with the lockdown- despite the frothing of certain online elements- I don't think you should read these polls like election polls. Majorities don't matter here- if 30% of people decide to ignore the lockdown that's a huge problem.
In Bridgend today traffic was heavier than most Fridays. Tesco Extra car park was almost full and the queue waiting to go into Lidl at 10.30 was thirty deep.
On the thread header, did they control for ownership of a garden?
I am only half joking, but seems to me there would be a lot less support for lockdown if it was pouring with rain every day. Many are getting, or will get, 80% salary to sit in their gardens. Maybe I am too cynical.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
I don't know Gordon Brown but can imagine him doing that. Reflects rather well on him. Makes me smile anyway.
Yes, he was the despair of his handlers. Tony would have finessed it - "I'm grateful to my honourable friend for his interest in this important issue, and look forward to briefing him as matters progress." Gordon had no patience for such arts.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
I don't know Gordon Brown but can imagine him doing that. Reflects rather well on him. Makes me smile anyway.
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
I don't know Gordon Brown but can imagine him doing that. Reflects rather well on him. Makes me smile anyway.
Yes, he was the despair of his handlers. Tony would have finessed it - "I'm grateful to my honourable friend for his interest in this important issue, and look forward to briefing him as matters progress." Gordon had no patience for such arts.
I've a sense that for all his disabling flaws as a politician he will be regarded as one of our better PMs in future years.
I really do not see the problem and if they are softball what about all those a PMQs from his side
Journalists just confirming why they are not fit for purpose or covid reporting
Good to see they are not hoodwinking the public
Exactly right, Big G. Get rid of the journalists and have the government asking itself questions every afternoon at 5pm.
Why not have a different lot of journalists every day ?
Or a random selection of questions from professional bodies or 'ordinary people' ?
Questions from ordinary people? I suspect you'd be praying for the journalists back within five minutes of the first event.
I think the Freakonomics guys have a thing about how the role of the press is misunderstood by intelligent people. It's not to educate the masses on the details of the complex issue of the day; it's primarily to get some sort of flavour across to the average person on the street, who wants to keep up to date but hasn't the time or patience to go and do the proper research themselves.
Okay, back from an hour's walk around the block after 25 days cooped up inside. Now I rememberer what fresh air is like, albeit breathing through a medical mask.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
There's going to be talk of almost nothing else but this target, for the next seven days.
In the eyes of the media and opposition, meeting (or missing) that target will define the whole government response to the pandemic.
You seem to overlooking the rather important fact that we are all talking about that figure because it was the the government that announced that target in the first place. Can't really blame anyone for criticising the government if it fails to meet a target it set for itself.
Why? What sort of world would it be if all targets were hit? Wouldnt the targets then be worthless?
There are plenty of scenarios where "stretched" targets are better than "achievable" targets. Even achievable targets should be missed a fair proportion of the time, otherwise they were unlikely to be needed in the first place.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
My son's school did an interesting questionnaire recently about fees for this term where there is some teaching online. From a range of options 40 odd per cent were content to pay the existing fees. I suspect that they were a combination of the rich and those who are not financially affected (there are a lot of doctors kids at the school). The next largest was those seeking a 20% discount which included me. Some went for smaller increases and about 70 are seeking bursaries on the basis that they can no longer afford the fees at all.
Which shows that the economic impact of this is currently very uneven and it is very likely to become even more so going forward.
Most English schools are voluntarily reducing fees. Prep schools around 15-20%, public schools more (Eton 30% for example).
Private schools are going to have a total nightmare. They are still incurring the vast majority of their costs during lockdown, lessons-from-home technology is an additional cost, and when lessons resume their variable costs are pretty much all staff salaries.
Most private schools (outside the top 'public' schools) operate as charities and don't have a lot of profit margin, savings or endowments.
Why would the top public schools not be charities? Eton and Winchester, frexample, are, and I'm guessing everyone else is.
Badly phrased on my part. Outside the top schools, few have big savings and endowments. They are pretty much all charities.
