Wednesday, May 6, 2020

Transparency?

In a recent letter to Science the authors request:

Given the necessity for rapid response to the coronavirus pandemic, we need many eyes to review and collectively vet model assumptions, parameterizations, and algorithms to ensure the most accurate modeling possible. Transparency engenders public trust and is the best defense against misunderstanding, misuse, and deliberate misinformation about models and their results. We need to engage as many experts as possible for improving the ability of models to represent epidemiological, social, and economic dynamics so that we can best respond to the crisis and plan effectively to mitigate its wider impacts. We strongly urge all scientists modeling the coronavirus disease 2019 (COVID-19) pandemic and its consequences for health and society to rapidly and openly publish their code (along with specifying the type of data required, model parameterizations, and any available documentation) so that it is accessible to all scientists around the world. We offer sincere thanks to the many teams that are already sharing their models openly. Proprietary black boxes and code withheld for competitive motivations have no place in the global crisis we face today. As soon as possible, please place your code in a trusted digital repository so that it is findable, accessible, interoperable, and reusable.

 We have been asking for this for months now. Like the stuff that character in London put out. I gather he is gone. Now the key point is the data. As I have been noting the data is recklessly unreliable. Thus model or no model there is no way to validate the results. New Jersey is a prime example and one wonders if one can trust anything from New York. 

There also is a fundamental question that we seem to have no answer. Namely; how is the disease transmitted. I can see the argument regarding person to person by exhaling virion laden liquid drops. But frankly after hundreds of papers there is no real answer. Isolation should have wiped out the disease. Face masks are nice but in every scene we see people without gloves. I have argued that the hands are the primary path of transmission but that is based solely on common cold vectors. 

Having just said the above, one then must question any model lacking a phenomenological structure. Regression models are useless since the rely on corrupt data. Regrettably these academics may be causing more long term deaths than the disease. Why? Just think of all the cancer patients not getting care. Take melanoma for one. A highly aggressive skin cancer, of increasing prevalence, yet controllable is caught early. But now one cannot go to a dermatologist. The list of cancers go on and on. Cancer centers are closed, until the Govs decide they had culled the herd?

Science demands transparency and the ability for others to analyze and critique the results. A trustworthy scientist must reveal all of their data and the means and methods to reach a conclusion. Knowledge is attained in this often combative process. Unlike the London Prof who seems to be playing Three Card Month with the world, as well as apparently all others this is not science, it is cultism.