I must say I am surprised at the number of whiny adults I am hearing. Boo Hoo I can not go fishing, Boo Hoo I don't want to wear a mask, Boo Hoo my favorite restaurant only has limited take out. Boo Hoo I need a hair cut and my nails done. Really folks my six year old neighbor ( that I talk to over the fence from a distance ) is doing better then some of you.
The New York Times is also reporting the drug Remdesivir is showing promising results in the treatment of COVID 19.
The New York Times is reporting that the manufacturer of Remdesivir claims promising results. Nowhere in the article was there mention of some of the deficiencies of the test, such as the lack of a control group.
When thousands of people are dying how to you establish a control group to take a placebo?
There are a lot of instances when a control group is not included in an experimental design. Prospective cohort studies are one, in which a group of individuals are tracked longitudinally. Another is observational studies wherein the study does not have an experimental manipulation, but rather participants are followed in their naturalistic setting, or alternatively, all participants undergo the same experimental protocol. Pilot studies are commonly set up this way. It's a good way to collect preliminary data without too much expense. Typically, in these kinds of studies analyses look at repeated measures, within-subject effects (i.e., before vs after).
Matt Hancock, the health secretary, admitted on Friday that he had a “high degree of confidence” that the number and proportion of Covid-19 deaths in care homes was higher than reported.
That has to be true. And the numbers of the dead (15,464 according to worldometer) are distressingly high without that. I think it was Chris Whitty who observed that 20,000 deaths in this epidemic wave would be a "good, if horrifying result" based on how bad things might have been. It's increasingly likely that we will exceed that total even without including the care homes deaths.
FWIW Canada, which has a lower rate of deaths per thousand population than many countries, including (by a factor of 4 or 5 times) the UK, is including nursing home/long term care home deaths in its figures. And those deaths make up nearly 50 per cent of total deaths, or did until Friday (have not heard whether the nearly 50 per cent figure is still valid -- it may well be). SO I'm guessing that you in the UK (or just in England?) need to just about double the number of deaths reported from hospitals to come anywhere close to a real death count.
When thousands of people are dying how to you establish a control group to take a placebo?
The control group is given the best medical treatment known and you evaluate day by day. Note that it is not unknown for a treatment being tested causing more death (or other problems) than the placebo.
NZ has had 12 deaths - all but 2 were contracted care homes. Both the others were contracted by contact with someone who came from overseas.
On Monday it will be announced whether we go from level 4 to level 3. For many of us the change won't make much difference, but it will be of economic benefit to the construction industry and some manufacturing. Also people will be able to travel outside their city, but not their region. There has been a huge effort in increasing testing, and the results from that look promising. The message though is still "stay home and keep in your bubble". The "bubble" is your household, but people living alone will be able to visit someone else living alone as long as they are exlusive.
It's easy for me to say, because I am retired and it won't affect my income, but I don't mind what the decision is. I think the decision making process is fairly robust. It won't change my everyday behaviour. Going down to level 2 when the library will be open is what I'm looking forward to, but not until it has been deemed safe.
X. Country - [# of known cases] ([active] / [recovered] / [dead]) [%fatality rate]
Fatality rates are only listed for countries where the number of resolved cases (recovered + dead) exceeds the number of known active cases by a ratio of at least 2:1. At the moment only China, Iran, Switzerland, Austria, and South Korea meet that criterion. Italics indicate authoritarian countries whose official statistics are suspect. Other country's statistics are suspect if their testing regimes are substandard.
If American states were treated as individual countries sixteen of them would be on that list. New York would be ranked at #2, between "everywhere in the U.S. except New York" (#1) and Spain (#3). New Jersey would be between Turkey and Iran.
Japan has joined the 10,000 case club since the last compilation.
The anti-viral drug Remdesivir is in a series of clinical trials. The only info I've found is case studies.
Bloke in the hospital up the road ( a bus driver of tourists who gave it to one another in the manner of bus tourists) was on a total breathing machine thingy for like days on end and they gave him that as a last resort/best guess/ can't hurt if it doesn't work and he got well. Anecdata - but it's given me a bit of hope
The anti-viral drug Remdesivir is in a series of clinical trials. The only info I've found is case studies.
Bloke in the hospital up the road ( a bus driver of tourists who gave it to one another in the manner of bus tourists) was on a total breathing machine thingy for like days on end and they gave him that as a last resort/best guess/ can't hurt if it doesn't work and he got well. Anecdata - but it's given me a bit of hope
A long way down, Ramdesivir is described as the 'most promising' drug for Covid-19. It seems to inhibit the viral genome replication.
Two things need to be held in tension here:
1) At the moment, there is no good evidence for anything - things are moving very fast. However, there is potentially some nice data that Ramdsivir could be really useful
2) Beware of anyone claiming a 'miracle cure' or whatever with any degree of confidence.
2) Beware of anyone claiming a 'miracle cure' or whatever with any degree of confidence.
AFZ
The MOOC I'm trying to follow on epidemiology has got as far as confidence intervals. Just looking at the formula for the 95% confidence interval sends my non-mathematical brain into a corner whimpering.
(I still claim the E I got in Maths A level was due to me spending all my time in maths classes trying to evangelise the guy at the desk next to me...)
I think of it generally as: 95% confidence interval (of , say, 67-73) means - if we repeat this 100 times, 95 times out of that hundred the result would fall within the those two numbers.
Thanks again AFZ. I always try to find reputable sources of information - and of course that's one of the best.
I read the "Managing Stress" sidebar and thought it could be useful. Nothing startlingly new, but a good reminder of techniques that might help. I know if I panic my IQ can immediately be expressed in negative numbers, and I can forget things I already know - which is why a reminder is always welcome.
I think of it generally as: 95% confidence interval (of , say, 67-73) means - if we repeat this 100 times, 95 times out of that hundred the result would fall within the those two numbers.
And the worry there is that if you're conducting 100 studies of different treatments on 95% confidence intervals you're likely to get false positives and false negatives.
2) Beware of anyone claiming a 'miracle cure' or whatever with any degree of confidence.
AFZ
The MOOC I'm trying to follow on epidemiology has got as far as confidence intervals. Just looking at the formula for the 95% confidence interval sends my non-mathematical brain into a corner whimpering.
(I still claim the E I got in Maths A level was due to me spending all my time in maths classes trying to evangelise the guy at the desk next to me...)
Confidence intervals can be mathematically very complex but the concept is really simple.
The whole point of any statistics is to answer this question about experimental data: Is this result real or it is the result of random chance alone?
A confidence interval (and it's usually 95% in medical literature) simply means We can be 95% confident that the real result lies between these two numbers
(That's it. Feel free to skip the rest).
For example if you wanted to know about the height of the UK population, in theory, you could measure every single one of the 40 odd million adults in the UK. That would be very expensive and very time-consuming. OTOH if you sampled 100 people you could then work out what the mean of that sample was. The question would be how close is this sample mean to the True mean for the whole population.
The confidence interval is derived from this sample mean. And thus you could be 95% confident that the mean height of British people lies between 178.2cm and 184.6cm (made up numbers). If you sample 1000 people, your confidence interval will get smaller and might be 180.6-182.1 for example. (Confidence intervals get smaller with larger sample sizes). Or to put it another way; there is a 5% chance that the true mean lies outside of those numbers. That's confidence intervals. If you're interested, this works in a large part because of a phenomenon called Central Limit Theorem.
Here's another example of this at work... (This is me being really sad!) I have ended up running the tech to enable our church to have a live virtual service every Sunday morning. I wanted to know how many people are in our virtual congregation. Facebug tells me how many streams are running simultaneously but there's no way of knowing how many people are watching each stream. So I conducted a survey. Based on around 50 responses I know that the mean number of people watching each stream is 1.98 and thus I simply multiply that by the number of streams to get an estimate of the congregation size. I took that a step further and worked out the confidence interval. The 95% CI is 1.69-2.27. Thus is we have 100 streams running, my central estimate would be that it equates to 198 people watching. I can be 95% confident that the real number is between 169 and 227. I.e. there is only a 2.5% chance that it's less than 169 and 2.5% chance that it's higher than 227.
In the medical literature, the p-values (which is a similar, but slightly different concept) and the 95% confidence intervals are the key to understanding whether the data shows a real result or whether the truth is that it's just chance alone. For example if we go back to our height measurements again if I have a sample of 100 people with a mean height of 181cm and you have a sample of 100 people with a mean of 179cm, we can see intuitively that these are probably the same population and it's just chance that meant you got a slightly lower result than I did. Conversely, if we both had samples of 10,000 with those numbers then we're probably talking about genuinely different populations. The stats is all about calculating the probability of a result being real or true.
Isn't it true that we will get a very small proportion of false positives and false negatives from any highly reliable diagnostic test. And we will get a very small proportion of very dangerous side effects from any effective treatment. Or any effective vaccine. I read the disclaimer advice for my regular medication and get disclaimer advice for my annual flu jab. I've had one very bad reaction to a previous flu jab. That's life. I don't expect 100% guarantees.
For the COVID-19 virus, given my age and pre-existing conditions I'm probably looking at (at least) an 8% risk of death and double that for serious illness. I'll take 95% remedial or preventive success rates. YMMV.
FWIW Canada, which has a lower rate of deaths per thousand population than many countries, including (by a factor of 4 or 5 times) the UK, is including nursing home/long term care home deaths in its figures. And those deaths make up nearly 50 per cent of total deaths, or did until Friday (have not heard whether the nearly 50 per cent figure is still valid -- it may well be). SO I'm guessing that you in the UK (or just in England?) need to just about double the number of deaths reported from hospitals to come anywhere close to a real death count.
Scotland is including care home deaths, but with a time lag - hospital deaths are updated daily, but care home deaths are weekly. The next update will be on Wed.
As of 12 April, 62% of deaths happened in hospital, 24.6 in care homes, 13.3% at home, and the remaining death was in an "other institutuion" - prison, AIUI.
Possibly of more relevance is the difference between deaths from all causes per week, compared to the average of deaths from all causes over the previous five years. Scotland had 1969 deaths from all causes last week (w/b 6 April) compared to an average of 1100 for the same week in the previous five weeks. Hence we had 869 "extra" deaths.
Isn't it true that we will get a very small proportion of false positives and false negatives from any highly reliable diagnostic test. And we will get a very small proportion of very dangerous side effects from any effective treatment. Or any effective vaccine. I read the disclaimer advice for my regular medication and get disclaimer advice for my annual flu jab. I've had one very bad reaction to a previous flu jab. That's life. I don't expect 100% guarantees.
For the COVID-19 virus, given my age and pre-existing conditions I'm probably looking at (at least) an 8% risk of death and double that for serious illness. I'll take 95% remedial or preventive success rates. YMMV.
That's a different 95% though. We're talking about a 95% chance that the effectiveness lies within a certain range. If the bottom end of that range is just above placebo then there's a 2.5% chance it's either no better or indeed worse than placebo. If you have a hundred potential treatments that are all, in reality, duds, you might well expect to get results for 2-3 of them that appear effective on a 95% confidence interval simply by chance. The recovery rate for covid-19 is high enough that any trials need hundreds of participants in order to get meaningful data.
Finding a lot of participants is really not a problem at the moment.
Sure, but it means that you need to keep an eye out for the reports of 30 or 60 people being treated and it showing "promising" results. And even if you've got enough participants large scale trials require more manpower to monitor and set up.
We agree that we need the data from trials to give more accurate information about both effectiveness and risks. And I side with DT re human trial volunteers. But we will not get through this effectively if we apply 100% reliability standards. I don't want to be stuck in self isolation any longer than necessary.
There are levels of risk I'm perfectly happy to accept as a trade off for improved quality of life, but not at the cost of worsening the risk for others. Both from a personal and a social viewpoint, we need the data to inform these choices.
Some newspapers reporting a relaxation of the lockdown in May, e.g., schools, some shops. I am curious as to whether this is a govt leak, or pressure from business interests, or both. Of course, the danger is a new spike. Mail on Sunday, Sunday Times.
Following watching Merkel's speech, I've decided that in my head, my preferred leader for current situation would go (of the UK candidates).
Ed Mil
Thatcher or Brown
Corbyn (would be up with Ed, except that the Sun would tell people to breach social distancing, etc...)
Major
Blair (we know what he'd do if Britain started one)
Teresa May/M Howard/etc...
Lord Buckethead
Johnson/IDS/etc...
...
/Farage/etc...
I inexplicably thought of Major this morning too. Major & Brown would work if you were looking for the old guard, but would also go for Starmer/Sturgeon/Major (there not being a single Tory in the current government I'd trust to run a whelk stall).
Re technopeasant and not having a cell phone. Both cell phones and internet are essential infrastructure here. Province twice the size of the UK with 1.1 million people. You'd have a free cell phone here. Cell data plans are also free for the infection duration. Pre-virus wifi was generally available within cities and towns. Not fast but usable for basic communication. Work and school are as online as possible now.
I wonder about considering this essential infrastructure everywhere.
Chemist, would at least understand the concepts. (plus I wanted to be fair, and also the Merkel video was high in my mind).
She'd still lose to a supported Corbyn, because he has a much better grasp on reality.
Oh and Stugeon would go v high, good call, (again, my experience of old people is that they'd willfully avoid listening, though) Starmer is still a bit unknown.
Much as her influence on just about everything was utterly dire, when the science and the economics clashed she went with the science. Sadly she taught her admirers to do as she said not as she did on that point.
She didn't have that public school education that has taught so many of the current crop that they can just bullshit through everything. I'm not saying there's anything else to be said in her favour.
One obvious problem is the need for necromancy. I know the dark lord Cummings is probably capable of that, but I'd much prefer it if we didn't have to have him involved in any way.
Beautiful day in London, blue sky, warm sunshine. Normally, there would be a rush to parks and the river, well, there is, but distancing isn't too bad, except for the dreaded joggers. People have absorbed the warnings. I wonder for how long.
For example if we go back to our height measurements again if I have a sample of 100 people with a mean height of 181cm and you have a sample of 100 people with a mean of 179cm, we can see intuitively that these are probably the same population and it's just chance that meant you got a slightly lower result than I did. Conversely, if we both had samples of 10,000 with those numbers then we're probably talking about genuinely different populations. The stats is all about calculating the probability of a result being real or true.
Or possibly systematic bias, where the sampling isn't quite random. For example, we have detailed height records of the well-off attendees of various British military academies from the nineteenth century. We also have detailed height records of various charitable organizations that supplied orphans to the British Navy of the time. In one sense these were the same population, but due to differences in nutrition the military academy graduates were, on average, about eight inches taller than their age-peers from orphanages. A sample which skewed too heavily towards one of these groups would be biased in its result.
We're still well below freezing overnights and just above in the day, with about 18 inches of snow left. It's melting but did lightly snow yesterday. Canada geese are out on the ice puddling where there's water. They're always in pairs this time of year. I'm enjoying their honking. And there's precious little competing noise.
We're still well below freezing overnights and just above in the day, with about 18 inches of snow left. It's melting but did lightly snow yesterday. Canada geese are out on the ice puddling where there's water. They're always in pairs this time of year. I'm enjoying their honking. And there's precious little competing noise.
They are noisy critters. As they're an invasive species in the UK, I think, I find their noisiness a bit over the top. Greylags are a lot quieter.
I heard and saw some Canadas around our local pond, which is over 2500 metres round, but is a pond. Most people 'get' six feet, but some cyclists blithely seem unaware of what that distance is. But the vast majority get it.Thankfully, because they had threatened to close it to maintain distancing.
Gavin Williamson just announced that households of children in exam years and carers will receive 4G broadband until the lockdown eases.
That's certainly an expensive way of providing bad broadband.
Faster than ADSL and more widely available than fibre. Expensive is going to depend on how metered it is and what sort of discount the government have managed to wangle from the phone providers.
We use 4G for our internet connection and it's lightyears ahead of what we could get via landline.
4G here offers about 20+ Mbs down and about 5 up. It is all free here too as I've mentioned, as the main cell phone company is a Crown Corporation (totally government owned) as private companies are only in it for profit and won't serve most of the province.
This is a very good public health decision. But it is should really be free to everyone. Realizing that if it is privately owned, it's costly. This is when I think nationalization of things which never should have been private.
Gavin Williamson just announced that households of children in exam years and carers will receive 4G broadband until the lockdown eases.
That's certainly an expensive way of providing bad broadband.
Faster than ADSL and more widely available than fibre. Expensive is going to depend on how metered it is and what sort of discount the government have managed to wangle from the phone providers.
Expense has to also factor in the cost of the device vs that of an ADSL/VDSL router (and in general mobile bandwidth is more expensive than wired bandwidth - and you will easily use a fairly large amount of it streaming classes and the like).
We use 4G for our internet connection and it's light years ahead of what we could get via landline.
Yes, I think it's fine as a stopgap, just a shame that a general upgrade of the network was deemed to be broadband communism.
Comments
When thousands of people are dying how to you establish a control group to take a placebo?
There are a lot of instances when a control group is not included in an experimental design. Prospective cohort studies are one, in which a group of individuals are tracked longitudinally. Another is observational studies wherein the study does not have an experimental manipulation, but rather participants are followed in their naturalistic setting, or alternatively, all participants undergo the same experimental protocol. Pilot studies are commonly set up this way. It's a good way to collect preliminary data without too much expense. Typically, in these kinds of studies analyses look at repeated measures, within-subject effects (i.e., before vs after).
FWIW Canada, which has a lower rate of deaths per thousand population than many countries, including (by a factor of 4 or 5 times) the UK, is including nursing home/long term care home deaths in its figures. And those deaths make up nearly 50 per cent of total deaths, or did until Friday (have not heard whether the nearly 50 per cent figure is still valid -- it may well be). SO I'm guessing that you in the UK (or just in England?) need to just about double the number of deaths reported from hospitals to come anywhere close to a real death count.
The control group is given the best medical treatment known and you evaluate day by day. Note that it is not unknown for a treatment being tested causing more death (or other problems) than the placebo.
Visors don't trap particulates (or anything)
On Monday it will be announced whether we go from level 4 to level 3. For many of us the change won't make much difference, but it will be of economic benefit to the construction industry and some manufacturing. Also people will be able to travel outside their city, but not their region. There has been a huge effort in increasing testing, and the results from that look promising. The message though is still "stay home and keep in your bubble". The "bubble" is your household, but people living alone will be able to visit someone else living alone as long as they are exlusive.
It's easy for me to say, because I am retired and it won't affect my income, but I don't mind what the decision is. I think the decision making process is fairly robust. It won't change my everyday behaviour. Going down to level 2 when the library will be open is what I'm looking forward to, but not until it has been deemed safe.
The listings are in the format:
X. Country - [# of known cases] ([active] / [recovered] / [dead]) [%fatality rate]
Fatality rates are only listed for countries where the number of resolved cases (recovered + dead) exceeds the number of known active cases by a ratio of at least 2:1. At the moment only China, Iran, Switzerland, Austria, and South Korea meet that criterion. Italics indicate authoritarian countries whose official statistics are suspect. Other country's statistics are suspect if their testing regimes are substandard.