I've just discovered that you can't always rely on experts. This article, by a professor of urban studies, informs us that England is not as densely populated as the Netherlands. In fact England is more densely populated than the Netherlands. The article is using UK data not English data despite referring to England specifically. (I was trying to find information on the population of England and the Netherlands and this article was one of the first to come up on a search).
I've just discovered that you can't always rely on experts. This article, by a professor of urban studies, informs us that England is not as densely populated as the Netherlands. In fact England is more densely populated than the Netherlands. The article is using UK data not English data despite referring to England specifically.
There was a fascinating map posted a couple of days ago. What I hadn't realised is that there isn't anywhere in Germany as underpopulated as the highlands.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
Yes. Many people are enjoying it. And with most of those who aren't, or who are very concerned about the economy, the message that it is saving lives RIGHT NOW will still be in the box seat.
Quite. At the moment it is uneconomic to make anyone redundant when you can furlough them. I’ve been advising my employer clients that it’s pointless making anyone redundant at the moment. Many are just waiting until July to do it.
Say you furlough somebody who is on £120,000 per annum.
Government pays £2,500 of their £10,000 per month salary.
You don't top up. So they take a £7,500 per month cut.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
My son's school did an interesting questionnaire recently about fees for this term where there is some teaching online. From a range of options 40 odd per cent were content to pay the existing fees. I suspect that they were a combination of the rich and those who are not financially affected (there are a lot of doctors kids at the school). The next largest was those seeking a 20% discount which included me. Some went for smaller increases and about 70 are seeking bursaries on the basis that they can no longer afford the fees at all.
Which shows that the economic impact of this is currently very uneven and it is very likely to become even more so going forward.
Most English schools are voluntarily reducing fees. Prep schools around 15-20%, public schools more (Eton 30% for example).
Private schools are going to have a total nightmare. They are still incurring the vast majority of their costs during lockdown, lessons-from-home technology is an additional cost, and when lessons resume their variable costs are pretty much all staff salaries.
Most private schools (outside the top 'public' schools) operate as charities and don't have a lot of profit margin, savings or endowments.
But right now a lot of their staff are furloughed, they are not incurring heating bills, etc, not using anything like the same amount of stationary, materials etc. not running school trips all around the country, etc etc. In my son's school at least iPads have been in use as a teaching tool for several years now and the additional IT cost will be minimal.
Ah okay. I had assumed the staff were all working on remote lessons rather than being furloughed. Housekeeping and gardening will still need to happen, maybe catering in day schools could have been saved, and the utility bills will be a little lower. When I were a lad, stationery and trips were additionally charged.
Each department has some teaching staff doing the remote lessons but not the full amount. Certainly in the company I am involved in we are rotating staff on a 3 week basis, being the minimum amount of time that you are allowed to be furloughed. I suspect the school will do likewise for many of their staff. Not sure about housekeeping but certainly no lunches etc.
My son's school has cut fees by 20%.
Private school, Nigel? Odd choice for a progressive icon of the site.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
Yes. Many people are enjoying it. And with most of those who aren't, or who are very concerned about the economy, the message that it is saving lives RIGHT NOW will still be in the box seat.
Quite. At the moment it is uneconomic to make anyone redundant when you can furlough them. I’ve been advising my employer clients that it’s pointless making anyone redundant at the moment. Many are just waiting until July to do it.
Say you furlough somebody who is on £120,000 per annum.
Government pays £2,500 of their £10,000 per month salary.
You don't top up. So they take a £7,500 per month cut.
Is that legal under a normal employment contract?
I believe you can’t be automatically furloughed. You have to agree to it.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
Yes. Many people are enjoying it. And with most of those who aren't, or who are very concerned about the economy, the message that it is saving lives RIGHT NOW will still be in the box seat.
Quite. At the moment it is uneconomic to make anyone redundant when you can furlough them. I’ve been advising my employer clients that it’s pointless making anyone redundant at the moment. Many are just waiting until July to do it.
Say you furlough somebody who is on £120,000 per annum.
Government pays £2,500 of their £10,000 per month salary.
You don't top up. So they take a £7,500 per month cut.
Is that legal under a normal employment contract?
I believe you can’t be automatically furloughed. You have to agree to it.
