Shockley Semiconductor Laboratory

~ the eulogy ~

Beckman Instruments gave up on the lab and sold it to Clevite in 1960, and Clevite gave up in turn and sold it on to ITT in 1968. ITT could not find a buyer of its own, so it closed the place. That is the ending of the first company in Silicon Valley to work on silicon, and the reason it reads as a formality is that the thing had already been gutted eleven years earlier by eight men walking out of the building.

William Shockley set the lab up in Mountain View in 1956, funded by Arnold Beckman, in the same year he shared the Nobel Prize in Physics for the transistor. He hired physicists directly out of Bell Labs and the universities, and because the man had the biggest name in the field he got the people he wanted. Gordon Moore and Robert Noyce both took jobs there. It was the first high technology company in the area to work on silicon devices rather than germanium ones.

The lab shipped almost nothing. It is remembered as a birthplace rather than as a manufacturer, and the companies it seeded were all built by people leaving it rather than by anything it sold.

~ the dossier ~

~ the four layer diode ~

Shockley's bet was a four-layer device, one more layer than a transistor, which would latch into the on or the off state and stay there without any further input. A switching network that needed three transistors could be done with one of these. The idea was sound, the thyristor descends from it, and the problem was that building the things reliably in 1957 sat beyond what a new company could manage.

The staff said so. They wanted a dependable three-layer product out of the door first and the hard project second. Shockley became convinced the diode would matter as much as the transistor had, and then he kept the whole project secret from his own employees. The paranoia went further than the secrecy. After a secretary cut her finger he decided it was a plot to injure him and ordered lie detector tests on everyone in the company.

~ the walkout ~

The youngest members of the staff went over Shockley's head to Beckman and asked for him to be replaced. Beckman seemed to agree at first and then made a run of decisions that backed Shockley instead. Eight of them broke ranks: Julius Blank, Victor Grinich, Jean Hoerni, Eugene Kleiner, Jay Last, Gordon Moore, Robert Noyce and Sheldon Roberts. In 1957 they found backing from Fairchild Camera and Instrument on the East Coast and started Fairchild Semiconductor to make silicon transistors.

Shockley called them the traitorous eight and said they would never be successful. Over the next twenty years sixty five companies were founded by first and second generation teams tracing back to his lab, and the eight themselves went on to seed the rest. A 2014 recount of Don Hoefler's 1971 family tree found ninety two public companies among the descendants, worth more than two trillion dollars between them.

~ what the address became ~

After the lab closed, the building at 391 San Antonio Road was turned into a retail store and spent part of its life as a produce market. It was still standing in 2006 and still selling fruit. Demolition came later, the site was redeveloped, and by 2017 there was new signage calling the spot the Real Birthplace of Silicon Valley. Facebook's building 391 occupies it now. Two IEEE milestone plaques mark the address, one for the valley itself and one for the work Gordon Moore did inside that building before he left it.

Every machine I have ever written code on sits at the end of a silicon lineage that runs through Fairchild and back into this building, which shipped almost nothing and is remembered for the people who left it. The place sold produce for longer than it made semiconductors.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »