Hacker News new | ask | show | jobs
by TheCoelacanth 757 days ago
Occam's Razor isn't proof, though. It's a rule of thumb for how to create the most useful hypotheses.

A more complex hypothesis isn't useful until you come up with a method of testing that will distinguish it from the simpler hypothesis. That doesn't mean you have disproved the more complex hypothesis; just that you shouldn't use that hypothesis until you actually have a need for it.

2 comments

I have already said that there is nothing to "prove" (i.e. no proof) but that it is a necessary corollary of applying the heuristic of Occam's Razor(https://en.wikipedia.org/wiki/Occam%27s_razor).

> A more complex hypothesis isn't useful until you come up with a method of testing that will distinguish it from the simpler hypothesis.

No, the condition is even stronger; a more complex hypothesis should not even be considered until the simpler hypothesis fails for some data/evidence.

It is one from a set of heuristics called philosophical razors (https://en.wikipedia.org/wiki/Philosophical_razor) and uses Abductive reasoning (https://en.wikipedia.org/wiki/Abductive_reasoning).

Yes, this section of Wikipedia is spot on in my opinion:

https://en.wikipedia.org/wiki/Occam's_razor#Controversial_as...