Outbreak – playable simulations of a disease outbreak | Hacker News Reader

Outbreak – playable simulations of a disease outbreak

meltingasphalt.com

355 pointsgringoDan66 comments

Remember pandemic2 flash game?

It was almost impossible to kill all humans without starting the infection from Madagascar. If you started from somewhere else Madagascar always had time to close the borders.

Lo and behold: Madagascar is one of the few countries without infections in Global Cases tracker. https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.h...

And greenland as I recall? But yeah - thikn the trick was low mortality so they took too long to respond

Long incubation period, start in southeast Asia, airborne, create delays for the scientists...

Still playable so long as flash still exists:

https://www.newgrounds.com/portal/view/448950

I remember that game so well. Your get so far and then Madagascar closes it's borders and it's game over for you.

Turns out it's true.

The other takeaway from that game was that stealth symptoms with low fatality does the best out of any pandemic.

Well everyone gets the common cold!

We're also getting another classic part of any Pandemic play through cruise ships forever circling the ocean looking for a place to dock.

https://www.cnn.com/travel/article/cruise-transatlantic-coro...

My favourite cruise ship fiasco this week has been the newly imposed rule in New Zealand - they can’t dock, go away. So hours before the ban was imposed one docked and dumped all it’s passengers in the capital. https://i.stuff.co.nz/national/health/coronavirus/120298962/...

I got some stern comments that "laws should never apply retroactively" for saying this, but I still believe: all those border closings should apply immediately from the second they're announced, and all quarantine requirements for returning passengers should apply retroactively, from a day or two before they were announced.

Because the story is the same all over the world. A country announces it'll close the borders in 2 days. A day after the announcement, every citizen of that country that was in COVID-19 hotspot returns to their homeland. In some cases avoiding quarantine.

Kind of related problem... A major city becomes an outbreak hot spot. The city enacts strict quarantine recommendations but they are not enforced by law and domestic travel is still permitted. People then flee the hot spot city and rapidly spread disease to other areas. This was a mistake Italy made that the US is repeating in places like NY, San Francisco, etc.

The border closings need to give notice, it doesn't really matter if some people notice and respond early to take advantage of the "loophole" because ultimately it's about controlling the spread, not stopping 100% (which is unrealistic)

The issue with immediate border closings is they create a lot of confusion. What happens to planes in the air or taking off or who does the policy actually apply to (see the initial iterations of the muslim ban where even the border agencies didn't have a consistent view of their own policy. The policy has to get out to the people at the border actually enforcing it too so it's either distributed early (meaning it will probably leak early and cause a rush then) or they only learn about it when everyone else does causing issues.

You think that's bad?

A cruise with 200 people was denied port in Portugal, so the ship went to Spain, dropped all the passengers and they took busses to enter Portugal.

One of the many reasons why the border is being shut

And lo and behold, again: Madagascar has just closed their borders.

https://www.nytimes.com/reuters/2020/03/14/world/africa/14re...

> "To prevent the outbreak entering in Madagascar, all flights connecting Madagascar to Europe are suspended for 30 days," Madagascar President Hery Rajaonarimampianina said in a statement.

> Madagascar, one of the world's poorest nations where malnutrition is rife and outbreaks of deadly diseases are common, will also suspend air links to the nearby islands of La Reunion and Mayotte, he said.

https://pics.me.me/president-madagascar-a-man-in-brazil-is-c...

Or Plague Inc. on ios or android

Bullfrog's classic game "Theme Hospital" had really great emergent vomit cascades.

https://www.youtube.com/watch?v=Y69QTjTvp1w

Two Point Hospital does fantastically as a spiritual successor to Theme Hospital. https://www.youtube.com/watch?v=jmlaOYMU8qA

Cool, thank you! I'll check in and check it out. (Installing it now!) Does it have a "Bloaty Head" treatment? ;)

https://www.youtube.com/watch?v=Le_znuXcP2M

I played a LOT of Theme Hospital when working on The Sims 1, and aspired to make the Sims editing tools as easy to use and "clicky" as Theme Hospital was. That and some of Peter Molyneux's older games like Dungeon Keeper, with architectural editing and a lot of independent characters running around at the same time, had a lot of influence on The Sims.

“emergent vomit cascades” gave me the laugh I needed today.

I am not a huge fan of this (or the WaPo) simulator, since I they seem to have chosen their models for the nice look (regular grid, bouncing balls) instead of for their accuracy.

While you can give people a rough idea of how different containment efforts will work, the models are so far removed from reality that I don't think that it really helps very much. Especially the WaPos'comparison' between different containment measures is IMHO borderline negligent without putting huge caveats in front of them.

Given how much normal people know about these issues (nothing whatsoever, or even less than nothing, somehow) I think the shortcomings of their model, vs. full physical simulation, is of no importance. It's a fantastic intuition-builder.

