On Fri, 2010-09-03 at 12:22 +0200, Damiano G. Preatoni wrote: > IMHO the problem is: write code for QGis only, or stick with an architecture > based on R for the calculation engine and QGis for a GUI? > > This is because HR calculations either are "single shot" (i.e., few animals, > experimenting with methods, etc, and is better to habdle them from inside > QGis) or "batch" (i.e. thousand of fixes and hundreds of animals handled > automatically inside R). > > > Any ideas (say, Paolo, Anne, Cleḿent, Boris?)
Hi all, I see the advantage of pure QGIS implementations for small-sized analysis - that introduces code duplication, however. OTOH the dependency hell of QGIS+R+Python+rpy2+Rpackages can be justified when data are really a lot and there is the need of a simple GUI. If data are a lot and there are programming skills available, R/scipy scripting is usually more effective. I'd like to see a more organic development of animal movement (sensu lato) tools for QGIS, from the first steps, so it would be nice that we all use the wiki and the mailing list also for initial brainstorming whenever possible: the tools can already be present, like it has been the case for triangulation. For the integration of the tools in a common toolbar/library, Borys and me have already some ideas, anyone who wants to participate is mostly welcome. My 2c, Anne -- http://wiki.osgeo.org/wiki/Anne_Ghisla
signature.asc
Description: This is a digitally signed message part
_______________________________________________ AniMov mailing list [email protected] http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
