Page 61 of 82 FirstFirst ... 1151596061626371 ... LastLast
Results 1,801 to 1,830 of 2446

Thread: Disease and pandemics thread (because it's science)

  1. #1801
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by George View Post
    Thanks! That's educational and logical.

    As a novice at evolution, I tend to the basics. I humorously recall Darwin calling Kelvin an "odious specter" because Kelvin restricted the age of the Earth to less than 100M yrs. and that was too short a time in Darwin's view for mutations to form varieties that later form new species.
    Which is odd, because I guarantee there is no way anyone can say evolution couldn't happen that fast based on anything but the rate that it happened on Earth (which requires knowing the age of the Earth!). The reason it is impossible to know the rate that evolution can occur theoretically is our good friend the Anthropic Principle, which in this case says you have no idea how fast evolution can happen on a single planet where it happened until you know the probability distribution for the potential environmental constraints on such planets. We are not looking for a likely story, we are looking for the most likely story, because we can see it happened (and the universe is quite vast).

    The argument here is interesting and rarely appreciated. Let's say there are only 100 planets similar to Earth in the universe (where "similar to Earth" is tricky to define, but I mean has all the key ingredients that were relevant to our evolution here, such that we can expect similar conditions over that entire class), and let's say it takes on average, in that class, a few billion years for intelligent life to evolve. Then we could expect to say the universe must be about a few billion years old, because if it was much younger, we wouldn't expect to be able to be here, and if it was much older, we should have come and gone by now (or we should have evidence of long-lost civilizations, like the dinosaurs). So that would make Darwin's point seem reasonable.

    Now ask, what if there are 1015 planets very similar to Earth in the entire universe (perhaps most well beyond anything we can observe-- and the number would be way way higher if there is a multiverse!). If the mean time for evolution of intelligent life is a few billion years, can we now conclude that the universe must be on the order of a few billion years old? No, we cannot, because any calculation that uses both the assumption that we are generic, and the age of the universe, is an incorrect calculation. We can assume we are generic for those Earthlike places, or we can use the observed age of the universe, but we can never use both those elements in the same argument, because those assumptions could be rendered contradictory by additional constraints on the problem that we have not included because we don't know about!

    The key point is, arguments like the one Darwin used always assume more than just the knowledge of how evolution "should happen" in some generic case, they also assume knowledge of some kind of underlying distribution of possibilities. It all comes down to what Sherlock Holmes said: "When you have eliminated the impossible, whatever remains, however improbable, must be the truth." What happens is what is vastly more likely, but only after you have investigated all the constraints! The job of the detective is to investigate the constraints, not simply assume them. If it turns out to be true that stars suffer internal instabilities that routinely cause total extinctions every 10 million years (or if it is common for comets to do that in almost all other planetary systems), then we would have to know the underlying probability distribution of systems that don't create such common complete extinctions, versus the underlying probability of a few planets that manage to have a vastly accelerated evolution process, to know which type we should expect the Earth to be.

    Arguments about what we should expect always forget the need to know underlying probability distributions of all the externals, not just the effect under study-- I see it all the time, it's the failure to recognize the unknown unknowns in any Bayesian analysis. (End of off topic.)
    Last edited by Ken G; 2020-Apr-16 at 03:30 PM.

  2. #1802
    Join Date
    Sep 2003
    Posts
    13,162
    Quote Originally Posted by Strange View Post
    This article uses a linear scale to compare covid-19 deaths to various other causes: https://www.thenewatlantis.com/publi...ashes-not-like
    You are right, the visual impact is higher. But you do lose detail at the low end of the y-axis.
    That's a great graph at that link which makes a big impact for those who fail to see the proper comparison with the flu.
    We know time flies, we just can't see its wings.

  3. #1803
    Join Date
    Jun 2004
    Location
    The Great NorthWet
    Posts
    15,966
    538 and Zach Weinersmith of SMBC have teamed up for a comic explaining how Covid-19 models work. Or don't.

