User blog comment:MilkyWay90/I'm ready for the easiest OCF!/@comment-32697988-20180821002553/@comment-30754445-20180822220935

Exactly, Fefjo.

I think people here (from both groups) are putting far too much emphasis on formal proofs. What we're basically looking for, is any kind of actual justification that the claims are true.

An actual formal proof would be even better, of-course, but that's not a hard requirement.

So, how do we go about justifying our claims? There are no rules here. No special hoops you're required to jump through. No tedious bereaucratic process you must follow. Just justify your claims with some kind of logic, and make sure that the logic is convincing even to people who don't have access to your thought-process.

It isn't any different, really, than justifying your opinions on any other topic.

@Ecl1psed276

"For example, what justification would one have to give to convince you that the limit of pair sequence BM2 is psi(W_w)? A formal proof? A detailed analysis?"

Whatever works. It's up to you.

There's no ironclad rule that states whether "a detailed analysis" will be convincing or not. What matters isn't the format of your argument, but how good it is in covering all the bases. The problem with most of the analyses seen here, is that they don't give the reader any reason to believe they are true. They're just a long list of expressions and a corresponding values, without even a hint at how those values were calculated or even conjectured.

A formal proof, of-course, will always do the job. The downside is that formal proofs are usually much harder to come by. Another downside is that even a tiny technical error in a formal proof will render your entire argument useless.

You can view proofs as really delicate tools that require a very steady hand to work properly. They are really cool and really useful... but if all you want to do is hammer a nail into a piece of wood, a simple hammer will do just fine.