Correct, unless the employment contract specifically allows for it which would be extremely rare.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
Yes. Many people are enjoying it. And with most of those who aren't, or who are very concerned about the economy, the message that it is saving lives RIGHT NOW will still be in the box seat.
Quite. At the moment it is uneconomic to make anyone redundant when you can furlough them. I’ve been advising my employer clients that it’s pointless making anyone redundant at the moment. Many are just waiting until July to do it.
Say you furlough somebody who is on £120,000 per annum.
Government pays £2,500 of their £10,000 per month salary.
You don't top up. So they take a £7,500 per month cut.
Is that legal under a normal employment contract?
I believe you can’t be automatically furloughed. You have to agree to it.
Correct, unless the employment contract specifically allows for it which would be extremely rare.
On topic, surely this is because the economic impact of coronavirus has not really been felt?
The weather is lovely. We haven’t run out of food. The government is paying everyone’s wages. Life is actually quite pleasant, certainly tolerable, at the moment, for many.
Wait til people move from furlough to the dole.
My son's school did an interesting questionnaire recently about fees for this term where there is some teaching online. From a range of options 40 odd per cent were content to pay the existing fees. I suspect that they were a combination of the rich and those who are not financially affected (there are a lot of doctors kids at the school). The next largest was those seeking a 20% discount which included me. Some went for smaller increases and about 70 are seeking bursaries on the basis that they can no longer afford the fees at all.
Which shows that the economic impact of this is currently very uneven and it is very likely to become even more so going forward.
Most English schools are voluntarily reducing fees. Prep schools around 15-20%, public schools more (Eton 30% for example).
Private schools are going to have a total nightmare. They are still incurring the vast majority of their costs during lockdown, lessons-from-home technology is an additional cost, and when lessons resume their variable costs are pretty much all staff salaries.
Most private schools (outside the top 'public' schools) operate as charities and don't have a lot of profit margin, savings or endowments.
Why would the top public schools not be charities? Eton and Winchester, frexample, are, and I'm guessing everyone else is.
Unless the Israeli prof (and a few others) are right, then the only way to guarantee not have some kind of peak is to continue in lockdown until, erm, maybe next April.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
There's going to be talk of almost nothing else but this target, for the next seven days.
In the eyes of the media and opposition, meeting (or missing) that target will define the whole government response to the pandemic.
You seem to overlooking the rather important fact that we are all talking about that figure because it was the the government that announced that target in the first place. Can't really blame anyone for criticising the government if it fails to meet a target it set for itself.
Why? What sort of world would it be if all targets were hit? Wouldnt the targets then be worthless?
There are plenty of scenarios where "stretched" targets are better than "achievable" targets. Even achievable targets should be missed a fair proportion of the time, otherwise they were unlikely to be needed in the first place.
Well, there is nothing to stop the government explaining this.
Unless the Israeli prof (and a few others) are right, then the only way to guarantee not have some kind of peak is to continue in lockdown until, erm, maybe next April.
I am more than ever convinced after Shapps' 'no risk' assertion that Boris' return on Monday will coincide with the fantastic news that the lockdown is over.
It's not like the US mainly because of the unsustainable cash to stay at home. After a year, the school numbers would (will?) look different. But naturally people want to work from home and keep the old folks away. That part of the political equation will be very difficult to change, maybe even after the virus.
Unless the Israeli prof (and a few others) are right, then the only way to guarantee not have some kind of peak is to continue in lockdown until, erm, maybe next April.
There would still be a risk that the virus would mutate and a 2nd peak would happen. It is a ridiculous and meaningless condition.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
South Korea has also had the virus far more under control than we've ever managed.
The UK is painfully slowly moving in the right direction with testing and surveillance. I wonder how much of a distraction this testing target has been though.
I bet it has been a distraction. Targets should be kept for the gun range IMO. I have never come across one in other matters that has led to anything but needless stress, false measurement, linguistic contortions and poor outcomes.
She must have amazing skills and spent all of April on it to visually see such a small week to week difference and be able to note its statistical significance.
Or her "seeing" it for herself might be meaningless.
She must have amazing skills and spent all of April on it to visually see such a small week to week difference and be able to note its statistical significance.
