John Pickens

My years working with Glen were in my formative years, both professionally and personally. Before marriage. Before internetworking and systems and research and development as career choices. Before the opportunity to construct teams myself -- trying to remember the many lessons learned at Glen's feet -- trying to model Glen's mentoring style myself toward others.

My recollections therefore are mostly personal.

My two most intense memories are of Glen as coach and of Glen as visionary -- especially his vision of the ideal environment for team growth and creativity -- which he worked constantly to translate into reality.

I first met Glen in 1967 when I was a Junior in Electrical Engineering at UCSB. Before EE became EECS. That summer when hearing of an interesting project on campus, and knowing nothing of systems programming, and needing summer income, I applied for a job as a junior (an overstatement) programmer in the team led by Glen Culler. I was hired to work on the portation -- and expansion -- of the Online System (OLS) on the IBM system 360 mod 50. In the computer center. Before the BSU protest and takeover of the computer center building after which the exterior window was bricked over. Hands on mainframe time from 12 midnight to 5am in the morning -- wasn't "360" actually the cost per hour in dollars for rental of the mainframe?

Development was primitive. Punched cards. BAL assembly -- "C" was the stuff to the west and south of the campus, which licked the mussel coated rocks on which I became a surf fisherman of some repute. Two hour batch job turnaround. After teething under the tutelage of Chuck Irby (and together with my fraternal software twin -- Dale Taylor) -- and driving Ron Stoughton crazy by ferreting out his bugs which he swore were not -- I was given the responsiblity for designing and implementing COL (Card Oriented? Language). On numerous occasions Glen and I would discuss the ins and outs of this facility; how it should be mapped to an easy to use user interface, how it would interface with the batch system, and how my frequent mistakes and screwups should be corrected. I don't know if Glen remembered the day I submitted a batch job that beat the computer center's 30 minutes nuke-if-hung hog-job-partition timer by sleeping 29 minutes and 59+ seconds and waking just long enough to not be canceled and not be charged.

By 1969 or thereabouts Glen left the campus to form a startup company, Culler-Harrison (CHI Inc.) to develop the world's first commercial speech recognition system. Well, actually it became signal processing hardware/software tools. $500K investment from Harrison, I recall. I also married. I was fortunate to be invited to join Glen among the first round of employees -- my first startup. At $11K annual salary I was affordable. And 50,000 shares of founders stock -- or was it 10,000 or 100,000 -- was the clincher -- Glen's selfless wish to make us all wealthy. I lived within a few minutes biking distance of CHI. I don't know if Glen noticed, but there were days too numerous when, after lunches with my new spouse when I biked home, for some unannounced reason I did not return to work until late afternoon -- if at all. If he noticed, he never asked. After a year I and my wife moved from an apartment to our first purchased house -- $25K. Rolling in the dough and a healthy stock option. My first son was born.

My most memorable learning came while I was at CHI.

One of my first projects was to design and implement the disk I/O system for the OLS online system on the SEL Modcomp minicomputer. I thought I understood all the requirements. I designed. I implemented. I tested. I finally finished the project, and was proud. But after showing Glen my masterpiece my joy was short lived. After only a few minutes -- seemed like milliseconds -- of using the system, he called me into his office and gently challenged me on its poor performance. It was 4-5 times too slow, said he. I would not believe it! Besides, how could he know? I was the task expert. Glen's turf was math, algorithms, high level systems design, team building. Mine was in the specs, details, implementation. Reluctantly I examined the actual read timing performance with the hardware help of Ray Bjorklund. To my horror I learned that my driver was slipping the inter-sector gap, every gap, and that only one sector -- out of five -- was being read each rotation. I learned that day two lessons. First, don't consider delivery of base functionality the sole criteria of success. Second, learn to perform independent external rules-of-thumb sanity checks on delivered performance, especially in a module that is performance driven. Thanks to Glen and his attention to details.

On another occasion I was complaining that one of my system designs was becoming more and more complex and logically twisted. In a brilliant teaching moment Glen stopped me in my tracks with the following lesson -- "If a design continues to grow more and more warts and twisted appendages, it is probably an indication that you took a wrong turn some ways back. Throw out your design and start over." I did.

Glen's vision for team creativity and individual growth was one that has stuck with me to this day. While many leaders seem bent on advancing their own career, sometimes at the expense of their subordinates, Glen's goal was to grow individuals as his peers. (As if that were possible.) During my few years with Glen at CHI he shared with me his vision of an ideal working environment. A building in the shape of a wheel, with an open air center courtyard, with offices facing inward, inviting direct contact and dialogue between peer scientists. Preferably in a wooded area. Remote. Sweet. Perhaps that vision will become a reality.

Beyond these serious memories I have others more humorous.

Eventually CHI fell upon hard times and exhausted its startup funds. I returned to campus to obtain my Masters degree and PhD. My second son was born. I obtained a $21K job with Bell Northern Research and moved to the San Francisco Bay area. A daughter was born. I began to make my own way in the professional world. And raise a family. I learned to mentor others and lead teams. I made my own way and blazed my own path. And this summer our 25th wedding anniversary was marked by a surprise party thrown for us by our children.

But I always remember those few years with Glen Culler. He I consider the penultimate teacher and model mentor. I am proud and fortunate to have been associated with him in the formative years of my own personal life and professional career.


Previous
Tribute
Tribute
Contents
Next
Tribute
Back to
Glen Culler Honorary Lecture
Home Page
Back to
College of Engineering

kk October 2, 1995