If American states were treated as individual countries sixteen of them would be on that list. New York would be ranked at #2, between "everywhere in the U.S. except New York" (#1) and Spain (#3). New Jersey would be between Turkey and Iran.
Japan has joined the 10,000 case club since the last compilation.
AFZ
Bloke in the hospital up the road ( a bus driver of tourists who gave it to one another in the manner of bus tourists) was on a total breathing machine thingy for like days on end and they gave him that as a last resort/best guess/ can't hurt if it doesn't work and he got well. Anecdata - but it's given me a bit of hope
This is a really good site I've just found: https://www.hopkinsguides.com/hopkins/view/Johns_Hopkins_ABX_Guide/540747/all/Coronavirus_COVID_19__SARS_CoV_2_
A long way down, Ramdesivir is described as the 'most promising' drug for Covid-19. It seems to inhibit the viral genome replication.
Two things need to be held in tension here:
1) At the moment, there is no good evidence for anything - things are moving very fast. However, there is potentially some nice data that Ramdsivir could be really useful
2) Beware of anyone claiming a 'miracle cure' or whatever with any degree of confidence.
AFZ
(I still claim the E I got in Maths A level was due to me spending all my time in maths classes trying to evangelise the guy at the desk next to me...)
I read the "Managing Stress" sidebar and thought it could be useful. Nothing startlingly new, but a good reminder of techniques that might help. I know if I panic my IQ can immediately be expressed in negative numbers, and I can forget things I already know - which is why a reminder is always welcome.
And the worry there is that if you're conducting 100 studies of different treatments on 95% confidence intervals you're likely to get false positives and false negatives.
Confidence intervals can be mathematically very complex but the concept is really simple.
The whole point of any statistics is to answer this question about experimental data:
Is this result real or it is the result of random chance alone?
A confidence interval (and it's usually 95% in medical literature) simply means
We can be 95% confident that the real result lies between these two numbers
(That's it. Feel free to skip the rest).
For example if you wanted to know about the height of the UK population, in theory, you could measure every single one of the 40 odd million adults in the UK. That would be very expensive and very time-consuming. OTOH if you sampled 100 people you could then work out what the mean of that sample was. The question would be how close is this sample mean to the True mean for the whole population.
The confidence interval is derived from this sample mean. And thus you could be 95% confident that the mean height of British people lies between 178.2cm and 184.6cm (made up numbers). If you sample 1000 people, your confidence interval will get smaller and might be 180.6-182.1 for example. (Confidence intervals get smaller with larger sample sizes). Or to put it another way; there is a 5% chance that the true mean lies outside of those numbers. That's confidence intervals. If you're interested, this works in a large part because of a phenomenon called Central Limit Theorem.
Here's another example of this at work... (This is me being really sad!) I have ended up running the tech to enable our church to have a live virtual service every Sunday morning. I wanted to know how many people are in our virtual congregation. Facebug tells me how many streams are running simultaneously but there's no way of knowing how many people are watching each stream. So I conducted a survey. Based on around 50 responses I know that the mean number of people watching each stream is 1.98 and thus I simply multiply that by the number of streams to get an estimate of the congregation size. I took that a step further and worked out the confidence interval. The 95% CI is 1.69-2.27. Thus is we have 100 streams running, my central estimate would be that it equates to 198 people watching. I can be 95% confident that the real number is between 169 and 227. I.e. there is only a 2.5% chance that it's less than 169 and 2.5% chance that it's higher than 227.
In the medical literature, the p-values (which is a similar, but slightly different concept) and the 95% confidence intervals are the key to understanding whether the data shows a real result or whether the truth is that it's just chance alone. For example if we go back to our height measurements again if I have a sample of 100 people with a mean height of 181cm and you have a sample of 100 people with a mean of 179cm, we can see intuitively that these are probably the same population and it's just chance that meant you got a slightly lower result than I did. Conversely, if we both had samples of 10,000 with those numbers then we're probably talking about genuinely different populations. The stats is all about calculating the probability of a result being real or true.