Or her "seeing" it for herself might be meaningless.
I've a sense that for all his disabling flaws as a politician he will be regarded as one of our better PMs in future years.
No he most certainly won't. He was probably the worst since WWII. His unique combination of control-freakery (every decision had to go to him personally in No 10) and incapability of actually taking a decision was unparalleled.
Luckily Peter Mandelson came in and rescued things to a large extent.
What is there to be won by these numbers going down further?
Also, do these "appalling" journalists fall randomly fall from the sky and are then imposed on the news outlets, or is it possible that news outlets spend considerable effort on finding out what their customers want to be fed with, and then select and instruct their journalists what to say and write to please their audiences? Who's at fault then, news outlets, customers or the "appalling journalists"?
Why "oh dear"? It's a mistake, sure, but not by Hancock, and those all look like sensible questions to me. Hard to see that as a "gaffe".
I think the notion of answering ones own questions is the issue. A bit like setting ones own 'A' level paper.
It's pretty common. I had a colleague who was asked to ask Gordon at PMQs (having got a precious slot to be called) a question about whether there was progress in such-and-such an area. He asked, and Gordon replied, "No." My colleague swore he'd never ask a planted question again.
I don't know Gordon Brown but can imagine him doing that. Reflects rather well on him. Makes me smile anyway.
Yes, he was the despair of his handlers. Tony would have finessed it - "I'm grateful to my honourable friend for his interest in this important issue, and look forward to briefing him as matters progress." Gordon had no patience for such arts.
I've a sense that for all his disabling flaws as a politician he will be regarded as one of our better PMs in future years.
It’s all the newbies trying to slot in under him in the league table.
The media have already blown up their own argument about going into the lockdown too slowly and too late, because they now ask every day when it will end and we have only done a month.
"More than 2.6million people in London may have already caught the coronavirus and recovered from it, data suggests.
Early results from antibody surveys - which reveal how many people have had the illness already - suggest the true death rate for COVID-19 may be anywhere between 0.1 and 0.6 per cent."
Very noble of her! Those most likely to try Dr. Trump's remedy certainly didn't vote for Hillary.
Good point. It's a safe bet that 100% of the people who will now kill themselves drinking bleach will be 2016 Trump voters who would have been sure to vote for him again in 2020. Wonder if it could cost him Alabama?
While I do think Mike's right that public support remains with the lockdown- despite the frothing of certain online elements- I don't think you should read these polls like election polls. Majorities don't matter here- if 30% of people decide to ignore the lockdown that's a huge problem.
Which will reinforce the resolve of others not to out because it's a pretty safe bet that those that decide to break the lockdown will be the same people who haven't bothered taking the precautionary measures seriously from the outset.
The last people I would go into a pub with will be those that rush there the moment they think they can get away with it.
The media have already blown up their own argument about going into the lockdown too slowly and too late, because they now ask every day when it will end and we have only done a month.
As soon as it ends it will have been lifted too early in the media questioning.
The one entity everyone wants you to think is in charge, though, is The Science. Of all the seven phrases of magnetic fridge poetry that are rearranged each day to fashion a 5pm briefing, the most ominous is surely “We have, at all times, been led by the science.” This is untrue, no matter how many times they say it. We have, at all times, been led by the government, whose job it is to take the scientific advice and make political decisions based on it. The ministers do not, in fact, all work for “The Science”, the most suspiciously shadowy overlord since Keyser Söze.
“The Science” is not some monolithic stone tablet that gets handed to Raab to read out to the masses – and all scientists themselves know this. The claim that government would be “led by the science” was reassuring many weeks ago, at the start of all this. But now, through suspicious overuse to the exclusion of all other considerations, it has become troubling. Whenever some politician standing next to him says it these days, my overactive imagination fancies it sees a flicker behind the eyes of Chris Whitty. Is he realising that they are fitting him up to be the guy who needs at least 36 changes of shirt for the inevitable public inquiry?
Another encouraging drop (577) in hospital admissions related to Covid-19 today - down 3.5% (was similar yesterday). Hopefully this continues over the next few days.
What is there to be won by these numbers going down further?
Also, do these "appalling" journalists fall randomly fall from the sky and are then imposed on the news outlets, or is it possible that news outlets spend considerable effort on finding out what their customers want to be fed with, and then select and instruct their journalists what to say and write to please their audiences? Who's at fault then, news outlets, customers or the "appalling journalists"?
The media management need to look at themselves and the widespread criticism and take action to become relevant and useful.
She must have amazing skills and spent all of April on it to visually see such a small week to week difference and be able to note its statistical significance.
Or her "seeing" it for herself might be meaningless.
The plural of "anecdote" is not "data".
If nothing else, portions of the construction industry are tentatively moving again.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
There's going to be talk of almost nothing else but this target, for the next seven days.
In the eyes of the media and opposition, meeting (or missing) that target will define the whole government response to the pandemic.
You seem to overlooking the rather important fact that we are all talking about that figure because it was the the government that announced that target in the first place. Can't really blame anyone for criticising the government if it fails to meet a target it set for itself.
Why? What sort of world would it be if all targets were hit? Wouldnt the targets then be worthless?
There are plenty of scenarios where "stretched" targets are better than "achievable" targets. Even achievable targets should be missed a fair proportion of the time, otherwise they were unlikely to be needed in the first place.
I should have used the word "committed to" rather than "targeted". As I recall Hancock told us he would do 100k a day not that he was aiming to do 100k tests a day. Hence the mad scramble to try and save face now.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
South Korea has also had the virus far more under control than we've ever managed.
The UK is painfully slowly moving in the right direction with testing and surveillance. I wonder how much of a distraction this testing target has been though.
I bet it has been a distraction. Targets should be kept for the gun range IMO. I have never come across one in other matters that has led to anything but needless stress, false measurement, linguistic contortions and poor outcomes.
I would actually say the reverse. A target is a good way of communicating across an organisation, if used correctly.
It was a pretty arbitrary target anyway. S. Korea managed to control their outbreak with around 20k a day, which suggests it's not all about the raw numbers.
Yep, but the media narrative is such that it has to be 100k per day no matter the cost, no matter the science.
There's going to be talk of almost nothing else but this target, for the next seven days.
In the eyes of the media and opposition, meeting (or missing) that target will define the whole government response to the pandemic.
You seem to overlooking the rather important fact that we are all talking about that figure because it was the the government that announced that target in the first place. Can't really blame anyone for criticising the government if it fails to meet a target it set for itself.
Why? What sort of world would it be if all targets were hit? Wouldnt the targets then be worthless?
There are plenty of scenarios where "stretched" targets are better than "achievable" targets. Even achievable targets should be missed a
fair proportion of the time, otherwise they were unlikely to be needed in the first place.
I should have used the word "committed to" rather than "targeted". As I recall Hancock told us he would do 100k a day not that he was aiming to do 100k tests a day. Hence the mad scramble to try and save face now.
He pledged 100,000 tests a day by the end of the month on 2nd April.
Comments
Sad to hear the news about Lynn Faulds Wood. She was doing the BBC or Sky paper review just a couple of weeks ago. RIP.
A lot of the Home Counties has high rates for its population density. Commuters I suppose.
The very lowest areas are places with very little commuting, Devon, Lincs, Dorset Humber etc. I note Hartlepool is low, but Newcastle and Sunderland high.
Trains, trams and buses must have been teeming with the virus. Take care Sunil!
For example, yes, about 10 years ago the idea of using NN started to be revived and were initially based upon work done in the 80s, now that much better computational power was available. But standard deep learning techniques, such as GANs, are a massive leap forward (albeit not without huge issues) on anything from the 80s.
This isn't my field, so very happy to be educated, but it's not immediately obvious to me what the past dataset looks like.
I see she outlived a diagnosis of stage 3 bowel cancer by 31 years. Highly encouraging for those in the same boat.
But the limiting factor wasn't really just a computational power, there were some fundamental mathematical issues about how to actually take what seems like a nice idea, with some very attractive properties, and turn it into something useful.
That work has only really occurred in the past 15 years or so. Sure having bigger computational power has certainly driven interest, but a number of significant advances have occurred e.g GPLVM for probabilistic dimensionality reduction, which in turn sparks lots of new avenues / techniques.
