Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Corona Simulator

washingtonpost.com

Post reply on HN
Print view View original

31–40 of 81 posts

Re: Corona Simulator

#31

Post by totalZero »

gamegod wrote: post #28
glaive123 wrote: post #25

Earlier quoted context omitted.

except it may not be successful yet you are just taking for granted the data provided to you by an authoritarian state

You don't need to rely on the data by the Chinese government, for several reasons: 1) Lots of epidemiological papers about the Wuhan epidemic use data from phone carriers (not the government) to track the spread and make predictions, and they were generally pretty good with their forecasts. The numbers reported by the government are consistent with the forecasts from scientists. 2) The numbers reported by China are a…

I think the point is that you have to wait until the city gets back to normal to judge the efficacy of the lockdown.

Re: Corona Simulator

#32

Post by robotresearcher »

One thing about this sim bothers me. Having points bounce off each other means they migrate more slowly and increases the effective contact graph diameter. Bouncing points aren't meaningful for the system being modeled. Me staying at home doesn't restrict your movements: our decisions are independent.

IMHO he should turn off the collision model. The effect will still be there, but the logistic parameters will be different.

Re: Corona Simulator

#33

Post by totalZero »

cl42 wrote: post #7
raynguyen wrote: post #3

The thing is, with these models. Everyone will eventually catch the virus and either dies off or recovers. If this happens in real life, the consequences are dire and catastrophic. Social distancing and avoiding large groups will obviously save lives and we flatten the curve to allow for treatment of individuals that will require hospitalization. But once the novelty of social distancing wears off. Will the number of…

That’s actually not correct, and the point of the simulations is to show that. By social distancing, you reduce the likelihood of everyone getting the disease. The use of social distancing brings the probability that everyone gets infected at some point closer to 0.0 but doesn’t make it 0.0. It’s possible to have a result where you either have people who were never infected, or ones who are fully recovered. That’s wh…

The major error with the model used in the article is that, in real life, all the points will eventually start to move again.

Re: Corona Simulator

#34

Post by Fricken »

glaive123 wrote: post #25
nyolfen wrote: post #15

> “As Lawrence O. Gostin, a professor of global health law at Georgetown University, put it: “The truth is those kinds of lockdowns are very rare and never effective.”” uh, except... the one we all just watched be successfully implemented in wuhan?

except it may not be successful yet you are just taking for granted the data provided to you by an authoritarian state

It's uncanny, because the various testing regimes implemented in the places where there have been outbreaks ought to be providing wildly inconsistent data, but the data from around the world is all showing a strikingly similar pattern. Once there's a community outbreak, the starting gun is off and it plays out like clockwork. The best you can hope to do is gum up that clockwork with draconian quarantine measures, because the starting gun is off all over the continental United States. Wash hands across America.

Re: Corona Simulator

#35

Post by Rapzid »

There seems to be a very well defined separation between those who have a relatively high chance of becoming critically ill, and those who have a very low chance. Depending on what you are optimizing for it could be more important to keep those groups separated.

I wonder if having known that for weeks now has lead to better social distancing between those groups, even if interactions within the low risk group has stayed rather high. Could lead to a lot of spread with very little detection because we largely only test for and catch the really bad cases.

Re: Corona Simulator

#36

Post by maxerickson »

glaive123 wrote: post #27

The people who don't understand this stuff, aren't failing to understand it because no one wrote about it. It's because they are not white collar or college graduates and are not likely to read The Washington Post in the first place. My father is the perfect example of this. He's dyslexic and so he struggles to read detailed journalism and gets all of his information from cable TV news like Fox News.

There's really not a 1:1 relationship between "college" and "white collar", and "curiosity".

Plenty of people with their fancy college degree are incurious, and plenty of people that barely made it out of high school are curious.

It's an appealing shorthand, but it's not a great one.

Re: Corona Simulator

#37

Post by jdietrich »

nyolfen wrote: post #15

> “As Lawrence O. Gostin, a professor of global health law at Georgetown University, put it: “The truth is those kinds of lockdowns are very rare and never effective.”” uh, except... the one we all just watched be successfully implemented in wuhan?

>uh, except... the one we all just watched be successfully implemented in Wuhan?

China have successfully paused the epidemic, but they haven't stopped it. Unless we fully eradicate the disease globally (highly unlikely given the presence of asymptomatic carriers and the spread of the disease to low-income countries), they'll be back to the beginning of the exponential curve as soon as the lockdown is lifted. That's better than nothing, but it's a million miles from a solution.

Re: Corona Simulator

#38

Post by maxerickson »

azinman2 wrote: post #12

What I don’t get is that we keep hearing quarantine isn’t effective, yet it seems to be in China and South Korea: https://www.businessinsider.com/coronavirus-lockdowns-work-s...

Who is saying quarantine isn't effective?

Re: Corona Simulator

#39

Post by rsynnott »

raynguyen wrote: post #3

The thing is, with these models. Everyone will eventually catch the virus and either dies off or recovers. If this happens in real life, the consequences are dire and catastrophic. Social distancing and avoiding large groups will obviously save lives and we flatten the curve to allow for treatment of individuals that will require hospitalization. But once the novelty of social distancing wears off. Will the number of…

> The question is when will this virus go away? (if ever).

Probably never. It seems fairly likely there’ll be a vaccine in the next couple of years.

Re: Corona Simulator

#40

Post by hawkice »

hn_throwaway_99 wrote: post #20
nyolfen wrote: post #15

> “As Lawrence O. Gostin, a professor of global health law at Georgetown University, put it: “The truth is those kinds of lockdowns are very rare and never effective.”” uh, except... the one we all just watched be successfully implemented in wuhan?

Except what was really implemented in Wuhan was really not just quarantine, but was also an extreme form of social distancing, where essentially the whole province of Wuhan was locked down.

*Hubei
Post reply on HN