If we're talking about disease-based games, I highly recommend Pathologic 2. Nothing realistic there in terms of modeling disease spread or cure mechanics, it's a very artistic and somewhat abstract game, but it conveys certain social realities of being in a contagion zone really well. Among other things, it show the importance of not collectively freaking out.

I second this recommendation. I'm currently playing through Pathologic 2 for the first time. Playing through the game in the wider context of a pandemic makes for a very powerful experience.

How can I model the type of interaction between people. For example :is school a stronger vector of propagation than workplace ?

I ask because everybody in my country talks about what to stop (restaurants, culture,...) but never explain why working is considered less of a problem... (I have my little idea but well, ain't sure)

when economy shuts down, more people die of malnutrition than of the virus.

Covid19 doesn't seem deadly enough to get us to that stage, thankfully. Unlike MERS/SARS/Ebola.

School is a bigger issues because children are usually are much closer together, are in the same room with far more people, and have far more interaction that adults at a typical job.

Tell a child to stop touching things and see how many things they touch, where they put their fingers, how many times they pick their nose etc and then you'll know.

It'll only take a few minutes.

My friend was in an airport recently. A mother dutifully used hand sanitizer on her 3 year old's hands. Within one minute the child had put their hands on every surface within a meter: arm rests, seat cushions, under the seat where the boogers and gum are, the floor, etc.

This https://www.pnas.org/content/116/27/13174 is a good paper, it gives numbers from actual school closures (in Russia), compares with areas where schools weren't closed, creates models for some scenarios too.

It would be interesting, possibly instructive to take a SIMS town and use it to model an epidemic.

Thx for the article. I've skimmed through it. Looks like closing schools averts 30% of cases. Not bad. But interestingly, in my country, I've never heard of a school closed because of influenza...

Apparently school closures work very well because it also forces adults to stay home from work to take care of their children.

Source: https://samharris.org/podcasts/190-respond-coronavirus/

You should totally add this to https://coronavirustechhandbook.com/

Super fun: In the full model { 5%, 3%, 5, 3, 0.25, 5, 14 } led to a walking sprawl of disease in a healthy population that lasted a long time, turning down the incubation days by 2 made it slower. The rate of new infections eventually eventually tapered off because the virus encountered its own sprawl like a growing structure in Go. Seems like a STD.

We had a small demo to simulate pandemic outbreak at malls. The simulation was mostly for demonstration purposes to illustrate how distancing impacts spread, and potentially some ways to help mitigate it.

http://socialdistance.ai/

Much less visual than yours though, I also made a tool to highlight the exponential effect of social distancing.

Exponential growth is often counter-intuitive and lot of people don't seem to get it.

https://www.spreadsim.com/

This is cool but your hospital beds figure needs to be reduced dramatically because I believe you have taken the number of total hospital beds, not intensive care beds for the critically ill. In my country, the number of total beds is about 48 times the number of intensive care beds.

This is missing a couple things:

1) Cells should have random demographics such as male/female and age, and a virus could respond differently to them

2) The virus should have some capability for mutation as it spreads. (probability of changing variables from the initial configuration by some percent)

Great work. Is the code available to look at? I see its a Creative Commons license but I cant find a link to a code repo.

This stuff drives me crazy. I know the author of this (and of the linked Washington Post article) are trying to help people gain clarity, but it worries me that it only serves to accelerate fear and panic.

These articles are making it sound as if the exponential part of the curve will just continue on forever until it consumes the whole population. The (self-admittedly) overly simplified simulations aren't helping either.

Front this article >

And here's the kicker: even if we manage to "flatten the curve" enough to meaningfully space out the case load, we're still positioned to lose millions and millions of lives.

From the WaPo article >

This so-called exponential curve has experts worried. If the number of cases were to continue to double every three days, there would be about a hundred million cases in the United States by May.

Still, without any measures to slow it down, covid-19 will continue to spread exponentially for months. To understand why, it is instructive to simulate the spread of a fake disease through a population.

It will continue to spread exponentially for months!? Millions & millions of cases & or deaths? These are really sloppy and dangerous ideas to put in peoples head. They are just plain wrong, and not how pandemics play out in reality.

Here is what the CDC says about how pandemics spread (Under the 'COVID-19 Now a Pandemic' heading) https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/summ...

Pandemics of respiratory disease follow a certain progression outlined in a “Pandemic Intervals Framework.” Pandemics begin with an investigation phase, followed by recognition, initiation, and acceleration phases. The peak of illnesses occurs at the end of the acceleration phase, which is followed by a deceleration phase, during which there is a decrease in illnesses. Different countries can be in different phases of the pandemic at any point in time and different parts of the same country can also be in different phases of a pandemic.

