Sue Denim (a pseudonym) – Lock Down Sceptics May 7, 2020

Imperial finally released a derivative of Ferguson’s code. I figured I’d do a review of it and send you some of the things I noticed. I don’t know your background so apologies if some of this is pitched at the wrong level.

My background. I wrote software for 30 years. I worked at Google between 2006 and 2014, where I was a senior software engineer working on Maps, Gmail and account security. I spent the last five years at a US/UK firm where I designed the company’s database product, amongst other jobs and projects. I was also an independent consultant for a couple of years. Obviously I’m giving only my own professional opinion and not speaking for my current employer.

The code. It isn’t the code Ferguson ran to produce his famous Report 9. What’s been released on GitHub is a heavily modified derivative of it, after having been upgraded for over a month by a team from Microsoft and others. This revised codebase is split into multiple files for legibility and written in C++, whereas the original program was “a single 15,000 line file that had been worked on for a decade” (this is considered extremely poor practice). A request for the original code was made 8 days ago but ignored, and it will probably take some kind of legal compulsion to make them release it.  Clearly, Imperial are too embarrassed by the state of it ever to release it of their own free will, which is unacceptable given that it was paid for by the taxpayer and belongs to them.

The model.  What it’s doing is best described as “SimCity without the graphics”. It attempts to simulate households, schools, offices, people and their movements, etc. I won’t go further into the underlying assumptions, since that’s well explored elsewhere.

Non-deterministic outputs. Due to bugs, the code can produce very different results given identical inputs. They routinely act as if this is unimportant.

This problem makes the code unusable for scientific purposes, given that a key part of the scientific method is the ability to replicate results. Without replication, the findings might not be real at all – as the field of psychology has been finding out to its cost. Even if their original code was released, it’s apparent that the same numbers as in Report 9 might not come out of it.

Non-deterministic outputs may take some explanation, as it’s not something anyone previously floated as a possibility.

The documentation says:

The model is stochastic. Multiple runs with different seeds should be undertaken to see average behaviour.

“Stochastic” is just a scientific-sounding word for “random”. That’s not a problem if the randomness is intentional pseudo-randomness, i.e. the randomness is derived from a starting “seed” which is iterated to produce the random numbers. Such randomness is often used in Monte Carlo techniques. It’s safe because the seed can be recorded and the same (pseudo-)random numbers produced from it in future. Any kid who’s played Minecraft is familiar with pseudo-randomness because Minecraft gives you the seeds it uses to generate the random worlds, so by sharing seeds you can share worlds.

Clearly, the documentation wants us to think that, given a starting seed, the model will always produce the same results.

Investigation reveals the truth: the code produces critically different results, even for identical starting seeds and parameters.

I’ll illustrate with a few bugs. In issue 116 a UK “red team” at Edinburgh University reports that they tried to use a mode that stores data tables in a more efficient format for faster loading, and discovered – to their surprise – that the resulting predictions varied by around 80,000 deaths after 80 days:

Continues …