    ETA: 538 is a political site but there are no politics in the comic.
    Cum catapultae proscriptae erunt tum soli proscript catapultas habebunt.

  4. #1804
    Join Date
    Sep 2003
    Posts
    13,162
    Quote Originally Posted by Strange View Post
    This article uses a linear scale to compare covid-19 deaths to various other causes: https://www.thenewatlantis.com/publi...ashes-not-like
    You are right, the visual impact is higher. But you do lose detail at the low end of the y-axis.
    Oops, how did they come up with 40 deaths per million in their last week of data for the US? There were 13,631 deaths in the last 7 days, which is the highest so far and it's climbing, but that is only about 14 per thousand.
    We know time flies, we just can't see its wings.

  5. #1805
    Join Date
    Sep 2003
    Posts
    13,162
    Quote Originally Posted by Ken G View Post
    Arguments about what we should expect always forget the need to know underlying probability distributions of all the externals, not just the effect under study-- I see it all the time, it's the failure to recognize the unknown unknowns in any Bayesian analysis. (End of off topic.)
    Your post is worth exploring on another thread. I'll comment because I think you make a good case in distinguishing the difference between deductive reasoning (Sherlock) and abductive reasoning (Mrs. Hudson on a good day?). I'm more with Mrs. Hudson in this thread. I think too many in the media are with me (but don't see it) and it seems we don't have to go far to prove it, if my math is correct in reviewing that flu comparison graph.
    We know time flies, we just can't see its wings.

  6. #1806
    Join Date
    Sep 2003
    Posts
    13,162
    Quote Originally Posted by George View Post
    Oops, how did they come up with 40 deaths per million in their last week of data for the US? There were 13,631 deaths in the last 7 days, which is the highest so far and it's climbing, but that is only about 14 per thousand.
    Nevermind, my error.
    We know time flies, we just can't see its wings.

  7. #1807
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by grant hutchison View Post
    I think the scorecards aren't in yet. We won't actually know who got it right, who got it wrong, and who just plain got lucky, for a year at least, and it will probably take longer than that to add up all the mortality and morbidity resulting from the economic impact of actions taken to acutely moderate the disease, and the health impacts of various styles of lockdown.
    I agree the good results don't mean they weren't just lucky, but I think we can already know places that went very wrong. There's no way New York City is ever going to be glad they waited to lock down, it already seems clear that COViD-19 will be the leading cause of death in NYC in 2020, and the way things are going it might end up leading all other causes of death combined. That's got to be worse than whatever happens from lockdown. Of course one does not have the benefit of hindsight to know which US cities should have locked down sooner, but had they to do it again, both the US and the UK would have locked down a month sooner. There's no advantage to waiting until it is out of control and then doing it, the problem was there was not the recognition that it would ever be necessary. Now we know-- the longer you wait, the more draconian and the longer will be your lockdown. What this looks like sustainably, once you have things back under control, is what remains to be seen-- and I agree will take a long time to assess. But it sure looks like the answer is, just do everything South Korea is doing-- we have a roadmap for what works, I don't think we even need to wonder whether it's the right way to go, but it likely requires that the case load gets dramatically reduced first. And that's the part where earlier response would have been so much better, though it's just hindsight at this stage.

    And I don't just mean actions by governments, I mean the whole state of the rhetoric and buy-in from the common people. In the early stages, I heard so many times "but there are no deaths in our state as of yet", or "the case number in the US is still low." And every time I heard that, I thought, why do I need to be told that, every single time the disease gets mentioned? Do they really think that widespread panic is really the threat here? That's where I put a lot of the blame for letting this get out of hand-- failure to achieve early buy-in. South Korea has been so successful because of their history with epidemics, and the "once bitten twice shy" principle.
    Last edited by Ken G; 2020-Apr-16 at 04:35 PM.

  8. #1808
    Join Date
    Jul 2005
    Posts
    19,664
    I think you can say that if you overwhelm your health care system, either in its ability to cope with COVID or in its ability to cope with other life-threatening illnesses because of COVID, then you've done something wrong.
    But without knowing how this will end, we don't know how long countries are going to have to keep dealing with endemic COVID, so we don't know the final death toll.

    Grant Hutchison

  9. #1809
    Join Date
    Oct 2001
    Location
    Clear Lake City, TX
    Posts
    12,973
    Huh.

    Pregnant women without symptoms are testing positive for coronavirus, study says

    https://nypost.com/2020/04/14/pregna...navirus-study/

    NewYork-Presbyterian and Columbia University Medical Center screened more than 200 women for the illness upon admission between March 22 and April 4, according to the study.

    Among the 33 patients who tested positive, 29 of them had no symptoms.
    That's ~14% of the women tested who were positive but without symptoms. It's dangerous to extrapolate from such a small (and specific) "test group" but it is worrisome and indicates a need for more - and more widespread - testing.
    Never attribute to malice that which can be adequately explained by ignorance or stupidity.
    Isaac Asimov

    You know, the very powerful and the very stupid have one thing in common. They don’t alter their views to fit the facts. They alter the facts to fit their views.
    Doctor Who

    Moderation will be in purple.
    Rules for Posting to This Board

  10. #1810
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by Jim
    That's ~14% of the women tested who were positive but without symptoms. It's dangerous to extrapolate from such a small (and specific) "test group" but it is worrisome and indicates a need for more - and more widespread - testing.
    It's 14% of the full sample of women, apparently selected for testing only because they were pregnant. That is quite a high fraction. Is that because it's NYC where the disease is rampant, or might they have become infected in the hospital itself? Or it might mean pregnant women are more susceptible to getting infected, and were caught prior to getting symptoms. If that few ever get symptoms, it might mean pregnancy is protective of the symptoms of COVID-19, though you'd probably expect the opposite since pregnant women can have reduced immune responses to protect the fetus. I wonder how many samples of people they have that are tested for any reason independent from COVID-19, samples like that are very diagnostically significant but rarely done. (I just heard a state say they will soon have the ability to test everyone with symptoms, and I thought, that's good for tracing, but it would sure be nice to know the situation in the asymptomatic population.) We can't test everyone, but a statistically significant sample of randomly chosen people would sure be nice and wouldn't cost that many tests.
    Last edited by Ken G; 2020-Apr-16 at 06:17 PM.

  11. #1811
    Join Date
    Oct 2005
    Posts
    27,229
    I notice that France just announced their cases have "plateaued." I find that language significant-- most people say "peaked", suggesting an expected decline. I see a growing awareness that the long-term solutions may look more like flat plateaus than like SIR-type peak and decline. At the very least, they are saying they don't expect a sharp decline any time soon-- the plateaus seem to last much longer than SIR-type peaks, presumably because SIR-type models peak due to herd immunity not due to governmental measures. Why lockdowns produce long plateaus rather than declines is presumably because SIR peaks have herd immunity working for them, which is quite a powerful force, whereas lockdowns still have the disease spreading within households. If it's only that, then a decline should appear after it has run its course through households, and that should take about a month. The longer term issue is likely to be that as cases start to drop, so will the willingness to maintain draconian measures, so again it seems likely that plateaus, not drops, will be what the West should expect. I know there are already protests in the US against having lockdowns, so you can see the likelihood that case levels will ever be made to significantly drop. This is an issue for the medical modeling-- what assumptions are they making about human nature?

