How it worked is that whenever a post was made, an infection had a 50% chance of passing from the new post to the previous post, or from the previous post to the new post. In other words, it was passed between adjacent posts. (But the random number generation was deterministic, so you couldn't "force it" by repeatedly deleting&reposting something.)
Asymptomatic infection had been happening for about 12 hours before symptoms started appearing. There was only one "patient zero", Sanitough, who I chose by looking at the
latest posts at an arbitrary point in time and picking the most recent one which mentioned "virus". It was basically random, and definitely not meant as any statement about Sanitough.
On Apr 1, symptoms started appearing, though the spread of the virus continued exactly as before. After 30 minutes, your name got corrupted; after 3 hours, your posts got increasingly corrupted depending on the amount of time since you were infected. The roughly 150 people infected before Mar 31 @ 23:30 had their symptoms shifted forward in time, so they ended up as "asymptomatic carriers" for up to ~12 hours.
Here's data on the virus's spread. Note that I designed this "forum virus" to be easy-to-implement and interesting; my goal was
not to create any sort of realistic coronavirus model. I suppose it could perhaps serve as an illustrative example of some general concepts, though, such as exponential spread.
Happy April Fools, and stay safe & sane in these unusual times!