But I think his point was that those people who can't solve those problems in infinite time are at such a low end of productivity that almost every decent progr

But I think his point was that those people who can't solve those problems in infinite time are at such a low end of productivity that almost every decent programmer is some coefficient c better than the one that can't do it, where c >> 10. There are tough problems in programming, but unless you are breaking ground in p-np complete type proofs, are there really any day to day problems in almost any development setting that a decent programmer can't attack?