  12. #1812
    Join Date
    Jul 2005
    Posts
    19,664
    Here in the UK, we've been talking about looking for a plateau ever since initial control measures were introduced. Some journalists seem desperate for a "peak", but the replies they get from government scientific advisers always make it clear that that isn't what will happen (and it's now perfectly bleedin' obvious that it isn't what's happening). Then the next day they're back asking about a "peak" again, sometimes immediately after there has been a careful and clear data presentation explaining why there isn't a peak. (I haven't done a statistical analysis, but my perception is that there's a direct correlation between the idiocy of a journalist's question and the height of the moral high ground they try to adopt while asking it.)
    Another thing that's going on in many countries, just as they flatten out the cases-per-day curve, is that expanding testing is capturing community cases that would not previously have been picked up. That's often not clearly reported (and another reason why trying to compare CFR between countries is a bad idea), but if the goalposts for what constitutes a case are steadily being pushed into milder disease, then that's another reason for the curve to hang before it descends.

    Grant Hutchison

  13. #1813
    Join Date
    Oct 2005
    Posts
    27,229
    That's also a good point, in the US there are a lot more tests becoming available, so it stands to reason there will be more cases reported. What makes this tricky, though, is that places like South Korea had a lot of testing early (much more testing than the US did), yet their case numbers stayed low, in part because of the response that testing made possible. So it sounds like early testing reduces total cases, whereas ramping up testing after the horse has left the barn only makes the case numbers higher, without much positive impact on containing disease. It seems like all there is to do now is lock down and hang on as long as the economy can sustain, and then hope you can get the cases down to a point where you are not so overwhelmed by positive tests and you can actually do something about them. At that point, instead of nation-wide kinds of strategies, you might need to use "granular" strategies, where you can test-and-track in regions of low case numbers, and just lock down the explosive regions as long as you can tolerate. There might even be types of granular response within communities, where if you have explosions going off in nursing homes and packing plants, you can try to use effective measures in those areas that you don't use elsewhere. Maybe every person who has to come to work in a packed plant or a nursing home should have to be tested every single day, would that really be such a strain on testing compared to testing everyone who is symptomatic? I would tend to think we should treat everyone with symptoms as if they had the disease and not bother to test them, saving the tests for asymptomatic people who are identified as potential superspreaders. It is starting to look to me like something we are doing is just not as effective as it needs to be, given the associated economic fallout and how long it takes to show progress.

  14. #1814
    Join Date
    Dec 2011
    Posts
    3,577
    Quote Originally Posted by Ken G View Post
    ... I would tend to think we should treat everyone with symptoms as if they had the disease and not bother to test them, saving the tests for asymptomatic people who are identified as potential superspreaders.
    Its starting to look like 'asymptomatically infected' may actually be 'presymptomatically infected', which supports the strategy of treating everyone as though they are infected because after a week or so 'asymptomatic' becomes 'symptomatic' anyway.

    There just seems to be no easy escape from lockdowns unless a statistically significant rolling random sample is repeatedly tested over short periods of time (like: every week or so)?

  15. #1815
    Join Date
    Jul 2005
    Posts
    19,664
    And just to highlight the issue of "doing it right" versus "doing it wrong":
    Singapore, which a few weeks ago was touted as a great success story for early testing, tracing and lockdown, is now undergoing an exponential rise in its cases (https://www.worldometers.info/corona...try/singapore/)
    Whereas Sweden, portrayed as pretty much the daft laddie of Europe for its continuing refusal to lock down hard, is at present tracking mortality between Germany and South Korea (both lauded for early extensive testing and control), when compared at the same numerical stage of their epidemics (the graphs commence on the first day 50 deaths were reported):
    coronavirus mortality.jpg

    Grant Hutchison
    Last edited by grant hutchison; 2020-Apr-16 at 09:44 PM.

