That time World of Warcraft helped epidemiologists model an outbreak

Revisiting a seminal 2007 paper modeling the <em>WoW</em>'s Corrupted Blood incident

Read the whole story

 

[Meathim]

Was waiting for something on that, it's a very interesting event.

Incidentally, got back to WoW (been off since Dreanor) and it is so bewilderingly stupid these days. But Classic is awesome, it is so much better than whatever is going on now.

 

[glutto]

Did they have toilet paper in WoW?

 

[AggroBoy]

Typo in the instance name: it was Zul'Gurub, not Zul'Burub

 

[ManuOtaku]

I do wonder if there was also people with the attitude “I don’t take responsibility at all,” like in real life.

 

[Voo42]

Oh that was back in the days when I played. It was hilarious to watch.

 

[carrellk]

I remember this. Mostly I was annoyed by the folks who deliberately acted to spread it. As a NE Rogue, I spent a lot of time skulking around trying to avoid infection. I called my wife (the Doc) in to observe what was going on. She hated WoW, but thought the plague was interesting. Stupid Docs.

And yes, there are people today who oppose trying to stop or slow-down COVID-19, ignoring the consequences of having yet another endemic disease resident in the human population.

And yes, there are countries which allow exotic meat markets, whether it is out of dietary need or to support traditional medicine, without regard to the hazard this practice creates for all of humanity. There are strong reasons to believe that SARS, MERS, and now SARS2 (COVID-19) are all the result of exotic meat markets.

Nihilistic and selfish behavior is alive and well outside of games.

 

[0xD34DC0DE]

I don’t think real-world parallels can be drawn when comparing real life where you have a single life and games where you can die and run back to your corpse and revive... just like FPSs and real-world combat parallels can’t be drawn. When the player knows there are little consequences to death, they don’t mind doing stupid things.

 

[Vala]

This always seems to come up, but I think people behave quite differently when it's a game than in reality. In these write-ups it's often talked about as though it was an accidentally spread epidemic, but my experience was definitely that it was 99% deliberate.

Sure, someone found out it worked that way by accident. The fact that hunters would dismiss their pets with the debuff only to re-summon it in town again, and again, and again wiping out all the nearby low-level players repeatedly was not an accident.

I just don't see that many parallels; the equivalent would be if large swathes of people traveled to Wuhan for the express purpose of catching the virus to bring it to capitol cities around the world, rapidly wiping out large portions of the city then doing it again a few hours later. Additionally the virus itself would have to be practically incapable of spreading accidentally; it's vastly different.

 

[TheArsTrev]

I doubt that hoarding and hiding will prevent hardly anyone from contracting it. The only real question is how hard will it hit you.

Fair...they've been saying all along that 70 to 80 percent of us will get it...most with mild symptoms...but 'hoarding and hiding' will slow down how fast we reach 70 to 80 percent, thus giving our medical teams time to deal with the x% that need help because it wasn't mild.

 

[rosen380]

" In the end, at least three servers were affected, and Blizzard had to reboot the entire game to correct the problem."

Maybe I'm severely missing the mark in how these games work on the backend, but I'd think that most player attributes were essentially stored in a database.

UPDATE [WOW].[dbo].[player_attributes]
SET [CorruptedBlood] = 0;

?

 

[Deleted member 274645]

Some article corrections. The initial infection was not due to players teleporting out; there were several mechanisms in place which prevented this. Dungeons were separated from the wider world by "instancing"; once you enter that dungeon, you generally cannot interact with the wider world until exiting it and you are not permitted to exit while in combat.

Additionally, once a boss monster has become aggressive towards you ("gained aggro") by either being attacked or approached too closely, it will remain so until either you and your party or dead or until you slay it. And you are considered "in combat" for the duration, so you are not permitted to leave the dungeon "instance".

This programming worked correctly, so there was generally no way for the corrupted blood disease to spread. Except for hunters.

Hunters had as one of their skills the ability to tame "pet" animals, and then summon or dismiss them even in combat. Generally whatever their status was when dismissed, it remained so when resummoned later.

