Subject: | |
From: | |
Reply To: | |
Date: | Wed, 14 Feb 1996 13:09:13 -1000 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
The loss of Microsoft's DOS FORTRAN compiler is an inconvenience, but
should not be a major set-back to our modeling. It has at least prompted
a re-evaluation of programming languages, which is good. Thanks, Brian.
I know at least some of the IBSNAT/ICASA modelers have experience with
object-oriented programming languages. For a lot of reasons, I hope the
IBSNAT/ICASA network will start to use object-oriented modeling
techniques. While FORTRAN-90 might be used for this, C++ is my personal
choice at this time.
I have recently sealed off a version of some C++ software that I am using
for most of my modeling work. The software is called JanuSys and is
available on the World Wide Web at:
http://agrss.sherman.hawaii.edu/caldwell/janusys.htm
JanuSys provides an object-oriented framework for modeling and simulating
hierarchical systems. It helps organizes the creation of families of
submodels (e.g., crop models) that can then be integrated to simulate
agricultural systems.
I am releasing the JanuSys source code now, and announcing it
especially to the IBSNAT/ICASA group, because I have verified that
(1) we have recreated the plant component of CERES-Wheat within the
JanuSys framework, and (2) we can simulate any sole crop of wheat
as a mixture of any practical number of competing sub-populations
(i.e., the rule I have been using in the development of CropSys).
Some in the IBSNAT/ICASA network may be curious how an IBSNAT model
might look in C++. You are welcomed to download the source code
and use it in any way you would like.
Aloha,
Robert M. Caldwell, [log in to unmask]
Department of Agronomy and Soil Science
1910 East West Road
Honolulu, HI 96822
Phone: (808) 956-6835; FAX: (808) 956-6539.
|
|
|