15
Fall 2013 Newsletter of the FCIAAO
REAPING THE BENEFITS
My name is Lee Herbst and I am the Director of Information Resources for the
Marion County Property Appraiser. I am a Certified Florida Evaluator and a Mi-
crosoft Certified Solutions Developer, which means that I am trained to design
and develop enterprise class software using Microsoft languages and tools. I
have more than 30 years of IT experience and I have been working for the
Marion County Property Appraiser for 19 years.
When I started working for the Property Appraiser in 1994, we were running
commercial CAMA software called OASIS on an IBM mainframe. In 2002, we were asked to cut costs
and the director at the time decided to eliminate the most expensive component on the mainframe,
VSE, but that left us with VM and forced us to convert to an in-house developed CAMA system running
on mainframe web servers. We developed Wisard and it went live in time for the 2004 roll year.
In 2006, we determined that Wisard, was no longer meeting our needs. Although, less expensive than
running OASIS, hardware and software maintenance still exceeded $100,000 per year. Wisard had
also been poorly designed, its performance was mediocre and it was difficult to maintain due to a lot of
redundant code. A single change often meant modifying six or seven different programs. This caused
severe bugs whenever one or more changes were missed. We had become reluctant to add new fea-
tures that we needed and wanted.
We needed a solution, so we turned to the various CAMA vendors. One rose to the top of our short list.
It provided the functionality we were looking for and demonstrated excellent performance. It was ex-
pensive and would cost us more than 2.8 million dollars over the next 4 years. We were tempted but,
at the end of the day, the high cost turned us toward another direction.
The Property Appraiser, The Honorable Villie Smith, CFA, was reluctant to consider an in-house solu-
tion after the disappointment of Wisard, however, he approached me and asked if I thought we could
accomplish our goals in-house. I felt that we had the skills necessary to avoid the mistakes of the past
and create a CAMA system that he could be proud of. We agreed that we could limp along on Wisard
until a new system could be completed. We could design the new system exactly the way we wanted it
and continue to reap the benefits of an in-house staff for support. He decided to move forward with the
idea and Merlin was born.
Our goals for Merlin included no redundant code, high performance, flexibility, and lower costs. I de-
cided on a client/server architecture using object oriented methodologies. The client/server architec-
ture would allow us to move to less expensive PC servers and using object oriented methodologies
would allow us to develop the system without redundant code.
During the development process we faced several challenges; the first of which was training. It would
have been impossible for me to do all the work by myself in a reasonable amount of time. Luckily, my
PC tech, Sherry Sherouse, said she was willing to learn, so we got a set of training videos and she
started on the road to become a programmer. She picked it up quickly and started coding in less than
two months. She proved to be very capable and between the two of us, we were able to complete the
development of Merlin.
Another challenge we ran into was data conversion. Most of it went fairly smooth; however, some of
(Continued on page 16)
Lee Herbst
