Monday, February 14, 2011

Computer Systems - BruceF

This note covers my career in computer systems development.  There is another note for my career in teaching and financial management.  See: Finance  - BruceF

During my senior year at college, we took a field trip to JL Hudson Co. in Detroit. On the 14th floor there were four wings in a X formation. We looked east and west, empty. We looked north, there were woman sitting at desks - 4 across and 6 deep - typing payables invoices. We then looked south. A Big Blue IBM computer with printer was spitting out page after page of recievables and payroll invoices. The Hudson representative said - "If you had visited 6 months ago, each of these wings would have looked just like the north wing with 24 desks and people. If you come back in 3 months, the north wing will also be empty and only the IBM computer will be here". This was my first introduction to computers and their impact on business automation.

If you are curious about what it is like to work in computer systems development and / product management, read on.

After teaching, I joined a bank in souther California and worked in financial planning and budgeting.

As a user of financial (MIS) data, I was asked to participate in a 24 person meeting to review and recommend improvements to a state of the art commerical credit system the bank built from scratch. A new manager, Joel, was responsible for credit policy and he heard that this $5mm credit system (now his responsiblity) had problems. He assembled 24 people from systems, credit, finance and loan departments, and said "we've got a problem and I want you people to fix it", and left the room. We chewed on the problem all morning. After lunch I voluntered to organize our problems list into four areas and said I would lead the MIS group. A few days later, I received a call from Joel. The loan deparment group was lost, would I sit in and help them organize? A few days later, Joel called again. The credit group was stuck. It had two members, both of whom expected to be "the next president of the bank", and could not agree on anything. Would I please sit in and help? Within 2 weeks I was running 3 of the 4 task forces working on the credit system problems. A month later Joel asked me to join the credit department (THE department of most banks) and I made my career change from finance to systems development. Did knowing how to program help? Sure, but the real key to sucess was 1) understanding the big picture of how systems are designed and built and 2) knowing how to work with people who know their business but do not understand how computer systems work or what is possible (financially) and what's not.

It took three years to rewrite the credit system. Upon completion, Joel said - "if you want to be the next president of the bank, you have to become a loan officer in this bank". Ok, off to credit class to become a loan officer. While in class I got a call. The credit system just began double billing loan custormers. Help! New credit system manager was lost. She hired two "programming consultants" to document the credit system I had just designed. I sat for 12 hours listening to these guys read me psyco-code, trying to find where the problem was. Alter I heard the same code read to be for the 3rd time, I said - stop! This is going nowhere. I want to speak with the lead programer. "What have you been working on for the past 2-3 weeks". I listen and then said, "somewhere there has been a code change, that has triggered the repeat of the billing cycle. Go look and tell me what you find." A day later, he reported he thinks he has found a switch that triggers the second billing cycle. New system manager says we'll have to reprogram and run the test version of the system to see if that's the problem. I said - "how long with that take?" Two weeks. I said NO; we can't have bad or duplicate bills going out to our custormers again. Make the change now, to the production system, and I will be there when the loan bills are produced. If they are bad, we don't send them out. If good, they are sent. The bills were good. This was the first of many risk-taking experiences where I learned that the way you manage /mitigate risk is not by avoiding it, but by knowing your subject / business and understanding how to respond when things don't go as you planned or hoped.

I became a commercial loan officer and accumulated some really interesting (bad and good) stories about how people behave. It sure is not what you read about in economic text books. The role of personality (the person's views of self and situation) often times makes the economics of the venture secondary. I also learned I did not like being a loan officer. It's a strange job. You can't make loans to the people who really need them and the credit worthy ones can borrow from you or a competitor - so you march to the customer's tune. Funny business, but not for me.

I left LA in 1981 to work for a larger SF bank in loan systems development. Their systems were antiquated and I could leverage my LA credit system development expericence.  Two years into my new system development, the president of the bank changed. The European Division head became president. He had sponsored an internation banking system while in Europe. He asked the systems group - "why can't you use this european designed system in the US rather than build your own?" Being both financially and systems literate, I did some analysis: build my US focused loan system - cost $7mm, modify the european system for use in the US - cost $40mm. I thought this difference big enough to be significant. I flew to LA to discuss this with the North American division head; - thanks, but no thanks. He had just lost the race to be president of the bank to the man from Europe division. You don't buck city hall. I learned this SF bank was so big that $30mm didn't matter when it came to the politics of the top positions. (Five years later, one of my former managers asked me what loan package I was going to use; the bank acquired it because of the weakness in the international banking system.)

