I'm pleased to announce the release of Ecolab.4.D25. This is the first
major release of EcoLab for almost 2 years, and consists of a major
effort to eliminate all known bugs in the package, prior to starting
work on EcoLab 5. There are a few new features added as well. See
http://ecolab.sourceforge.net
Please log any problems and bugs with the sourceforge bug tracker.
EcoLab 5 represents an opportunity to redesign/refactor the package,
retaining all the good points of Ecolab 4. At this stage, no code has
been written, but it is just a concept in my mind to make this an even
better C++ ABM environment. I'd like to open up discussion with the
modelling community to see what people would like. I'm particularly
interested to hear from people interested in using C++, and who use
EcoLab now, or have tried EcoLab and found it wanting, just what it is
you'd like in a C++ modelling package.
My preliminary thoughts at this stage are:
* Make a more uniform syntax for declaring objects acessible from the
scripting interface: At present we have NEWCOMMAND, make_model and
TCL_TYPE for static declarations and Tcl_Obj for dynamic creation.
* Improve the handling of arguments to be more like native C++
(instead of the current TCL_args)
* Generalise scripting to other languages/environment. Environments
I'd like to target as well as TCL are Perl, Python and
Java. Possibly use SWIG to provide support for the range of
environments supported by that package.
* Integrate EcoLab engine into a package like RePast to allow RePast
visualisation tools to work with EcoLab compiled models in
interactive mode. Batch processing will still be done independently
of RePast.
Cheers
--
----------------------------------------------------------------------------
A/Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 hp...@hp...
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------
|