One hunter facing Hakkar with their party had the bright idea to get the pet infected, and then dismiss it. The party died to hakkar, and when they revived he then went to a major human city and summoned his pet... who was infected, and promptly started spreading the infection.

---

One other correction. Blizzard did not "reset the game". WoW is divided into dozens of servers, which run their own instances of the world. What happens on one server never affects another (barring character transfers, which are a lengthy process). Three servers were affected, so those three servers were reset, clearing the diseases and ending the epidemic. Pets were later made immune to corrupted blood so that it could not happen again.

 

[Deleted member 274645]

" In the end, at least three servers were affected, and Blizzard had to reboot the entire game to correct the problem."

Maybe I'm severely missing the mark in how these games work on the backend, but I'd think that most player attributes were essentially stored in a database.

UPDATE [WOW].[dbo].[player_attributes]
SET [CorruptedBlood] = 0;

?

Im not sure that a database is the best way to store lists of debuffs, which are transient and rapidly changing. Even if it were, this would ruin in-progress Hakkar encounters.

 

[torp]

" In the end, at least three servers were affected, and Blizzard had to reboot the entire game to correct the problem."

Maybe I'm severely missing the mark in how these games work on the backend, but I'd think that most player attributes were essentially stored in a database.

UPDATE [WOW].[dbo].[player_attributes]
SET [CorruptedBlood] = 0;

?

Im not sure that a database is the best way to store lists of debuffs, which are transient and rapidly changing. Even if it were, this would ruin in-progress Hakkar encounters.

If you pay attention to what resets on maintenance in a MMO, you'll notice that most buffs and debuffs are not saved and only kept in RAM. MMO devs have it easy with the database stuff.

 

[aksdb]

Did they have toilet paper in WoW?

Only before the outbreak.

 

[Null_Space]

I doubt that hoarding and hiding will prevent hardly anyone from contracting it. The only real question is how hard will it hit you.

In all this talk about “flattening the curve” I haven’t seen any discussion or analysis on how it might/will change R naught.

I would think that it would have to.

If I’m presymptomatic and previously would have had , say, 300 person-hours of close contact while contagious. And now I’ll have 10.

Wouldn’t that reduce R naught by ~30X? At a statistical/population level?

What am I missing?

 

[Elektriktoad]

One hunter facing Hakkar with their party had the bright idea to get the pet infected, and then dismiss it. The party died to hakkar, and when they revived he then went to a major human city and summoned his pet... who was infected, and promptly started spreading the infection.

So exotic meat started the WoW epidemic too. Too bad we can't patch our meatspace the same way Blizzard did.

 

[Pusher of Buttons]

We need Kazzak to purge the city

https://youtu.be/fWZyw1xRY58

 

[tenocticatl]

I doubt that hoarding and hiding will prevent hardly anyone from contracting it. The only real question is how hard will it hit you.

In all this talk about “flattening the curve” I haven’t seen any discussion or analysis on how it might/will change R naught.

I would think that it would have to.

If I’m presymptomatic and previously would have had , say, 300 person-hours of close contact while contagious. And now I’ll have 10.

Wouldn’t that reduce R naught by ~30X? At a statistical/population level?

What am I missing?

These statistics are (for now at least) empirical, so I would expect so. The Washington Post had an interesting article about the effect of social distancing that included some simple but illustrative simulations: https://www.washingtonpost.com/graphics ... simulator/

 

[kaidenshi]

(Long, detailed, factual response)

THANK YOU. I haven't played WoW since 2007 and I was never on one of the affected realms, but most of the article didn't mesh with my memory of the event; your description does.

 

[SimonRev]

I really feel like the author overdid the scare quotes in paragraphs 2-4. Do we really need quotes around corpse in "virtual 'corpse'" to know that we aren't talking about real corpses? Also in the context of video games, things like "hit points" and to "die" and even "end boss" are standard terms and shouldn't need scare quotes to indicate that the words are being used in a non-standard fashion.

