786                                                                         

 System  -  Author's Projects 3
Author (of AZMAT System)

Continued . . . .

6.  At the same company there was another system developed by internal staff, which produced a daily calculation report for dealers that took 3 minutes after the prices were entered by the dealers.  Until that calculation report was produced in the morning, the dealers could not begin trading.  

The dealers obviously would have liked a faster system but as they had not known any other system, they accepted it as a fact of life that the calculation report always took 3 minutes.  The author examined what they were trying to achieve, looked at what they were doing and suggested that he could do better.  The senior manager had no doubts whatsoever in the author's ability to produce fast and really efficient working systems, after his earlier work where he cut the report time from 18 minutes to under 3 minutes. 

It was another weekend's work.  The senior manager was also very interested and curious.  He couldn't wait till Monday morning and actually came on Sunday evening to see how the work was progressing.  It must be stressed that the working relationship was such that the manager was not checking to see whether the author was working; there was mutual respect and total trust between them.  The author simply would not work anywhere if he even suspected that the trust was not there. 

The author was in the middle of testing the new program.  He 'ran' the new report program and the senior manager exclaimed that there must be something wrong as he had never seen that report printed so fast in his life before!  The author had designed and developed the new system that took 7 seconds, instead of 3 minutes, to prepare the calculation report.  The dealers and of course the senior management were all very happy indeed; the new system enabled the dealers of that company to start trading just 7 seconds after they had entered their prices.   Being ahead almost 3 minutes in the hectic financial markets, when competing with rival dealers all over the world, can make a huge difference to the profitability of a company. 

7.  At another company there was a team of highly educated personnel; two people had 1st class MSc in Mathematics, another two had PhD degrees and the fifth person had two PhD degrees.  That team had developed a very good set of calculation routines (programs) in 'C' programming language over a period of several years.  However, less than half a dozen dealers in the London office of that company could make use of those calculation routines. 

That was because the users also had to be highly educated in order to understand the relevant calculations, and a great deal of technical knowledge was also required before they could use those routines in their spreadsheet models.   Furthermore, once such spreadsheet models were constructed by such users, in their absence nobody else could make use of those spreadsheet models; only the person who had constructed a particular spreadsheet model knew how to use that model.  Also, certain calculations were carried out according to each person's knowledge and ability to transfer that knowledge into a spreadsheet model.  Thus the company did not have a uniform method of calculations in spreadsheet models. 

The author joined that team and began his work on a system that people of any ability could use, and one that would make use of the existing calculation routines.  There was no written program specification and he had to use his own initiative and imagination.  In the financial field as anywhere else, such a situation is fairly common; a system developer can not simply rely on the users to tell him or her what the system should do.  The users might describe in very broad terms what they want, but a good system requires a lot more than that; it is like an iceberg - there is a lot of background design and programming work which is necessary but is usually not seen by the users.  The systems which fail are usually the ones which do not take care of this background work in both system design and programming areas. 

The author, using his own method of working, did eventually provide the users with a system that was fast, flexible and user-friendly.  He designed and developed the system which used the original team's calculation routines but one that could be used by almost anyone with minimal knowledge of spreadsheets.  The system did not require from the user knowledge of how calculations were carried out.  The system also ensured that the company had a uniform method for all spreadsheet calculations.  When finished, his system was being used by several hundred people in that company all over the world. 

This practical example shows that the author managed to do what the original team of highly qualified personnel clearly could not, over a period of several years.  Thus, you do not need to have a PhD degree to develop a good financial system.  Indeed, if you do have a PhD degree it is not alone sufficient to enable you to develop a financial system.  You do need other skills, and not just programming skills either. 

Continued . . . .


System  -  Author's Projects 3
AZMAT System   Copyright ©  1997-2017  Micro Minerva Consultants Ltd.   All Rights Reserved.