Computer code used to model the spread of diseases including coronavirus “must meet professional standards”, an industry lobby group has said, after criticism of some of the systems used by ministers who ordered the UK lockdown.
BCS, The Chartered Institute for IT, said that a lack of widely-accepted software development standards in scientific research, risks undermining confidence in computer modelling, including high-profile models which were used to inform Covid-19 policy.
Dr Bill Mitchell, the group’s director of policy, said better modelling could help determine which lockdown measures were likely to have the greatest public health benefits.
He warned that currently “the quality of the software implementations of scientific models appear to rely too much on the individual coding practices of the scientists who develop them”.
The British Computer Society (BCS) is a professional body which represents people working in information technology and computer science.
The group’s comments come weeks after data scientists criticised the code behind one of the most high-profile Covid-19 models in The Sunday Telegraph.
The model released by Imperial College, London, was seen as instrumental in the UK’s decision to close down the economy in March. The Imperial model prompted a U-turn by the Government, after it estimated as many as 250,000 people could die from the deadly virus if no further action was taken in the UK.
However, Imperial College’s model had initially been based on “thousands of lines of undocumented C” code which were written “13+ years ago to model flu pandemics”, one of the institute’s scientists Neil Ferguson had revealed in March.
After the code was cleaned up and released publicly, some questioned the reliability of its conclusions. Data scientists from WANdisco, a British software company, questioned whether it should have played such a large role in decision-making. They said the code was “something you wouldn’t stake your life on”.
Criticism has not just been targeted at Imperial, however. Other universities have since come under fire for how computer code is developed inside their institutions, with one expert claiming Britain suffers from a “culture of mediocrity” in the field.
Chris Greenshields, director of UK open source software firm The OpenFOAM Foundation, told The Telegraph earlier this month that it had become “all too much the norm in academia” for experts to be allowed to rely on badly-written software.
The BCS said making sure code met professional standards, and was open-source, would allow research groups to build on each other’s software, as well as ensure research findings could be reproduced and provide “reassurance” that policy decisions were based on robust evidence.
A spokesperson for the Imperial College COVID19 Response Team responded to criticism of its code by saying the Government “has never relied on a single disease model to inform decision-making”.
“As has been repeatedly stated, decision-making around lockdown was based on a consensus view of the scientific evidence, including several modelling studies by different academic groups.
“Within the Imperial research team we use several models of differing levels of complexity, all of which produce consistent results. We are working with a number of legitimate academic groups and technology companies to develop, test and further document the simulation code referred to. However, we reject the partisan reviews of a few clearly ideologically motivated commentators.”