 

[Boskone]

Just a note: the link for hoarding was actually price-gouging.

I don't think it's fair to equate panic hoarding with deliberately driving up prices. (Even though neither is helpful.)

 

[Defenestrar]

Go Cougs!

 

[A_Very_Tired_Geek]

I do wonder if there was also people with the attitude “I don’t take responsibility at all,” like in real life.

What responsibility? It's a virtual world and a game at that. Most people didn't give much of a damn and those that were ultra anal probably quit.

I was playing WoW at the time and I barely even remember such an event. I also didn't much care about it at the time. It was more of an annoyance than anything else.

Say it after me, "fiction is not real life" no matter how accurate the dev team tried to make it. Take a big deep breath. Let it out.

This is real life now, there are real stakes. End results matter. People are going to react differently.

 

[terkans]

I don’t think real-world parallels can be drawn when comparing real life where you have a single life and games where you can die and run back to your corpse and revive... just like FPSs and real-world combat parallels can’t be drawn. When the player knows there are little consequences to death, they don’t mind doing stupid things.

In general, most people probably will behave better in real-life than in a game.
Nevertheless, there are individuals out there who go around explicitly rubbing their spittle on public handrails or aggressively coughing at groups or at the officials trying to stop them from doing so. At least one such idiot in NL was arrested for it.

So while percentage wise many fewer people will behave that badly IRL, knowing that such behavior will occur can still improve your models for the spread of a new disease.

 

[cerberusTI]

Some article corrections. The initial infection was not due to players teleporting out; there were several mechanisms in place which prevented this. Dungeons were separated from the wider world by "instancing"; once you enter that dungeon, you generally cannot interact with the wider world until exiting it and you are not permitted to exit while in combat.

Additionally, once a boss monster has become aggressive towards you ("gained aggro") by either being attacked or approached too closely, it will remain so until either you and your party or dead or until you slay it. And you are considered "in combat" for the duration, so you are not permitted to leave the dungeon "instance".

This programming worked correctly, so there was generally no way for the corrupted blood disease to spread. Except for hunters.

Hunters had as one of their skills the ability to tame "pet" animals, and then summon or dismiss them even in combat. Generally whatever their status was when dismissed, it remained so when resummoned later.

One hunter facing Hakkar with their party had the bright idea to get the pet infected, and then dismiss it. The party died to hakkar, and when they revived he then went to a major human city and summoned his pet... who was infected, and promptly started spreading the infection.

---

One other correction. Blizzard did not "reset the game". WoW is divided into dozens of servers, which run their own instances of the world. What happens on one server never affects another (barring character transfers, which are a lengthy process). Three servers were affected, so those three servers were reset, clearing the diseases and ending the epidemic. Pets were later made immune to corrupted blood so that it could not happen again.

I found out about this at the time when a friend I shared an apartment with kicked my door in, demanding that I stop working and log into wow.

She had found out about the issue, and while neither of us had a hunter, we both had rogues. We thought maybe sneaking up to him, getting the plague, then vanishing out together would let us get out of combat and pass it back and forth. It did not really work, and we died a few times trying.

When we hearthed back (the teleport they speak of) to instead try to assemble a minimum viable raid to get it that way (not that hard for us as we both had priests with a bit of gear, and knew plenty of hunters) we found that someone had already unleashed it while we were doing so.

It pretty much just murdered any low level player or npc, but as they could pass it back and forth, and higher level npcs had enough health regeneration to not die from it, any decently high level npc standing within proximity of another npc would keep it active, and spread it to anyone passing by.

Pets were the source of more than one of these. I recall the same thing happening at an earlier point where one of the bosses in the release raid instance could make a pet "the living bomb". This hurt friendly players nearby too when it went off, and if you dismissed the pet with the countdown you could resummon it in the middle of the auction house later to cause mass chaos.

 

[Tech Person]

This article doesn't explain why/how the disease was transmitted. It says that the boss cursed characters with the disease, but doesn't explain that it was transmittable. How did characters transmit it? Did they have to come within a certain distance of other players?

