Graham Seibert Autobiography draft Jan 15, 2013 Page
Download 412.6 Kb.
|
AuthorshipI was intensely involved for for quite a while, working at the Honeywell facilities. I came to appreciate that the Oracle documentation was terrible. They told you very little about how to use their software, although the software itself was quite good. I saw a major need for a book. I compiled a syllabus for the use of the people on the Honeywell team, and one of the women, Kathy Christian, said that it really should be a book I agreed with her and went to McGraw-Hill. They accepted the book, and after the project was completed I spent some time writing it. Working with Oracle Development Tools was published in 1989. It served as a good reference as I sought business. I enjoyed writing a great deal. It was a challenge, as I was working alone, just me and my wordprocessing system. I learned quite a bit about myself and about the English language. Ultimately the book was reasonably successful. It filled a significant need. After a few years, however, Oracle had changed the software so much that my book was no longer relevant, and they had made the online help that accompanied the new system good enough that my book was no longer really needed. Moreover, Oracle Press started publishing, and they had their own book on the topic. My first book gave me a taste for authorship. I next reflected on the difference in philosophy between developing with the Oracle software and developing with IBM software. The IBM software required a structured, hierarchical approach. There was a rule of thumb for developed by Frederick Brooks, one of the pioneers, that said that you had to take a waterfall approach. You needed to take a needed to write a functional design first, and get everybody to sign off on what the function should be. After that, you needed a detailed design – a design for each program to be written in the system. After that you would program the system. Brooks’ rule of thumb was that it if you made a mistake at any level, you incurred a 10 fold increase in cost remedying that mistake downstream. Therefore it was highly important to rigorously follow the structured path as you develop software. The real world doesn’t work like that. Functional needs change frequently in the course of the development of an application. As you start writing it, you realize how to improve it. You need flexibility. The Oracle system allowed that level of flexibility. In other words, Oracle favored a system of developing on the fly, improving as you go, without spending a great deal of time on the general design. They call it rapid application development, or RAD. This required a new way of thinking for established programmers. That was the title of my second book, Managing in an Oracle Environment. It required a different set of skills, generally smaller teams, and generally less hierarchical control been developing in a traditional environment. People were able to get applications up and running quite quickly with Oracle, and then improve them over the course of time. The second book was reasonably well-received as well, by the same development community. Once again, this is been overtaken by events. This rapid application development is now old hat, and there are many new philosophies that have incorporated it in supply and fit it. So once again, that book is no longer selling many copies at all. In the early 1990s you could see a change in direction and the programming field. More and more programming activity took place at the systems level. The programming of real end-user software like accounting, inventory control, accounts receivable, project management and so on was becoming standardized. It was being done using productivity tools called CASE – computer automated systems engineering. The idea was that you develop the requirements using a high-level schematic tools and they would automatically generate the computer code that you needed to execute it. In other words, it was a taking computer to computers to a higher level of abstraction in the design and creation of software. This looked to me like the way things would be going, so I decided to align myself with the movement. I found a couple of experts in the Washington area who had a company. The company was a startup, rather thinly funded, but they looked to me like they knew their stuff. I made and I made an offer to join their company, giving them some financial support. This turned out to be a mistake. These were not particularly honorable people. They were happy to use my financial support. They didn’t cheat me, but on the other hand they did not cut me in as a full partner in the organization and they didn’t trust me. I left them, and was out on the street once again looking for work in 1994. The other news at this time was the revolution in financial software. Almost a decade earlier I had been involved with computer sciences as we installed the McCormack and Dodge line of accounting software for this Saudi Ministry of Petroleum and Minerals. This accounting software had been written for mainframe computers, IBM 370 computers, in the COBOL language. Oracle Corporation decided to implement the same applications using their highly productive interactive tools. I was excited by this; it made a great deal of sense to me as a developer. I resolved to become an expert in the Oracle Financials software. The quickest way to learn was to get involved in a project to use it. I went back to work as an employee, the first time in 15 years, taking a W-2 from the Mitre Corporation in Virginia. They were involved in a large-scale implementation of most of the Oracle Financial products. On that project I learned quite a bit about how the software works. I also observed problems when software is oversold. The software did not exactly fit Mitre’s needs, and Mitre was not realistic about adapting their needs to the capabilities of the software. The result was a cost overrun that almost broke the Mitre Corporation. I, however, learned quite a bit from it. I formed an association with a fellow who had been working with the software quite a bit longer, a certified public accountant named Joe Costantino. The two of us went into business in 1995. Joe found my first client, Black & Decker in Towson