I was reassigned to work on the international systems in product managment. My system's staff was sent to Europe to learn the european system. I spent four (largely unproductive) years working on the CIS (customer) and Accounting parts of the international banking system. Got to visit London frequently - the only good part of the experience. I learned the IBM way of developing large scale computer systems may work (I have my doubts), but it is definately time consuming and expensive. After four years I was asked to report to a new manager who was a jacka... I said no thanks and transferred to the Money Management systems division.

In MM I began working with advanced personal computer systems, used by the bond and securities traders. I built a money trading and reporting system using very sophicated personal computer software (not Microsoft). I was amazed at how much a small (4 person) team of programmers could do. This ran circles around the IBM mainframe computer systems being used in the international banking and loan systems. I was hooked on personal computers. Ever since my teaching days when I use handheld Texas Instruments programmable calculators I had been looking for a machine (computer) that I could use to build a complete system and modify it quickly when it needs to change or dosen't yield the results you need. PCs are in.

I was too successful. The users of the MM system wanted a way for the positions recorded in the tradning system to interface with the general ledger thereby eliminating manual calculating and posting GL entries. When asked, I said this was doable. They asked my boss for this enhancement and he said NO!  History - he has been hired after the previous management of the MM division bought a computer system and loss control of its accounting. This was embarassing, cost the bank "big bucks" in failed settlement transactions, and got the previous managers fired. My boss was not willing to even think about being responsible for GL accounting in the PC system we built.

I said "nuts", this is a piece of cake, the users want it, and I did it anyhow. It worked fine, but I had just signed by walking papers. The bank was going though a very difficult financial period (one reason - the excessively expensive international banking system being adapted for domestic use). People were being laid off. I saw my department's new organization chart which assigned me to "strategic planning", a function that previously did not exist. From my budget experience in LA I knew what than meant. Time to look for another position.

I contacted the Personal Computing systems division. Two weeks later I was "canned" by MM and hired the following week by the PC division. Fortunately, my PC development work impressed the people I worked with in PC product management; the rehire was done almost before my terminaltion paperwork was dry.

To summarize the rest of my business career: I spent time in PC Product Management where I evaluated and introduced new PC hardware and software. In addition to being a dynamic time, it had real financial benefits. I was well informed of the "coming" products in PC hardware and software. This enabled me to place stock market bets on companies, some of which did very well (prior to the bust). When the bank introduced Novell networking, I moved to network systems. I planned, designed and managed the installation of networked PC sysems throughout California. In the last few years, I consulted and deployed e-mail systems. Then the bank "merged" with an east coast bank to go nation-wide. The east coast bank's approach to systems was to seek advise from IBM and blindly follow it. Today we call it outsourcing. This practice may be cost effective (not proven), but guarantees you lose the knowledge and experience that comes from hands-on involvement with sytem development and management.

As I watched system divisions shrink, I decided it was just a matter of time before I would have to either move to North Carolina (no thanks!) or leave. I chose to fire myself this time. I transfered back to PC product managment, an area that I knew would be eliminated sooner or later. When asked to give up managing server evaluation and switch to portable products, I said I would cover both spaces but would not give up servers. A month later I was terminated. I took early retirement in 2000 and expected to work at PC consulting for a few years. The systems community contracted significantly in the Bay area during 2000 - 2002 and I found more enjoyable things to do that commuting and providing 7x24 systems support services. Free at last to be my own boss!

ps:  I have had 23 bosses during my career: 4 were good, 3 stunk and the rest could be replaced one with another and not miss a step. So, if you enter the business world - find and follow a good boss; otherwise move on. You can have all the skills in the world, but if you don't get along with the "boss", life is not going to be fun and forget advancing.  In the academic world - you have lots of freedom, maybe too much. Just be sure your personal objectives and values align with the school or be frusterated!