 

[zelmak]

/pedant its Zul'Gurub, not Zul'Burub.

 

[cerberusTI]

This article doesn't explain why/how the disease was transmitted. It says that the boss cursed characters with the disease, but doesn't explain that it was transmittable. How did characters transmit it? Did they have to come within a certain distance of other players?

Every few seconds it attempted to spread to all friendly players or NPCs nearby.

It did expire, but you could catch it again. The goal was to make you quarantine yourself in the raid and not spread it to allies.

 

[A_Very_Tired_Geek]

This always seems to come up, but I think people behave quite differently when it's a game than in reality. In these write-ups it's often talked about as though it was an accidentally spread epidemic, but my experience was definitely that it was 99% deliberate.

Sure, someone found out it worked that way by accident. The fact that hunters would dismiss their pets with the debuff only to re-summon it in town again, and again, and again wiping out all the nearby low-level players repeatedly was not an accident.

I just don't see that many parallels; the equivalent would be if large swathes of people traveled to Wuhan for the express purpose of catching the virus to bring it to capitol cities around the world, rapidly wiping out large portions of the city then doing it again a few hours later. Additionally the virus itself would have to be practically incapable of spreading accidentally; it's vastly different.

Since it was a game most of the time infections like that broke out was some wit wanting to share the love. It was little different than kiting one of the world dragons up to the gates of Orgrimmar then disappearing while the guards intercepted. An annoyance, intended to grief or cause mischief to the other side. I think people either don't know, or forget, that WoW was also a world in a state of Cold War up till open warfare broke out again after Garosh took over as Horde Warchief. Griefing the other side was part of the game. In any way you could. Hell sometimes it was your own damned side. People got bored. The only consequence was you have/had to run back and grab your corpse or spend gold to retrieve it.

Eventually the devs made the world dragons reset after a short distance and return to their patrol routes. And, they realized that diseases that spread across continents was a great backstory, but made for lousy active game mechanics. Infections are limited to the instance in which they originate. Hell, iirc sometimes you quite literally had to "share the love" to keep the whole party from a wipe in certain boss fights.

 

[gitgud]

Did they have toilet paper in WoW?

No, because of the amount of shitty people, griefers and trolls - they ran out of TP in the first year of Vanilla and never recovered.

 

[Fred Duck]

The Corrupted Blood outbreak was not intentional.

That makes sense. It would be well, a bit different if someone made a game that challenges you to strategically spread and evolve diseases. I mean, the concept isn't exactly honourable.

An epidemiologist named Eric Lofgren, then at Tufts University, just happened to be an avid WoW player and was fascinated by the real-world parallels to how the epidemic played out in the virtual world.

So, the takeaway would be if this happened in real life, we need to ban teleportation, as well as take down this Hakkar fellow ASAP. Remember how a bunch of companies rushed to create virtual presences in the hit online game Second Life? Maybe The Who should set up shop in Azeroth.

Edit:

One hunter facing Hakkar with their party had the bright idea to get the pet infected, and then dismiss it. The party died to hakkar, and when they revived he then went to a major human city and summoned his pet... who was infected, and promptly started spreading the infection.

I don't have time to rewrite the joke now. =_=

I really feel like the author overdid the scare quotes in paragraphs 2-4.

I don't think they're "scare quotes" but rather just for emphasis.

In general, most people probably will behave better in real-life than in a game.

I would hope so. If you've seen the hit show Sword Art Online, wherein players in an MMORPG were trapped inside unless they figured out a way to escape...and in-game death or attempts to disconnect led to real-world death, there were a number of players who didn't believe it was true and still went around "griefing" others.

 

[Baumi]

It's sort of like getting rid of people who predict earthquakes because you haven't had an earthquake in a while. Well, yeah, you're gonna have another one.

That sentence threw me for a loop: We cannot predict earthquakes. We can offer reasonable estimates about the seismic risk at a given location, but we can't predict when the next quake will strike nor how big it'll be, so there are not "people who predict earthquakes" one could get rid of.

I know it was just an off-the-cuff comparison, but it's a rather awkward one, especially coming from someone with scientific training. A better one would probably be to say that one shouldn't relax building codes just because of a temporary lull in local seismic activity. Eventually, the next quake will strike, so it's better to prepare for that.

 

[ripelivejam]

I doubt that hoarding and hiding will prevent hardly anyone from contracting it. The only real question is how hard will it hit you.

Fair...they've been saying all along that 70 to 80 percent of us will get it...most with mild symptoms...but 'hoarding and hiding' will slow down how fast we reach 70 to 80 percent, thus giving our medical teams time to deal with the x% that need help because it wasn't mild.

And then what? We are to continue isolating ourselves and hoarding goods in panic mode indefinitely so the more severe cases can be tended to? This is just how we are going to live now?

I have an employer that depends on people visiting it, but I guess that is ny cross to bear.

I'm also of the mind that the rampant hoarding is nonsensical and is just going to lead to more fear/panic/paranoia that won't help anything.

 

[Null_Space]

I doubt that hoarding and hiding will prevent hardly anyone from contracting it. The only real question is how hard will it hit you.

In all this talk about “flattening the curve” I haven’t seen any discussion or analysis on how it might/will change R naught.

I would think that it would have to.

If I’m presymptomatic and previously would have had , say, 300 person-hours of close contact while contagious. And now I’ll have 10.

Wouldn’t that reduce R naught by ~30X? At a statistical/population level?

What am I missing?

These statistics are (for now at least) empirical, so I would expect so. The Washington Post had an interesting article about the effect of social distancing that included some simple but illustrative simulations: https://www.washingtonpost.com/graphics ... simulator/

Interesting, but I think their simulation has basically the same problem as using WoW as a model.

It's effectively modeling a disease with infinite transmissibility. Pass someone infected in the street within 6 feet, you're guaranteed to catch it. To maintain consistent terminology, it has a R naught of ~hundreds, thousands?

From everything I understand (which isn't a professional knowledge) real life doesn't work like that.

And I think that really matters. If people are like "80% are going to catch it regardless" it seems like it has a real danger of undermining effective social distancing. People would be more likely to maintain distance if the messaging was "significantly reduce the number of people that catch it".

 

[evilgardengnome]

"We often view epidemics as these things that sort of happen to people. There's a virus and it's doing things. But really it's a virus that's spreading between people, and how people interact and behave and comply with authority figures, or don't, those are all very important things. And also that these things are very chaotic. You can't really predict 'oh yeah, everyone will quarantine. It'll be fine.' No, they won't."

Or, to put it another way: viruses don't cause pandemics; people do.

Seriously, just listen to the medical experts. Stay home if you can, limit exposure with others (for their sake, if not yours), and don't assume "it's passed" until you get the all clear.

If you can, it's almost a civic duty to work from home. At the very least so you don't risk exposure to the people who can't do so.

 

[Zool26]

Some time ago, not too much, I read another story on this outbreak, with more details. I am 99% sure it was on Ars, but a quick search with the embedded search engine returned just a very old, very short story.

 

[rosen380]

" In the end, at least three servers were affected, and Blizzard had to reboot the entire game to correct the problem."

Maybe I'm severely missing the mark in how these games work on the backend, but I'd think that most player attributes were essentially stored in a database.

UPDATE [WOW].[dbo].[player_attributes]
SET [CorruptedBlood] = 0;

?

Im not sure that a database is the best way to store lists of debuffs, which are transient and rapidly changing. Even if it were, this would ruin in-progress Hakkar encounters.

If you pay attention to what resets on maintenance in a MMO, you'll notice that most buffs and debuffs are not saved and only kept in RAM. MMO devs have it easy with the database stuff.

RAM on the server or RAM locally? If local, wouldn't it be reasonably easy to turn things on and off that only get noted in local RAM?