I have heard of at least one story where an employee-programmer was compensated for his work and later claimed copyright-ownership over the code he was hired to

I have heard of at least one story where an employee-programmer was compensated for his work and later claimed copyright-ownership over the code he was hired to write, and was successful in getting some kind of court decision or settlement, so I don't think the mere existence of compensation alone is sufficient to prove transfer of ownership. But the MIT license part you suggested is reasonable.