Journalists just confirming why they are not fit for purpose or covid reporting
Good to see they are not hoodwinking the public
I have no idea what the point of the IHME model is - it's clearly not very useful to produce such wide ranges of deaths.
But the LSHTM/Imperial ones showed lots of people dying and by all accounts pushed the govt to action. The reason they didn't show this earlier seems much more to be we didn't have enough data on key elements such as fatality rate, transmission rate, how many cases there actually are etc. I don't think fancy techniques can get around that.
As suggested down thread, Gaussian Processes are made for these kind of problems. Gaussian processes are often built on the infrastructure of something like Tensorflow, which is the open source library Google uses for deep learning.
I think there is a fairly good reason why the Royal Society have brought together a load of new people from fields well outside the narrow range that was being used and who are now working these problems.
Aerosol Filtration Efficiency of Common Fabrics Used in Respiratory Cloth Masks
https://pubs.acs.org/doi/full/10.1021/acsnano.0c03252#
...The emergence of a pandemic affecting the respiratory system can result in a significant demand for face masks. This includes the use of cloth masks by large sections of the public, as can be seen during the current global spread of COVID-19. However, there is limited knowledge available on the performance of various commonly available fabrics used in cloth masks. Importantly, there is a need to evaluate filtration efficiencies as a function of aerosol particulate sizes in the 10 nm to 10 μm range, which is particularly relevant for respiratory virus transmission. We have carried out these studies for several common fabrics including cotton, silk, chiffon, flannel, various synthetics, and their combinations. Although the filtration efficiencies for various fabrics when a single layer was used ranged from 5 to 80% and 5 to 95% for particle sizes of <300 nm and >300 nm, respectively, the efficiencies improved when multiple layers were used and when using a specific combination of different fabrics. Filtration efficiencies of the hybrids (such as cotton–silk, cotton–chiffon, cotton–flannel) was >80% (for particles <300 nm) and >90% (for particles >300 nm). We speculate that the enhanced performance of the hybrids is likely due to the combined effect of mechanical and electrostatic-based filtration. Cotton, the most widely used material for cloth masks performs better at higher weave densities (i.e., thread count) and can make a significant difference in filtration efficiencies. Our studies also imply that gaps (as caused by an improper fit of the mask) can result in over a 60% decrease in the filtration efficiency, implying the need for future cloth mask design studies to take into account issues of “fit” and leakage, while allowing the exhaled air to vent efficiently. Overall, we find that combinations of various commonly available fabrics used in cloth masks can potentially provide significant protection against the transmission of aerosol particles...
8 weeks and a day since I visited Central London (Liverpool Street station - um, to trainspot a new Class 745 intercity unit
8 weeks since I rode the Tube anywhere (Gants Hill to Epping via Hainault and Woodford and back)
7 weeks and 6 days since I flew from Luton to Aberdeen (to see my brother, his wife and their newborn son!)
7 weeks since I did the train from Aberdeen to Inverness and back
6 weeks and a day since I rode any plane, train or bus at all (plane from Aberdeen to Southend, two trains from Southend Airport to Romford via Shenfield, bus from Romford to Gants Hill)
5 weeks and 6 days since I visited a really big Sainsbury's
5 weeks and 2 days since I did any shopping at all (local shops/pharmacy) - though I also visited Valentines Park
Sars - "twenty-first century science played a relatively small role in controlling SARS; nineteenth-century techniques continued to prove their value”.
People were isolated and contact traced. It was less infectious and showed up sooner when people were which helped enormously.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636331/
I want HMG held to account but with responsible and informative journalism, not the Peston, Rigby, Kunnesberg schoool of 'gotcha' politics
And that is why I was so complimentary of Starmer at his first PMG
"More than 2.6million people in London may have already caught the coronavirus and recovered from it, data suggests.
