Coronavirus

[RogerS]

For me, this "cock up" has been very revealing, and gratifying. It has made it abundantly clear that the government has followed the scientific advice it was given, scrupulously. That is very much a good thing. The science was initially flawed (well, the computer modelling), but the scientific community quickly corrected the error. This is science working the way it should. Thank goodness we live here and not under Trump, where science is adjusted for political reasons and scientists are routinely ignored.

No....totally wrong. Government took the Imperial College model and didn't bother to check it or question the key assumptions. How can you try and sweep away the monumental cock-up that Imperial made with a simple 'Well, that's science for you folks, shrugs shoulders".

It is NOT science working the way it should. Ever heard of 'peer review' ? Clearly that key element of science escaped those at Imperial College and the Govt.

 

[RogerS]

For me, this "cock up" has been very revealing, and gratifying. It has made it abundantly clear that the government has followed the scientific advice it was given, scrupulously. That is very much a good thing. The science was initially flawed (well, the computer modelling), but the scientific community quickly corrected the error. This is science working the way it should. Thank goodness we live here and not under Trump, where science is adjusted for political reasons and scientists are routinely ignored.

I couldn't agree more. The scientists, under great pressure, got the ball rolling and the government acted. Once assumptions were shown to be false, remedial action was taken.

I think that as a culture we have taken on the idea that "experts" always get everything right. They usually do ... eventually ... but a lot of models get discarded on the way. The problem with the current situation is that mistakes can lead to great personal tragedy for families as opposed to say, a new choke point for traffic being inadvertently created. Unfortunately there is no magic way of getting around the usual error-strewn path.

Never heard of 'peer review' then ? That IS what good science is all about. A 'peer review' is your magic way. It's not rocket-science. I can't understand why you are defending both Imperial and the Govt.

 

[MikeG.]

Settle down Roger. It certainly is the way science works: others check your published work. That is the self correcting nature of the beast. Mistakes are made, and even some fabrication once in a blue moon. That doesn't survive the repeatability phase of the scientific method, which is the very essence of the process. You seem to be on a mission to blame someone. I'm sitting here sweating with a throbbing head, supreme lethargy and a horrible cough. If one of us had reason to be clutching around for someone to blame, it's me, surely. I'm a bit disappointed that you can't be a little more understanding about this situation. Two months ago this disease didn't exist.

 

[SammyQ]

I'm with Jake and RogerS. Peer review is a fundamental, rock bottom, can't-get-past-it aspect of Science. The central point here is IC used flawed data: historic 'flu figures instead of contemporary C19. Are you serious?

Sam

 

[TFrench]

Not sure if anyone has seen it, but Johns Hopkins University has a very good map showing how many cases in each country, recovery rates etc. It really knocks it home to see how few cases Russia is reporting - either they've escaped it or they're telling porkies...

Sent from my SM-G973F using Tapatalk

 

[MikeG.]

......Never heard of 'peer review' then ? That IS what good science is all about. A 'peer review' is your magic way. It's not rocket-science. I can't understand why you are defending both Imperial and the Govt.

Peer review checks your methods, and checks that your conclusions match your results. It also checks that your methods are correct. It most certainly does not try to replicate your results. That's for others, post-publication. My daughter has published 8 or 10 papers and writes peer reviews for others. Don't ascribe magic to a very mundane method-checking process.

You also have to differentiate between scientific advice to government and published papers. Was the Imperial College stuff published in a scientific journal? I don't think it was. It therefore wasn't subject to the peer review process. Even so, its flaws were very quickly picked up by the rest of the scientific community, and corrected. No-one is defending Imperial College. We're defending the scientific method, though, which corrected the error in no time flat.

 

[MikeG.]

I'm with Jake and RogerS. Peer review is a fundamental, rock bottom, can't-get-past-it aspect of Science. The central point here is IC used flawed data: historic 'flu figures instead of contemporary C19. Are you serious?