  16. #1816
    Join Date
    Sep 2004
    Posts
    4,916
    I was wondering what happened to Singapore. What caused the increase?
    Do good work. —Virgil Ivan "Gus" Grissom

  17. #1817
    Join Date
    Mar 2004
    Posts
    19,241
    Quote Originally Posted by Roger E. Moore View Post
    I was wondering what happened to Singapore. What caused the increase?
    Here is an opinion piece on some of the issues:

    https://www.scmp.com/week-asia/opini...navirus-crisis

    One item stated is that they have a good number of low income foreign workers in crowded housing, which makes social distancing harder and this became a hotspot. Another issue is they underestimated how easily it can spread, and it’s hard for most people to grasp the effects of fast exponential growth. Once the virus got ahead of their procedures, it started to rapidly grow. There is more, but those caught my eye. Also, I believe Singapore has a high population density in general.

    "The problem with quotes on the Internet is that it is hard to verify their authenticity." — Abraham Lincoln

    I say there is an invisible elf in my backyard. How do you prove that I am wrong?

    The Leif Ericson Cruiser

  18. #1818
    Join Date
    Jul 2005
    Posts
    19,664
    Quote Originally Posted by Roger E. Moore View Post
    I was wondering what happened to Singapore. What caused the increase?
    Demographics, basically. Lots of people in close proximity, even in lockdown. They may well get away with fewer deaths, because the hot-spots are mainly among younger people at present. But there's a large vulnerable elderly population in Singapore, too, unfortunately.
    In contrast, Sweden has the demographic luxury that more than 50% of its households are single-person residences--self-isolation becomes very effective in those circumstances.

    Grant Hutchison

  19. #1819
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by grant hutchison View Post
    And just to highlight the issue of "doing it right" versus "doing it wrong":
    Singapore, which a few weeks ago was touted as a great success story for early testing, tracing and lockdown, is now undergoing an exponential rise in its cases (https://www.worldometers.info/corona...try/singapore/)
    Whereas Sweden, portrayed as pretty much the daft laddie of Europe for its continuing refusal to lock down hard, is at present tracking mortality between Germany and South Korea (both lauded for early extensive testing and control), when compared at the same numerical stage of their epidemics (the graphs commence on the first day 50 deaths were reported):
    coronavirus mortality.jpg
    But raw death numbers make for an irrelevant comparison, what matters is deaths per million. That means all those numbers should be divided by the national populations, which puts Sweden about 4 times worse than Germany and at least 20 times worse than South Korea. I agree that there are other factors here than just what the national policies are, we have no idea how much weather matters and luck also (and demographics, as you say), but given the evidence we have, Sweden is doing much worse than Germany and spectacularly worse than South Korea (but then, everyone is). It remains to be seen if Sweden will continue to do badly, my guess is that they are functioning on pure wishful thinking at this point (as did almost everyone at that same stage). But I would love to see that same chart divided by national populations, I have no idea why that isn't what is clearly needed and I think it would show much more similarity between nations-- except the gold standards like South Korea.
    Last edited by Ken G; 2020-Apr-17 at 02:29 AM.

  20. #1820
    Join Date
    Oct 2005
    Posts
    27,229
    To follow up on that, what seems to me is that every western nation has taken the same approach, the same strategy. I just don't see evidence that they are choosing different strategies, since it is not a different strategy to do the same thing once it has been demonstrated to be needed, it's only a different strategy to either anticipate the need or wait until it cannot be denied. Sort of like climate change, actually. This is even true of every individual state within the US. They all do the same thing-- they wait until it is proven that some step needs to be taken, and then they take it, seemingly without noticing that it would have been much better to take that exact same step sooner, before it was proven to be necessary-- like shuttering the windows based on the hurricane forecast rather than because the window glass has already been blown in. So there are not different strategies being taken, there are just nations that have had those strategies forced on them by now, and those that will get there soon-- except for the very few nations that took effective action ahead of time.
    Last edited by Ken G; 2020-Apr-17 at 04:26 AM.

  21. #1821
    Join Date
    Dec 2011
    Posts
    3,577
    Quote Originally Posted by Ken G View Post
    But raw death numbers make for an irrelevant comparison, what matters is deaths per million. That means all those numbers should be divided by the national populations, which puts Sweden about 4 times worse than Germany and at least 20 times worse than South Korea. I agree that there are other factors here than just what the national policies are, we have no idea how much weather matters and luck also (and demographics, as you say), but given the evidence we have, Sweden is doing much worse than Germany and spectacularly worse than South Korea (but then, everyone is). It remains to be seen if Sweden will continue to do badly, my guess is that they are functioning on pure wishful thinking at this point (as did almost everyone at that same stage). But I would love to see that same chart divided by national populations, I have no idea why that isn't what is clearly needed and I think it would show much more similarity between nations-- except the gold standards like South Korea.
    This chart can be sorted by deaths per 1 million population.

    The worst is San Marino, Andorra then Belgium then Spain, Italy, France. UK is the 8th worse and Sweden is 11th. USA is 14th Germany is 22nd .. (at the moment).

  22. #1822
    Join Date
    Jun 2005
    Posts
    1,869
    It is sad to think how many thousands of tragically unimaginable deaths would have been prevented if travel bans were enforced just days earlier than they were when various Health-Authorities would advice against it by uttering nonsenses like "Viruses don't recognize borders".
    To future generations reading this in case of a potential pandemic, sometimes it is wiser to err on the side of caution, rather than act specifically based on past learnt "Science".

  23. #1823
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by Selfsim View Post
    This chart can be sorted by deaths per 1 million population.

    The worst is San Marino, Andorra then Belgium then Spain, Italy, France. UK is the 8th worse and Sweden is 11th. USA is 14th Germany is 22nd .. (at the moment).
    That's a start, but what you actually need are the curves that use a start time pegged to a fixed number of deaths per million. If you only look at cumulative deaths per million, you miss that some places simply had their infections start later, creating the illusion they are doing better rather than simply walking the same path a little while later. Normalizing to per million is then key on both the horizontal and vertical axes, to make fair comparisons. It's very hard to get that information because sources don't seem to understand statistics all that well. (You know, "liars, darn liars, and statistics") But it's really obvious that all case and death numbers should be normalized to the populations sampled, that's just a no-brainer, yet virtually never done. I even see the total numbers for my own state, compared directly to the numbers in states with ten times the population! Yeesh.
    Last edited by Ken G; 2020-Apr-17 at 06:22 AM.

  24. #1824
    Join Date
    Dec 2011
    Location
    Very near, yet so far away
    Posts
    397
    Quote Originally Posted by Ken G View Post
    That's a start, but what you actually need are the curves that use a start time pegged to a fixed number of deaths per million. If you only look at cumulative deaths per million, you miss that some places simply had their infections start later, creating the illusion they are doing better rather than simply walking the same path a little while later. Normalizing to per million is then key on both the horizontal and vertical axes, to make fair comparisons. It's very hard to get that information because sources don't seem to understand statistics all that well. (You know, "liars, darn liars, and statistics") But it's really obvious that all case and death numbers should be normalized to the populations sampled, that's just a no-brainer, yet virtually never done. I even see the total numbers for my own state, compared directly to the numbers in states with ten times the population! Yeesh.
    Deaths per million is not a good comparison in my opinion . Even within a country there is a huge difference between population density in a city or the countryside. Using it as a comparison across countries with differing densities, different cultures and different approaches due to those mentioned differences leads to no useful data. Widely separated people are at less risk so being included in the nearest million people leads to mistakes on effectiveness of policy.

