Columbia University Computing History   

IBM's John McPherson, 1908-1999

Written by Herb Grosch for this site, 2004.06.01.

John McPherson was the monitor of all technical activities in "the IBM", as The Old Man liked to call it. He helped Ben Wood set up the Statistical Bureau at Columbia, and got him his slightly modified tabulator; he helped Wallace Eckert set up the T.J. Watson Astronomical Computing Bureau in the Pupin attic, and got young Red Dunwell to build him the first sequencing switch. He got Lake and Durfee and Piatt to build and rebuild the five Aberdeens [Pluggable Sequence Relay Calculators]; he got them to build the crypto gear now called MAGIC. He got Lake and Hamilton to build the ASCC for Aiken, and I believe contributed to the design. John was the first IBM executive I met after Eckert hired me in May 1945 – the second scientist in the company, the first employee of Wallace's Watson Scientific Computing Laboratory at Columbia (and the first IBMer in the U.S. with hair on his face!). He was helping Eckert set up the lab, choose and order the gear (making it the third most powerful computing center in the world), hire guys like me and Rex Seeber and Hilleth Thomas, and fresh young electronic talent from the [MIT] Rad[iation] Lab. He helped Eckert and Seeber and Hamilton design the SSEC, and helped Hamilton build it.

He helped with the older machines, and with the 603 and the 604 as electronics burst out. I was in his office the day he laid out the CPC for Northrop. He sent Pete Luhn and his one-off relay computers to the Watson Lab, and encouraged John Lentz to do the 1620 prototype there, and By Havens to design and build the first supercomputer [NORC] around the corner on Broadway and on 115th Street.

He was the door through which almost all the IBM inventors and engineers, and us scientific types, reached TJ (Bryce and Eckert had direct access). I call him a monitor; he didn't give orders, but he passed requests from la bas to Watson, and Watson's orders to nous ouvriers. He was unquestionably the central figure in the IBM technical escalade from 1930 to the early Fifties. Eckert had the scientific vision, Hurd had the marketing vision, but John put it all together and channeled the overwhelming power of The Old Man to the whole computing community.

My own relationship with him was unsymmetrical. From my first contact for over four decades I admired him, although I suffered occasionally from his stiffness; he thought me unruly – a tummeler, as they say in the Catskills. His access to Watson, and the trust The Old Man put in him, turned Young Tom hateful, and destroyed his later years. I'm happy that Emerson Pugh does so well by him in Building IBM. I'm very unhappy that he never made the NAE, just as Eckert never made the NAS, and regard it as a proof that those academies are far from impeccable. Palmer and Haddad and that ilk owed much of their recognition to John, as Brouwer and Clemence and Herget owed much to Wallace. I'm glad, proud, to have known him.