AFZ
For the COVID-19 virus, given my age and pre-existing conditions I'm probably looking at (at least) an 8% risk of death and double that for serious illness. I'll take 95% remedial or preventive success rates. YMMV.
Scotland is including care home deaths, but with a time lag - hospital deaths are updated daily, but care home deaths are weekly. The next update will be on Wed.
As of 12 April, 62% of deaths happened in hospital, 24.6 in care homes, 13.3% at home, and the remaining death was in an "other institutuion" - prison, AIUI.
Possibly of more relevance is the difference between deaths from all causes per week, compared to the average of deaths from all causes over the previous five years. Scotland had 1969 deaths from all causes last week (w/b 6 April) compared to an average of 1100 for the same week in the previous five weeks. Hence we had 869 "extra" deaths.
That's a different 95% though. We're talking about a 95% chance that the effectiveness lies within a certain range. If the bottom end of that range is just above placebo then there's a 2.5% chance it's either no better or indeed worse than placebo. If you have a hundred potential treatments that are all, in reality, duds, you might well expect to get results for 2-3 of them that appear effective on a 95% confidence interval simply by chance. The recovery rate for covid-19 is high enough that any trials need hundreds of participants in order to get meaningful data.
Sure, but it means that you need to keep an eye out for the reports of 30 or 60 people being treated and it showing "promising" results. And even if you've got enough participants large scale trials require more manpower to monitor and set up.
There are levels of risk I'm perfectly happy to accept as a trade off for improved quality of life, but not at the cost of worsening the risk for others. Both from a personal and a social viewpoint, we need the data to inform these choices.
Ed Mil
Thatcher or Brown
Corbyn (would be up with Ed, except that the Sun would tell people to breach social distancing, etc...)
Major
Blair (we know what he'd do if Britain started one)
Teresa May/M Howard/etc...
Lord Buckethead
Johnson/IDS/etc...
...
/Farage/etc...
I wonder about considering this essential infrastructure everywhere.
She'd still lose to a supported Corbyn, because he has a much better grasp on reality.
Oh and Stugeon would go v high, good call, (again, my experience of old people is that they'd willfully avoid listening, though) Starmer is still a bit unknown.
Starmer is still something of an unknown quantity, yes.
O for the chaos of a John Major government...
She didn't have that public school education that has taught so many of the current crop that they can just bullshit through everything. I'm not saying there's anything else to be said in her favour.
:eyeroll:
We went to the beach today, but shared it only with a solitary cow (who maintained a more than ample social distance).
Or possibly systematic bias, where the sampling isn't quite random. For example, we have detailed height records of the well-off attendees of various British military academies from the nineteenth century. We also have detailed height records of various charitable organizations that supplied orphans to the British Navy of the time. In one sense these were the same population, but due to differences in nutrition the military academy graduates were, on average, about eight inches taller than their age-peers from orphanages. A sample which skewed too heavily towards one of these groups would be biased in its result.
I heard and saw some Canadas around our local pond, which is over 2500 metres round, but is a pond. Most people 'get' six feet, but some cyclists blithely seem unaware of what that distance is. But the vast majority get it.Thankfully, because they had threatened to close it to maintain distancing.
That's certainly an expensive way of providing bad broadband.
Faster than ADSL and more widely available than fibre. Expensive is going to depend on how metered it is and what sort of discount the government have managed to wangle from the phone providers.
We use 4G for our internet connection and it's lightyears ahead of what we could get via landline.
This is a very good public health decision. But it is should really be free to everyone. Realizing that if it is privately owned, it's costly. This is when I think nationalization of things which never should have been private.
Expense has to also factor in the cost of the device vs that of an ADSL/VDSL router (and in general mobile bandwidth is more expensive than wired bandwidth - and you will easily use a fairly large amount of it streaming classes and the like).
Yes, I think it's fine as a stopgap, just a shame that a general upgrade of the network was deemed to be broadband communism.
Well yes, but then any number of things are worse as a result of the last decade of election results.