Maryland. I worked with them and deepened my knowledge of the system. Then we found some business with Watkins Johnson Corporation closer to home, in Gaithersburg, Maryland. Their problem was that in converting from their legacy system, which had run on a Hewlett-Packard 3000, to Oracle they were not able to convert the entire database. One of the most important parts of the database, the bills of materials, describing how their products (electronics components) were built, did not convert over to Oracle. They were between a rock and a hard place. They had Oracle software running their procurement their purchasing, their sales and accounting, but they did not have the ability to make products using the Oracle system. They still used the HP system for that. It was a very awkward stretch. They had tried several times to convert the HP bill of materials database, and each time failed. This was kind of a replay of the problem that I had solved for Computer Sciences. They needed an innovative approach to converting in a semi automated fashion from one system to another. Oracle needed more information than HP, and it could only be added manually. I proposed that instead of converting the entire database at once, they convert one product at a time, and I developed a custom program to convert one product at a time. The system I prompted them in a user-friendly way for the information that was missing. They were able to use the software to convert one product at a time. Like most manufacturers, the great majority of their production was concentrated on a relative handful of products, so they were able to convert the bills of materials for these products quickly and get themselves back in business under Oracle. They had of large catalog of products that they made infrequently, mainly older products. We resolve to convert them on an as-required basis. This was the insight – that the whole job could not be done manually, and that would only work if they manually tweaked each product as it was converted. At any rate, we got it converted over to Oracle. It turned out that there were several other small jobs that needed doing. I stayed on with Watkins Johnson for a total of about two years, and was able to bring in a team of as many as five or six people working on this project. We developed a good system, had a great time, and made some money and some friendships. Two of the people I brought onto the project got married, and two others used it to become independent, and are still independent these fifteen years later. It was really one of the best experiences in my professional life. However, all things must end, and at some point the accountants, the bean counters at Watkins Johnson added up how much we were costing and decided to replace us with employees, which is the usual fate of a successful consultant. I engaged in one more similar rescue mission with the Micros Corporation. They had similarly converted from a legacy system. They did their purchasing in their sales through the Oracle system, but they continue to do their manufacturing through the legacy system. This was what they had planned. What they had not realized was that the accounting system required that all accounting records be consolidated within one system. They were unable therefore to produce valid accounting reports. This was a major functional problem. I came in, and writing the same type of custom software bridge software, was able to write a series of programs which would import data from their legacy system and put it into Oracle so they could do financial reporting all out of the Oracle system. This worked on an interim basis until we were able to work together to get the entire manufacturing process running under Oracle. Once again, as with Watkins Johnson, I was with Micros systems for something over a year as we got all this done. I found a few other small things that needed to be done, and made myself generally useful. In the end, however, I was out of work. At this point I had quite a bit of experience working with the Oracle financials and I approached McGraw-Hill about writing a book on the subject. They said that they already had an author, a fellow named David James in England. David, however, was busy with client work and hadn’t made as great progress as he had hoped. I approached him and proposed that we work as co-authors. His expertise was in the financial area, and mine was more in the manufacturing area. It was a reasonable collaboration, and we came up with a book which is still in print as I write this, more than a decade later. Working on the book improved my knowledge of the system, especially the parts that they he knew, the accounting. That was a good education for me. Two things came out about this time. First, my partners in the business rather resented my work on the book. They said that it was not bringing in money. The fact was that I billed more than anybody, and I was responsible for bringing in the work on our ongoing projects. Nevertheless, the partner that I had chosen as a successor to Joe Costantino, as Joe left for the West Coast, left me out of the bonus distribution in my last year with the company. I had asked him to propose a distribution. He did more than propose it – he did his own arithmetic, computed the bonus pool, and that the distribution, leaving me with no bonus. That was an outrage. I quivered with rage. I told him that I was going to take my own bonus, which I did it out of earnings, and simply fold the company. That is what I did in 1998. Thus once again out of work, I surveyed the situation. I didn’t really need work because I had invested in Oracle and then Cisco Systems early in the 90s, and both companies have done extremely well. As you will note elsewhere in this biography, my marriage had not been doing terribly well, but my wife was quite successful in her work. She had more than adequate income. Also, my real estate investments had paid off, and had appreciated quite nicely. So among all of our activities we had more than enough income. I decided at that point that I really didn’t need to work anymore. It was a logical place to stop, and that’s what I did. I retired, except for writing books, and started doing things that I enjoyed. Download 412.6 Kb. Share with your friends:
|