In college I studied electrical engineering and focused on microprocessor programming. I learned Fortran, Pascal, COBOL and C programming languages as well as assembly code for MC6800 and MC68000 microprocessors. I interned as a Fortran programmer at a power industry consulting firm. When I graduated I had offers from several great companies. I graduated when EE demand was probably at the highest it has ever been. I had two top choices. One was Trane, an HVAC company, that wanted to start using microprocessors to control commercial HVAC equipment. They were ready to hire me straight out of college as an expert to help start their very first solid state controls department. It would have been a fantastic opportunity to be considered an expert right out of college and to introduce a technology into an industry.
Instead, I chose an offer to become an IC designer with Delco Electronics, an automotive company. Integrated circuit design was easy to identify as a growth career and I really like the idea of being involved with cars. I had no idea how to design an IC but they sent me to a one week crash course at the University of Waterloo in Canada. The course was an in depth study of a new design technology called VLSI and was based on a relatively new book authored by Mead and Conway. I went with one other NCG and when we came back we were two of the experts in this new methodology.
After a few design cycles using the VLSI methodology I discovered something called espresso which replaced Karnaugh maps with gate level optimization of logic expressions. Karnaugh map experts could not initially see the advantage but as soon as espresso started using more complex and more efficient logic gate structures which resulted in lower cost ICs, logic synthesis experts were sought. Soon after discovering the logic synthesis process, we discovered that a company called Synopsys was selling software that no only optimized logic but entire state machines and was using the Chapter 8 code from the Verilog logic simulator which we had been using for verification. Most people today don’t even realize that Verilog was originally a multi-state gate level simulator for schematically entered gate level designs. And, that the part of the language that has become well known for its ability to describe hardware was originally Chapter 8 of about 25 chapters from the language reference manual. Chapter 8 was original intended for describing the tests applied to the logic, something known today as a testbench. Synopsys, wisely chose a subset of the Chapter 8 code to be their hardware description language, as many testbench designers were already familiar with the syntax. I had designed some of the testbenches for the ICs I was involved with and was very familiar with Chapter 8 code, so voila, I was an expert.
Logic synthesis begat standard cell based design which required automatic placement and then automatic routing(APAR). That’s right they did not come at the same time initially and I actually manually routed designs that had only been automatically placed in rows with routing channels in between. One design like that was enough. Eventually we had a full front to back synthesis and APAR process with clock insertion, static timing analysis, a code coverage process and an Electronic Design Automation (EDA) department to maintain the constant new releases of the all the software from EDA companies. I was an EDA pioneer, and pioneers that survive to the destination are called experts.
So, after participating with the development of an EDA process at Delco Electronics and at the same time helping with the design and implementation of automotive electronic controls, I accepted an offer to take a position at Ford’s electronic division. I was sought by them as both an EDA and automotive expert. They had not yet adopted a complete EDA process and were embarking on a new powertrain microcontroller. They recognized that new IC architectures and the EDA process could leverage each other to create more cost effective products. This synergy between the design and the EDA process got the moniker ASIC development. A bit of a misnomer but then almost all of technology vernacular is misnomer. Again I was one of the ASIC pioneers and had several successful victories at Ford with gains in cost effectiveness of both product and product development. I was recognized as an expert in ASIC.
It was at this point that I started to understand the demands of being an expert. As an expert you are expected to not just contribute to the solution but must accurately identify the problem, then define the solution and quantify the results. At the same time an expert must continuously monitor and adopt the leading technological advances of other experts and effectively communicate his own. An expert’s resources are his knowledge, experience and access to outside resources. Experts begin to loose mentors and gain colleagues and depend heavily on an extensive library of information. When I first started my mentors were experts and they could almost be measured by their bookshelves of published texts and binders of notes from past efforts. Today that library is more commonly the internet. The internet has given all of us the bookshelf of an expert. I first started mining the internet when Ford Microelectronics bought a modem and allowed a few of us limited access. I discovered Archie and immediately started increasing my reputation as an expert. Not always because I had the answer but because I could get an answer and get it quickly. I also could sort through candidate answers to either find the best one or a selection of answers giving my peers and superiors options. Answers and options are the currency of an expert. It is the experience and knowledge of an expert that allows him to leverage the internet library quickly, efficiently and above all accurately. That is why the internet alone cannot really make everyone an expert.
Every position I have held since Ford was offered to me as an expert and each position gave me new experiences, new skills, new knowledge that allowed me to claim the needed expertise for the next assignment. As a consultant I haven’t always had immediate expertise in areas that my clients were seeking but now I am an expert at being an expert so I have demonstrated that I know where to look, where not to look in order to accurately identify problems and propose solutions with options. Even though I have not seen the exact problem or found the precise solution before, I can apply my knowledge and experience of finding other problems and solutions and the resources of colleagues and the internet to new situations more effectively than those without this experience. Technology moves fast and almost no one has direct experience any more. Whatever needs to be done has never been done before and once it has been done it does not need to be done again because it is quickly shared and absorbed by everyone that needs it and soon after it is obsolete. So an expert is not one that has done exactly what IS NOW needed before but has done what WAS THEN needed before. And therefore can confidently be expected to quickly identify, solve and communicate the problem at hand.