That's a classical/traditional view of science. There are many people who now recognise that science itself is a socially constructed process that is always subjective and uncertain. We can't eliminate those things. We can try and account for them.

I think that's too much of a nihilistic/ad hominem push in the other direction but it has some good points. Typically we can try and control for those things with statistical methods. There are various hacks around it but they're usually exposed by scientists rather than relativists.