Posted on 05/20/2020 9:56:15 AM PDT by grundle
Imperial College’s modelling... could go down in history as the most devastating software mistake of all time, in terms of economic costs and lives lost.
... those of us with a professional and personal interest in software development have studied the code on which policymakers based their fateful decision to mothball our multi-trillion pound economy and plunge millions of people into poverty and hardship. And we were profoundly disturbed at what we discovered. The model appears to be totally unreliable and you wouldn’t stake your life on it.
Imperial’s model appears to be based on a programming language called Fortran, which was old news 20 years ago and, guess what, was the code used for Mariner 1. This outdated language contains inherent problems with its grammar and the way it assigns values, which can give way to multiple design flaws and numerical inaccuracies. One file alone in the Imperial model contained 15,000 lines of code.
Try unravelling that tangled, buggy mess, which looks more like a bowl of angel hair pasta than a finely tuned piece of programming. Industry best practice would have 500 separate files instead. In our commercial reality, we would fire anyone for developing code like this and any business that relied on it to produce software for sale would likely go bust.
The approach ignores widely accepted computer science principles known as “separation of concerns”, which date back to the early 70s and are essential to the design and architecture of successful software systems. The principles guard against what developers call CACE: Changing Anything Changes Everything.
Without this separation, it is impossible to carry out rigorous testing of individual parts to ensure full working order of the whole.
(Excerpt) Read more at telegraph.co.uk ...
Fortran is still used for all kinds of simulations and modeling. Actively supported.
The “Hockey Stick” by Mann will always be the benchmark for bad models.
It’s just a journalist trying to use scientific sounding words. He doesn’t understand them.
Fauxi and Birx used fraudulent Imperial College model to urge shut down
I did the same with “C” and “Java” and got the same infinite hits.
The problem is with any computer language.
Well, we can’t have any program use 15,000 lines of code.
What a mess the world would be!
Why didn’t everyone concerned think to do another model? I would think several models should have been tested against this one.
The use of “FORTRAN” or “15,000” lines of code as the “source” of the problem is completely senseless.
It’s got to be something like what you state.
The problem is with the assumptions used and the direction the fashioned model goes with it. This guy is trying to shift the blame away from the biased humans behind it.
Agree!
ferguson lost the handle on his overall design, or his algorithm sucked to begin with, or he never understood the problem, or maybe he's just a crappy programmer.
but that the dumbest damn thing I've read in quite some time.
Bet it wasn’t a mistake...
I am supremely confident that ferguson would have f*cked it up even worse in c++
But if it saved even one life, wasn’t it worth it?
Just think of the lives saved if we would just lower all highway speed limits to 45mph, and all city limits to 15mph.
Boy this guy is a whinger !
“Imperial’s model appears to be based on a programming language called Fortran...”
There is nothing wrong with Fortran as an engineering language.
Geez !
I suspect an agenda was at play. The bigger modeling numbers, the more fear, the longer lockdowns. Just shut down long enough to get the dems to November.
We did a lot of planning for long-range to help with purchases and exercise planning. That used a lot of the long-range models - and those would sometimes be good - and sometimes crap - and it depended on stuff like MJO - AMO - all sorts of science and seasonalities.
They are only as good as the data you feed into them
For example, we were feeding data in from China and South Korea, the data from China was wrong, so our models were off.
All models are best guesses, because that is what a model is.
Could you build a model, based on the past 10 years of lottery number drawings predict next week winning lottery numbers?
If you could, you and all your scientific pals would be crazy rich right now.
My career was built on FORTRAN (and Assembly)
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.