[OSGeo-Discuss] Re: Staistical analysis support needed

Christopher Schmidt crschmidt at crschmidt.net
Mon Sep 13 06:08:36 PDT 2010

On Mon, Sep 13, 2010 at 04:02:44AM -0700, mayank_agarwal wrote:
> Thanks Christopher,
> I think R is a really good option to use, as GRASS is a desktop GIS.

Er... and? I mean, R is a 'desktop statistics program', right? :)

GRASS can be seen as a command line tool for performing analysis
operations on a set of geospatial data. It can also be many other
things, but calling it a "Desktop GIS" and dismissing it out of hand as
a result is not really fair. 

(You might say something like "Programmatic access to the statistical
options with R will be simpler than doing similar operations with
GRASS"; that *might* be true, though I'm not actually sure.)

In the end, for geospatial analysis, my expectation is that both GRASS
and R will behave similarly; both have a psuedo-language that you use to
operate on data and generate more data / statistics.  Each will have
different pros/cons, but "it's a desktop GIS" isn't really either :)

Christopher Schmidt
Web Developer

More information about the Discuss mailing list