My favorite Bob Widlar story is a huge misdirection prank that he pulled in the literature. In the late 1960s, many companies were designing IC voltage regulators, and there was a perceived demand for a monolithic device. Widlar convinced people that it couldn't be done, and then he did it.
In June 1969, he wrote a paper in EEE , where he laid out the case against monolithic voltage regulators. Using the (eight-pin) LM100 as an example, he convincingly argued three points, focusing on the detrimental effects of an integrated power transistor. In short, monolithic voltage regulators will never work because the die temperature will kill you, and if it didn't, then the temperature gradient would kill the reference, and even if it didn't, there's no way to package it (there weren't any standard high-power packages with enough pins). See page 93 of this article, under the heading "Power Limitations".
Then, in February 1970, he presented a paper at ISSCC , where he published the circuit of the LM109, the first monolithic voltage regulator. The solution, of course, was his bandgap voltage reference, which allowed the reference to coexist with the wide temperature swings caused by the integrated power transistor. Better yet, the whole thing fits into a three-terminal transistor package.
In February 1971, he published a paper in JSSC , which expands on the ISSCC paper. Given the various publication delays of these three publications, I imagine that these articles were submitted within days of each other.
 Robert J. Widlar, "Designing with positive voltage regulators," EEE, vol. 17, pp. 90–97, Jun. 1969.
 Robert J. Widlar, "New developments in IC voltage regulators," in IEEE International Solid-State Circuits Conference, Digest of Technical Papers, Feb. 1970, pp. 158–159.
 Robert J. Widlar, "New developments in IC voltage regulators," IEEE Journal of Solid-State Circuits, vol. 6, no. 1, pp. 2–7, Feb. 1971.
My second-favorite Widlar story is the one about the sheep.