[OSGeo-Discuss] C++ library for "gridding"

[Michael Barton]

This will be interesting to follow. First a question, then a couple things
to consider with respect to the GUI.

Will this new API be accessible from the wxPython (wxWidget) GUI platform?

With respect to a GUI, one of the great things about open source is that you
can adapt the code in many ways. However, you might want to also catch up on
the extensive GUI discussions that we¹ve had over the past year. A long
thread beginning about a year ago, and continuing over several months
focused on the design specifications for a next generation GUI for GRASS. A
follow up thread discussed the advantages and disadvantages of available GUI
platforms from the standpoint of interface potential, useability,
cross-platform support, and code maintainability within the GRASS
user/developer community. A couple items are worth summarizing.

First, just because ArcGIS has a very large commercial market share doesn¹t
mean that it has the best UI or that an open source project can¹t create a
better UI for GIS. There was considerable consensus around a number of
features that many of the GRASS community felt were essential in a UI but
which are lacking in ArcGIS (e.g., a CLI, efficient use of screen real
estate with a single control for multiple windows) and more diversity about
other design specs (tightly or loosely coupled 3D visualization and
cartographic capabilities).

The second is that QT and wxWidgets emerged as the leading GUI design
platforms. Of these, wxPython (Python implementation of wxWidgets) garnered
the most support because it is easier to develop in and more members of the
GRASS community have programming skills that can support it. Relatively few
people are QT experts, but many more people in the GRASS world have Python
experience.

Many of the initial GUI design specs have been implemented in the current
TclTk interface in order to test and refine this design. Others will require
a GUI platform with capabilities beyond TclTk. A wxPython GUI is already
present in the CVS in prototype test form and is actively being developed.
You might want to take a look at these to see where they are at and whether
(and how) they might interact with you new libraries.

Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution & Social Change
Center for Social Dynamics & Complexity
Arizona State University

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton




From: <discuss-digest-help at mail.osgeo.org>
Reply-To: <discuss-digest at mail.osgeo.org>
Date: 24 Oct 2006 01:40:30 -0000
To: <discuss at mail.osgeo.org>
Subject: discuss Digest 24 Oct 2006 01:40:30 -0000 Issue 120


From: <chenrg at lreis.ac.cn>
Date: Mon, 23 Oct 2006 09:57:20 +0800
To: Song <xfsong at gucas.ac.cn>, <discuss at mail.osgeo.org>
Subject: Re: Re: [Fwd: Re: [OSGeo-Discuss] C++ library for "gridding"]

Hi, all,

I had 2 days out at the weekend. Thank you for your attention. I'd like to
say, we just started the work,and the initial tech docs were in chinese.
I'll try to explain what we'll do for the project.
We're working on the current CVS-HEAD (GRASS 6.3). A new directory called
API was added to the main directory of GRASS. The high-level APIs will be
based on the files of GRASS commands. and will form 7 new dynamic libraries
(.so for linux, and .dll for windows), which will be updated along with
GRASS CVS. The ArcGIS-like GUI will call these APIs for data processing. The
work will join grass cvs once some progress made.

Regards

Chen


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/discuss/attachments/20061023/b3713195/attachment-0001.html>