I'd instead say some people are just good at making intuitive leaps and filling in the details later. Sometimes it works; often it doesn't. Sometimes we end up with alternate forms of mathematics with different sets of axioms, without infinities of any cardinality, or assuming or not assuming the continuum hypothesis.
The Collatz conjecture or 3x + 1 problem is unsolved and may remain so forever. One counterexample would indeed disprove it, but who's got time to wait around for that? Even though conceptually simple, it was described by Jeffrey Lagarias in 2010 as "...an extraordinarily difficult problem, completely out of reach of present day mathematics." My suspicion is that it is undecidable, although I don't believe it has been shown to be equivalent to the halting problem, for example. John Conway proved in 1972 that a natural generalization of the problem is algorithmically undecidable. More recent work seems largely to be based on establishing probabilistic constraints on stopping time.
The ultimate challenge. The 3x+1 problem.
Document Zbl 1253.11003 - zbMATH Open
The Collatz conjecture or 3x + 1 problem is unsolved and may remain so forever. One counterexample would indeed disprove it, but who's got time to wait around for that? Even though conceptually simple, it was described by Jeffrey Lagarias in 2010 as "...an extraordinarily difficult problem, completely out of reach of present day mathematics." My suspicion is that it is undecidable, although I don't believe it has been shown to be equivalent to the halting problem, for example. John Conway proved in 1972 that a natural generalization of the problem is algorithmically undecidable. More recent work seems largely to be based on establishing probabilistic constraints on stopping time.
The ultimate challenge. The 3x+1 problem.
Document Zbl 1253.11003 - zbMATH Open
Last edited: