1 Terry: You're saying we've accidentally released millions of copies of Cthulhu across all of time and space?
2 Simon: And alternate dimensions!
3 Terry: Well that can't be good.
4 Bonnie: It's the best thing that's happened since my Ph.D. study on global warming concluded civilisation will collapse within 25 years!

Nice understatement, Terry.

Bonnie is loosely based - very loosely^[1] - on one of the real Dr Simon Beard's real colleagues.

[1] As in, I've never met her and have no idea how accurate this portrayal is.^[2]

[2] Well, that's not quite true. I'm pretty darn sure it's close to 100% inaccurate.

The real Dr Simon Beard writes:

I would just like to point out that your two footnoes to today's annotation provide a very useful example of how much lower our level of uncertainty often is than we first assume. Even very simple reflections on a topic can reduce an 'I have no idea' to something pretty useful in informational terms. Often, as in this case, this is because issues about complexity mean that just about any useful statement we think we might be able to utter is almost certainly false. However the flip side of this is that that complexity also allows us to use general knowledge about only tenuously connected fields to establish pretty sound boundaries for what is and is not possibly true.

For instance, take Bonnie's PhD topic: How soon civilization will collapse. Obviously we know absolutely nothing about this and any attempt to understand how soon this will take place based on a direct assessment of the evidence base will be pretty useless, there just is no evidence base that we can use to try and work out how such an unprecidented event could occure. However, just by thinking about it was can pretty easily rule out some possibilities. For instance we can be certain that civilization will collapse some time (e.g. the heat death of the universe) but that it won't happen tomorrow. We can also use things that we already know to rule out many possible statements about the probability of civilizational collapse, e.g. That at present there are lots of things that could cause this to happen (from nuclear war to meteor strikes and global pandemics to a solar storm that destroyed our global electricity infrastructure). However on the other hand the probability of any of these things cannot be very high, because all of them have had the possibility of destroying civilization for at least 50 years or more and have not done so. Alas none of this information allows us to narrow the annual probability of civilization collapse all that well, but they suggest that something between around 0.0001% and 5% almost certainly bound the real value. Of course we know that these probabilities go up and down over time, and climate change is a big issue, but those are extreme bounds and I see no evidence to suggest that they will be exceeded in the next 25 years. All of which means that Bonnie's certainty of civilizational collapse within 25 years is also almost certainly 100% wrong (though there clearly is a risk that this will happen, and it may well be high enough that you are more likely to be killed in a civilizational collapse over the next 25 years than in a car crash).

Hence, not only do we learn a valuable lesson in today's comic about the limits of uncertainty. We also learn that either Bonnie never got her PhD (oh dear) or she has a special talent for bribing her examiners.