Different parts of the country are seeing different levels of COVID-19 activity. The United States nationally is currently in the initiation phases, but states where community spread is occurring are in the acceleration phase. The duration and severity of each phase can vary depending on the characteristics of the virus and the public health response.

So no, it will NOT continue exponentially for months. Especially given the amount of isolation/ quarantine measures and canceling of large events taking place. I think it is reasonable to expect these measures will help us progress through to the deceleration phase faster.

Further-more, there is more and more evidence coming out suggesting the in China where it started, they have already entered the deceleration phase. This should be encouraging news for us in USA where it is just beginning.

https://www.aljazeera.com/programmes/newsfeed/2020/03/big-dr...

What happens once China starts dismantling mitigations currently in place, given that the majority of the population is still naive to the virus? Is it reasonable to expect they can wait out the virus disappearing completely by tracking all active cases until they have recovered?

To try to answer my own question: it may be reasonable, however the problem of imported cases from other countries not doing the same still remains. This can restart the epidemic anew.

Yeah, I mean if they entered into another acceleration phase, that would be terrible. I suppose it possible another region in China not hit as hard still might enter one.

I guess we just have to trust they will keep things under wraps long enough.

It's probably better to err on the side of caution than the alternative.

The CDC link is interesting stuff I haven't read before, but where does it say the acceleration phase stops before it reaches a significant percentage of the population? China contained and stopped the virus, it didn't just naturally decelerate.

Yeah, that CDC page is one of the best resources I’ve seen. Explains everyone great detail.

I didn’t mean to make it sound like it happens naturally. I imagine it probably only happens through human intervention. It just says ” The duration and severity of each phase can vary depending on the characteristics of the virus and the public health response.

So, as we are entering the early stages, the public health response in the US right now is massive. Unlike anything most of us have seen in our lifetime. Except for the west coast, NYC, and a few others, the spread so far is still very low. Knowing that, and now seeing China enter deceleration, gives me hope we will minimize the duration and severity of the outbreak.

But on the other hand, I am more worried about the economic fallout. The aggressive response with isolation & canceling events, may be necessary, but it will cause many businesses to fail and many folks to lose their jobs.

There is nothing more important than human life.

That is certainly true, but the economy is connected to human life, in obvious (food to eat, medical insurance) and non-obvious ways (e.g. job loss linked to suicide https://www.theguardian.com/society/2015/feb/11/unemployment... )

So just stopping economic activity also has health downsides.

Ok, a reply to my own rant, LOL.

To be fair, the WaPo article is laced with caveats like ”without any measures to slow it down”, or ”If the number of cases were to continue to double every three days”. But there ARE measures to slow it down, and it certainly WON’T continue to double every three days.

This is really my gripe with these articles. I like the tech. The interactive ’simulations’ are cool & fun. But I worry they are only adding to the panic and confusion. I’ve had close friends spout ‘facts’ like this to me. Stuff like 60-70 of the population will be infected by the time it’s done.

The UK government says worst case 80% of people are infected. That is absolute worst case, based on their modelling, so highly unlikely, but it is coming from the Chief Medical Officer himself. It is not panic.

This is assuming that we don't take quick and drastic action.

Which is where we were as a country less than a week ago, and it didn't seem inevitable at the time that we were going to take this as seriously as we should be.

These articles and tools warn of what will happen if we take the wrong path, and if they serve their purpose successfully will look overblown and unnecessarily alarmist once this is over.

Very cool. Shouldn't the input transmission rate also be fixed for the last "test" ? It says to try to flatten the curve to reduce the % dead, but being able to control the transmission rate is cheating.

One of the first downloadable objects for The Sims 1 was a pet guinea pig, which literally had a downloadable computer virus.

If you neglected to feed and care for your guinea pig properly, it would bite you and give you a deadly contagious virus, which you could spread to other Sims by not following proper hygiene and getting enough rest, and even die from.

Kids who downloaded the binary Sims object from the internet and installed in their game were horrified that their beloved Sims dropped dead unexpectedly, but they learned a useful lesson.

Now deadly infectious guinea pigs are a tradition continued in later versions of The Sims!

The Sims Pie Menus

https://medium.com/@donhopkins/the-sims-pie-menus-49ca02a74d...

>Viruses

>You can download this guinea pig object, and if you don’t care for it it will make this secret hidden virus object that gives you a cold, and has with it the animations and sounds of coughing and going (cough cough).

>Every once in a while it will just interrupt what you do and cough, and it will be bad for your health. And if you don’t get enough sleep, you have to get sleep to get rid of it. This little program is literally a virus that runs and lives in your household and in your characters, and your characters can spread them to the neighbors, and they’ll bring them home to their families.

