long live Dr. William Fleischman

Unfortunately, this result is far too often misunderstood on what are its consequences.

Just because there exist no program that can take ANY program and check some properties on them, does not mean that there exist no program that can take SOME programs (with well-defined characteristics) and check some properties on them.

Which is good news, because now that I’ve said that, we can forget about this undecidability theorem which indeed means little in the real world, and concentrate on automatic proofs on USEFUL programs, which IMHO is a field with far too few studies, because every student is deterred to enter it when he is first taught about how this is not possible in the general case.

The general case does not matter AT ALL, because computers just don’t run randomly generated programs.

So now everybody please go back to work and write some cool programs that can resonate on useful ones.

Thanks for listening :p

But no matter how wise and sneaky they’ll be
When they construct their impervious P
There’s always a hack with nothing to do
Who goes on to devise an updated Q!