I agree. It's difficult to see any reason that someone who currently
uses R/GRASS would go to ESRI for spatial statistics. I come from a
long history as an ESRI user with an interest in spatial stats. As
others have said, ArcGIS may give you the basics quickly and easily, but
it is limited and expensive.
Actually, ESRI giving spatial statistics any kind of attention is a
bonus for R users. The new ArcGIS spatial stats functions may draw in
some typical GIS users to take a look (just like the ArcGIS GeoStat ext
did years ago), but once they need to do more intensive work, or modify
parameters, or deal with larger datasets, or get tired of crashes,
etc..., they will look elsewhere (again, like ArcGIS GeoStat did), and
R/GRASS is right there waiting.
- John
****************************************
John Callahan
Geospatial Application Developer
Delaware Geological Survey
University of Delaware
227 Academy St, Newark DE 19716-7501
Tel: (302) 831-3584
Email: john.callahan at udel.edu
http://www.dgs.udel.edu
****************************************
Thomas.Adams at noaa.gov wrote:
Paul,
Not to mention the huge cost of ESRI products and the fact that they run only on the MS-Windows platform. R also integrates well with GRASS GIS. Personally, I can not see any compelling reasons to go the ESRI route vs R/GRASS?
I feel a great debt of gratitude to the selfless developers of R & GRASS!
Tom
----- Original Message -----
From: Paul Hiemstra <p.hiemstra at geo.uu.nl>
Date: Thursday, July 3, 2008 7:54 am
Subject: Re: [R-sig-Geo] Impact of ArcGIS 8.3 beta's new spatial statistical functions in its Spatial Analyst Extension
Dear Hisaji,
I think that the choice of people for R is not only based on the
fact
that a certain method is only available in R. R is much more
flexible in
working with data, allows scripting, allows you to see exactly how
a
method works, allows you to extend or develop methods and if you
write
in script files it also provides a record of what you have done.
Apart
from spatial analysis, R also provides a very wide range of
"normal"
statistical methods that are not available in ArcGIS (linear
regression
and such).
So I don't think a lot of people will forsake R once a few more
methods
become available in ArcGIS.
regards,
Paul
Hisaji ONO wrote:
Hi.
What do you think about ArcGIS 8.3's new statistical
functions(Calculation of spatial or network, GWR,
Global/Local Moran, Gi etc.) in its Spatial Analyst
Extension described in
Current R's spdep's and spgwr's and others' end users
will loose interests of them and go to use ESRI's apart
from R's?
And will decrease activities of developments of
(geo)spatial data analysis tools in R?
Regards.