>Programmable Plug-In Objects

>Anyway, there’s quite a lot of interesting potential for what new plug-ins additions can do because of this programming languages.

>SimAntics Visual Programming Language

>There’s a built-in visual programming language called SimAntics, that is a control flow decision tree type of language. So the “Come and See” object, we’ll look at that.

https://www.youtube.com/watch?v=-exdu4ETscs&t=12m23s

Something Is Killing the Sims, and It's No Accident. By John Markoff, April 27, 2000.

https://www.nytimes.com/2000/04/27/technology/something-is-k...

Lethal guinea pig kills virtual people

http://news.bbc.co.uk/2/hi/sci/tech/746700.stm

The Sims 1: Guinea Pig Disease

https://www.youtube.com/watch?v=5O5cgXuqhnA

The Sims / Illness / Guinea Pig Disease

https://sims.fandom.com/wiki/Illness

How to Get and Treat Guinea Pig Disease

https://sims-online.com/how-to-get-treat-guinea-pig-disease/

The Sims 4 My First Pet: SimGuruGraham Talks “Guinea Pig” Disease

https://simsvip.com/2018/03/07/the-sims-4-my-first-pet-simgu...

A powerful boss in World of Warcraft could afflict the player with a plague called "Corrupted Blood", which rapidly drained your HP and could spread to nearby allies. The negative status effect was supposed to apply only in its zone of origin, but due to the way it was implemented, a player accidentally spread it outside its containment zone by dismissing and then re-summoning a pet with the effect, allowing it to spread. All hell broke loose on Azeroth, as entire cities would be contaminated with Corrupted Blood, and were subsequently abandoned or avoided by the remaining healthy players. Some players with healer-class characters would heal their allies and random strangers until the affliction passed. Others deliberately spread the affliction as a form of griefing. It was bananas, and epidemiologists used it as a model for possible human behavior in a real-life plague.

See: https://en.wikipedia.org/wiki/Corrupted_Blood_incident

Heh I published something very similar this morning. Looks like everyone has the same idea

https://reactfordataviz.com/articles/corona-simulation/

Interesting simulation but very slow at showing any effect. After 3500+ iterations (several minutes) there were still only 5 deaths in the first simulator (without distancing), all in the first iterations. I suggest to find a way to make it clear when it's done or speed it up to get to the point in one minute max.

That's weird, I get many deaths in just a few iterations. What were you running on? You seem to have found a bug.

Here's a video of when I run these: https://twitter.com/Swizec/status/1239599601537015808

Parent is probably referring to the "Covid19 with distancing" simulation.

Some feedback: these are very simplistic simulations and obviously not aiming for accuracy but please consider changing the default 20% "reinfectability" parameter.

All evidence and medical knowledge indicates that reports of reinfection are either:

- issues with testing (false positives/negatives)

- extremely rare

https://www.snopes.com/fact-check/covid-19-reinfection/

It's scary enough without reinfection. Seems unlikely.

> the 30,000-letter genetic code of the virus changes by one letter every 15 days [from link]

But could be like the common cold and flu, with new strains each year - but more contagious and deadly. Even then, we could still have new vaccines each year, and better treatment methods.

I want to be able to tailor survival rate according to availability of hospital support - it seems like an obvious thing to include.

e.g. say that there are X hospital beds, and an X% survival rate for critical patients if they can get a bed.

Anybody know an open source version of something like Plague Inc: Evolved ? I sure as hell am not playing computer games now. I'm going to get my crisis preparedness up properly and try to help my community. But it would be an interesting thing to have for the quarantine phase...

I'm a fan of Plague Inc on iOS and have unlocked all the pathogens...what's the difference between Plague Inc and Plague Inc Evolved?

Is the latter worth buying and playing?

Also: if there's a design doc out there for Plague Inc or a similar game, I'd love to see it.

Cool. I covered all the pathogens too, though I think I had quite a few genes left to unlock.

I've never played the original Plague Inc, so I can't tell you the difference from experience, but according to Wikipedia (https://en.wikipedia.org/wiki/Plague_Inc:_Evolved , isn't it amazing that it has its own Wikipedia page?):

The core game of Plague Inc: Evolved is the same as Plague Inc. - the player controls a plague which has infected patient zero. The player must infect and kill the whole world population by evolving the plague and adapting to various environments. However, there is a time pressure to complete the game before humans, the opponent, develop a cure for the plague. [...] It also includes all of the expansion packs from the original Plague Inc. game

Great game concept, though still with some untapped potential in my opinion. I keep getting inspired to replay it by the news right now, but we really have much, much more urgent things to attend to. Prepare, and help out, but stay safe!

A "design doc" would be the virtual lab where you can design your own scenario for Plague Inc Evolved. It is included in the Plague Inc Evolved game itself.

Reply on news.ycombinator.com