  25. #1825
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    9,237
    Deaths per hospital is the driver of policy, governments do not want deaths to exceed critical care facilities. Those are generally designed knowing the annual cycle of deaths, now they have to be ramped up in equipment and staff. It’s sad that these extreme interventions do not seem to save everybody, but the reaction is worse when people die without treatment. Those countries well prepared to contact trace and isolate early in the epidemic get much better results, for the others it’s lock down and wait for that drastic measure to bring control. Cities are bound to fare worse, but they should have more hospital beds waiting. Old people in care homes clearly are more vulnerable because of the social contact, and they are dying in greater proportion, often with no interventions at all, but protecting them would not hurt the economy, if there were to be a prior plan for inevitable epidemics. I do hope lessons are learned because this will not be the last epidemic.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  26. #1826
    Join Date
    Jul 2005
    Posts
    19,664
    Deaths per million is useful measure of the impact on society. But the trend in number of deaths pegged to a common datum is a useful measure of how well a society is flattening the curve--how long between the first fifty-death day and the first hundred-death day? Dividing that by the population just makes big nations look like they're doing well.
    Other useful measures are COVID hospital bed occupancy and ICU bed occupancy. These data are nationally reported at midnight in the UK, using a mechanism that's been in place for many decades, so they're immediate and pretty accurate (especially so for COVID, when one only needs to count the number of beds occupied in the area set aside for COVID), and (so long as resources are not overwhelmed) the case severities stay much the same from day to day.

    Grant Hutchison

  27. #1827
    Join Date
    Sep 2004
    Posts
    4,916
    Quote Originally Posted by Ken G View Post
    But raw death numbers make for an irrelevant comparison, what matters is deaths per million. That means all those numbers should be divided by the national populations, which puts Sweden about 4 times worse than Germany and at least 20 times worse than South Korea.
    While I see Grant's point that this makes big nations look good, what about graphing it on a log-10 scale? How would that look? Would it be meaningful?
    Do good work. —Virgil Ivan "Gus" Grissom

  28. #1828
    Join Date
    Jul 2005
    Posts
    19,664
    Quote Originally Posted by Roger E. Moore View Post
    While I see Grant's point that this makes big nations look good, what about graphing it on a log-10 scale? How would that look? Would it be meaningful?
    Log scales are commonly used, and (like all methods of data presentation) have their advantages and disadvantages. The slope of the log graph reflects doubling time--and if you're interested in doubling time, and the graph is appropriately marked up, then a log scale is useful.
    Here's one with doubling time slopes indicated, as an example of what I mean:
    ScotlandCumulativeCasesByRegion16April2020.svg.png
    The trouble is that people sometimes interpret the sectors between the doubling time lines as being a measure of doubling time, rather than looking at the gradients.
    Different graphs do different jobs, in other words, and there's no universal "correct" way of displaying data.

    Grant Hutchison

  29. #1829
    Join Date
    Apr 2020
    Posts
    46
    Quote Originally Posted by Roger E. Moore View Post
    I was wondering what happened to Singapore. What caused the increase?
    A few weeks ago, it looked like this thing was under control in Singapore, but then it increased dramatically.

    In the current wave, most of the people infected are foreign workers. There are a lot of foreign construction workers, living in shared facilities, close quarters, etc. Ideal setting for the spread of a virus.

    https://www.channelnewsasia.com/news...rkers-12652286

  30. #1830
    Join Date
    May 2008
    Location
    The Netherlands
    Posts
    15,669
    Quote Originally Posted by Ken G View Post
    (End of off topic.)
    When there have already been several warnings to stay on-topic, and yet you willfully post an off-topic that's bound to create more off-topic discussion, we have no choice but to issue infractions. And since this isn't the first time either, it leads to a suspension.

    Everyone, please, please stay on-topic. This thread is already difficult enough to manage with such a high-interest issue going on.
    ____________
    "Dumb all over, a little ugly on the side." -- Frank Zappa
    "Your right to hold an opinion is not being contested. Your expectation that it be taken seriously is." -- Jason Thompson
    "This is really very simple, but unfortunately it's very complicated." -- publius

    Moderator comments in this color | Get moderator attention using the lower left icon:
    Recommended reading: Forum Rules * Forum FAQs * Conspiracy Theory Advice * Alternate Theory Advocates Advice

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •