Programming

I interned at IBM in the late '80s at the TJ Watson research facility. I have no idea if that's still around or if it's still what it used to be, but at the time it was a pretty amazing place, filled with brilliant people doing stuff that may or may not have been directly related to the corporate bottom line. Benoit Mandelbrot (the chaos theory guy) had an office there. There was an unused scanning electron microscope parked in the hallway outside of our lab because there was nowhere else to put it. I learned to use CADCAM on enormous monitors; it was a blast to design something, send it electronically to the machine shop for fabrication and have it delivered on a cart the next day (sometimes the same day). I worked on a project repurposing these miniature electric punches that had been designed for ceramic green sheets (the way they built their mainframe cores back then) and then got to experiment creating a new hole-punching technique using pressurized fluids. They let you do whatever you felt like doing even if you were just an intern. There were no corporate idiots anywhere in sight there.

As far as I can tell, that part of IBM (the actual innovation) is gone.