Hacker News new | ask | show | jobs
by hn_asker 2004 days ago
I find it fascinating how we submit batteries of tests to code as software engineers. Yet when it comes to our own health as a species, the rigor suddenly falls.
3 comments

I regularly test programs by feeding them garbage to see if they'll die but I don't think that would pass by the ethics board in medicine
It'd be fine back when Jenner created the first vaccine over 185 years ago. There were no ethics boards back then. A lot of people died from snake oil treatments too.
It's not lack of rigor, it's ethics. We don't want to intentionally expose people to something that might kill them. Instead, we perform a trial on a large group and allow them to naturally avoid or be exposed to the virus as they will. Then, with full mathematical and statistical rigor (the best kind of rigor) we determine whether the vaccine has a demonstrated benefit.
We test code. We don't do destructive code tests in production.

Normal vaccine trials is like normal software testing. Challenge testing is like running dangerous tests in production. Its faster at detecting a problem than normal tests, but ultimately not needed and can blow up badly. The argument for doing it during covid is that time was of the essence and its faster than normal testing procedures. Instead we opted for the safer slower normal testing procedure.