Early results from antibody surveys - which reveal how many people have had the illness already - suggest the true death rate for COVID-19 may be anywhere between 0.1 and 0.6 per cent."
https://www.dailymail.co.uk/news/article-8250371/How-people-REALLY-caught-coronavirus.html
Here also I ought to leave a further remark for the use of posterity, concerning the manner of people's infecting one
another; namely, that it was not the sick people only from whom the plague was immediately received by others that were sound, but the well. To explain myself: by the sick people I mean those who were known to be sick, had taken their beds, had been under cure, or had swellings and tumours upon them, and the like; these everybody could beware of; they were either in their beds or in such condition as could not be concealed.
By the well I mean such as had received the contagion, and had it really upon them, and in their blood, yet did not show the consequences of it in their countenances: nay, even were not sensible of it themselves, as many were not for several days. These breathed death in every place, and upon everybody who came near them; nay, their very clothes retained the infection, their hands would infect the things they touched, especially if they were warm and sweaty, and they were generally apt to sweat too...
We want man bites dog, not dog bites man.
I was out on legitimate business.
I am only half joking, but seems to me there would be a lot less support for lockdown if it was pouring with rain every day. Many are getting, or will get, 80% salary to sit in their gardens. Maybe I am too cynical.
Or a random selection of questions from professional bodies or 'ordinary people' ?
I suspect the level of outrage on Twitter when this arrives on Sunday will provide some distraction from plague.
https://twitter.com/DawnHFoster/status/1253699037162803201?s=20
https://twitter.com/Fox_Claire/status/1253703067318484992?s=20
I think the Freakonomics guys have a thing about how the role of the press is misunderstood by intelligent people. It's not to educate the masses on the details of the complex issue of the day; it's primarily to get some sort of flavour across to the average person on the street, who wants to keep up to date but hasn't the time or patience to go and do the proper research themselves.
There are plenty of scenarios where "stretched" targets are better than "achievable" targets. Even achievable targets should be missed a fair proportion of the time, otherwise they were unlikely to be needed in the first place.
I've just discovered that you can't always rely on experts. This article, by a professor of urban studies, informs us that England is not as densely populated as the Netherlands. In fact England is more densely populated than the Netherlands. The article is using UK data not English data despite referring to England specifically. (I was trying to find information on the population of England and the Netherlands and this article was one of the first to come up on a search).
https://theconversation.com/think-your-country-is-crowded-these-maps-reveal-the-truth-about-population-density-across-europe-90345
Government pays £2,500 of their £10,000 per month salary.
You don't top up. So they take a £7,500 per month cut.
Is that legal under a normal employment contract?
There are no UK passengers left stranded on cruise ships he boasts.
He fails to point out that there are no passengers left on cruise ships globally as the Magnifca returned to port yesterday!
Unless the Israeli prof (and a few others) are right, then the only way to guarantee not have some kind of peak is to continue in lockdown until, erm, maybe next April.
https://twitter.com/benatipsosmori/status/1253592227512832001?s=21
Someone better tell XR.
Although it was slightly disguised within a sad anecdote about sunshine and a funeral.
Or her "seeing" it for herself might be meaningless.
Luckily Peter Mandelson came in and rescued things to a large extent.
Also, do these "appalling" journalists fall randomly fall from the sky and are then imposed on the news outlets, or is it possible that news outlets spend considerable effort on finding out what their customers want to be fed with, and then select and instruct their journalists what to say and write to please their audiences?
Who's at fault then, news outlets, customers or the "appalling journalists"?
The media have already blown up their own argument about going into the lockdown too slowly and too late, because they now ask every day when it will end and we have only done a month.
The last people I would go into a pub with will be those that rush there the moment they think they can get away with it.
“The Science” is not some monolithic stone tablet that gets handed to Raab to read out to the masses – and all scientists themselves know this. The claim that government would be “led by the science” was reassuring many weeks ago, at the start of all this. But now, through suspicious overuse to the exclusion of all other considerations, it has become troubling. Whenever some politician standing next to him says it these days, my overactive imagination fancies it sees a flicker behind the eyes of Chris Whitty. Is he realising that they are fitting him up to be the guy who needs at least 36 changes of shirt for the inevitable public inquiry?
https://www.theguardian.com/commentisfree/2020/apr/24/donald-trump-coronavirus-president-advice-bleach
If nothing else, portions of the construction industry are tentatively moving again.