Sam

Can you link to the peer-reviewed journal in which the work was published prior to being given to the government?

If you can't, then this is a red herring.

 

[Chris152]

Both, Chris. I see both of those things as a positive for the reasons I gave. Science worked, in that it self-corrected. The government did what it should in following the advice.

I'd say both govt and the scientists developing the advice failed in one very basic way - comparing their position to that taken by others. It's the first step in the development of knowledge in any existent discipline. If you find significant differences, to find the cause of those differences. Clearly in some circumstances that can take years - but in this case, if what Jake's written is correct, the error was simple and should have been identified easily. It wasn't and while we should have been taking stringent measures to contain the virus, we were washing our hands and singing songs.

 

[MikeG.]

.....It's the first step in the development of knowledge in any existent discipline......

An existent discipline? We've had a global pandemic of this nature recently, have we? As I said, a couple of months ago this disease didn't exist.

 

[Chris152]

.....It's the first step in the development of knowledge in any existent discipline......

An existent discipline? We've had a global pandemic of this nature recently, have we? As I said, a couple of months ago this disease didn't exist.

Fair enough - but in the space of a couple of months enough evidence existed to pick up that error specifically, and there was already evidence from other countries affected that the measures that the bad advice gave rise to would be inadequate.

 

[Andy Kev.]

I'm not sure what your point is - to defend the government for taking bad advice (in spite of the quite different, more stringent actions they could see other governments had already been taking), or to defend that part of the scientific community which developed the bad advice (in spite of existing research and data developed internationally that conflicted and was used to criticise the bad advice). Or both.

Your first point doesn't hold up if you think about it. Nobody's going to come up to the government and say, "We've got some bad advice for you here". And the govt. isn't going to reply, "Oh thanks, we'll take that". The advice will have been made and taken in good faith.

It's also probably true that at the time decisions were taken, it was not clear that what other govts. were doing would necessarily be more effective. Obviously the most guaranteed effective method would be instant, complete shutdown of the country, suspension of flights etc. However, can you imagine how the people who are complaining now would have reacted to that? (I'm afraid I'm implying here that there are people who complain come what may about anything which does not chime with their expectations/wishes.)

To err is human. Are we seriously expecting our govt. to work to super human standards? They should be better than the layman, certainly but one would have to be mad to think that one had a reasonable expectation of them getting everything right. Willfull negligence is unacceptable but I can see no reason to believe that we are dealing with that in this crisis.

There's also the matter of how govt./crisis management works. Read any detailed account of a period of govt. activity and you'll see what a role coincidence, chance, cock-up and pure luck play. If you want a light-hearted and entertaining example of this, I would recommend All Out War by TIm Shipman. If you can eliminate those factors, I'd vote for you.

 

[Andy Kev.]

For me, this "cock up" has been very revealing, and gratifying. It has made it abundantly clear that the government has followed the scientific advice it was given, scrupulously. That is very much a good thing. The science was initially flawed (well, the computer modelling), but the scientific community quickly corrected the error. This is science working the way it should. Thank goodness we live here and not under Trump, where science is adjusted for political reasons and scientists are routinely ignored.

I couldn't agree more. The scientists, under great pressure, got the ball rolling and the government acted. Once assumptions were shown to be false, remedial action was taken.

I think that as a culture we have taken on the idea that "experts" always get everything right. They usually do ... eventually ... but a lot of models get discarded on the way. The problem with the current situation is that mistakes can lead to great personal tragedy for families as opposed to say, a new choke point for traffic being inadvertently created. Unfortunately there is no magic way of getting around the usual error-strewn path.

Never heard of 'peer review' then ? That IS what good science is all about. A 'peer review' is your magic way. It's not rocket-science. I can't understand why you are defending both Imperial and the Govt.

Tell us all: how long in your experience does the peer review process take? Is it something to which decisions taken urgently under great time pressure are routinely subjected?

Perhaps you could enlighten us.

 

[SammyQ]

Mike, the model itself has not been published, but the IC paper has been referred to a few times.

Sam

 

[Andy Kev.]

Both, Chris. I see both of those things as a positive for the reasons I gave. Science worked, in that it self-corrected. The government did what it should in following the advice.

I'd say both govt and the scientists developing the advice failed in one very basic way - comparing their position to that taken by others. It's the first step in the development of knowledge in any existent discipline. If you find significant differences, to find the cause of those differences. Clearly in some circumstances that can take years - but in this case, if what Jake's written is correct, the error was simple and should have been identified easily. It wasn't and while we should have been taking stringent measures to contain the virus, we were washing our hands and singing songs.

I understand the point you're making there but comparison with other positions only becomes valid once we have the final results. Take for instance Italy: given that it is culturally common for three generations of a family to live in the same house, the relative failure (so far) of their response does not mean that precisely that response might not be ideal for another society. One could go on ad infinitum.

The science of the virus is universal. The art of the governmental responses to its effects must by definition be local.

 

[Droogs]

Not sure if anyone has seen it, but Johns Hopkins University has a very good map showing how many cases in each country, recovery rates etc. It really knocks it home to see how few cases Russia is reporting - either they've escaped it or they're telling porkies...

Sent from my SM-G973F using Tapatalk

As 1 of my Electronic Warfare instructors used to say the Oxford Dictionary definition of "Lies/liar" was "A Russian"

 

[MikeG.]

.........but in the space of a couple of months enough evidence existed to pick up that error specifically.....

Indeed. That's precisely what happened.

 

[Andy Kev.]

Not sure if anyone has seen it, but Johns Hopkins University has a very good map showing how many cases in each country, recovery rates etc. It really knocks it home to see how few cases Russia is reporting - either they've escaped it or they're telling porkies...

Sent from my SM-G973F using Tapatalk

As 1 of my Electronic Warfare instructors used to say the Oxford Dictionary definition of "Lies/liar" was "A Russian"

14, 13 or 9?

Me: Green Slime but I fled the Dark Side for the sunny uplands after 13.

 

[Jake]

For me, this "cock up" has been very revealing, and gratifying. It has made it abundantly clear that the government has followed the scientific advice it was given, scrupulously. That is very much a good thing. The science was initially flawed (well, the computer modelling), but the scientific community quickly corrected the error. This is science working the way it should. Thank goodness we live here and not under Trump, where science is adjusted for political reasons and scientists are routinely ignored.

There were multiple different teams with different models feeding into SAGE. A decision was made to rely on the flawed one, which, as the others are unlikely to have made the same sloppy mistake, must have been a decision based on the optimistic strategy it suggested, as an outlier which went against the epidemiological consensus. Not to audit the model properly in those circumstances is unforgivably sloppy and not "how science works". This was not some complicated science, it was one of the key critical assumptions in the model on which the whole strategy rested (ie what the NHS load would be, that needed to be managed in a mitigate policy).

There is no positive pro-government spin you can put on this. It was a catastrophic mistake which cannot now be reversed, which is why after 1 day of talking about suppression when the Imperial College paper was released they are now back to talking about compressing the peak and shifting it sideways (ie mitigation not suppression). Except now they know even Imperial College's modelling says they cannot achieve that, so it is just headlong into NHS capacity being massively exceeded, with no good science base for their policy.

 

[Jake]

Imagine the situation where government picks and chooses what advice to follow in circumstances like this....nightmare.

It did this. It preferred the negligent model over others because it liked the low intervention strategy. The latter preference is understandable and laudable even, but if you are picking an outlier non-consensus view on which to base policy there is an obvious duty to question why the model you like is producing results not reproduced by other modelling.

 

[Jake]

The advice will have been made and taken in good faith.

Good faith/bad faith is an honesty test, not a test of competence.