[R-sig-Geo] Getting started with Open Source GIS

[Dylan Beaudette]

On Sat, Oct 17, 2009 at 8:13 AM, Barry Rowlingson
<b.rowlingson at lancaster.ac.uk> wrote:
> On Sat, Oct 17, 2009 at 3:46 PM, Ezra Boyd <eboyd3 at tigers.lsu.edu> wrote:
>
>> Will I be able to complete the tasks described above using the the Open
>> Source GIS without a bunch of time consuming complications?  Or, should I
>> fork over some cash to extend my Arcview license?
>
>  What's Arcview going to get you? You will be able to do all your
> statistics in R - get a copy of Applied Spatial Data Analysis in R,
> which will tell you the basics and more about manipulating spatial
> data in R. Remember that statistics is never plug and play - even if
> there's statistics functions in Arcview, you still need to check
> diagnostics and other assumptions before your conclusion is valid. I
> don't think Arcview or other ESRI products will give you that
> flexibility.
>
>  Once your stats are done and checked and you want to make pretty
> maps, then you can create shapefiles from R and load them into Qgis
> for mapping with any other layers of information you want to put in.
> It is even possible to make publication-quality maps in R using the sp
> package and friends (as in the ASDAR book).


Good points so far, but I wouldn't be that quick to relegate GRASS to
obscure C-based tasks. I find the the combination of GRASS+R is an
efficient setup, as the two packages are very much complimentary.
Also, you are going to need GRASS if you would like to work with data
sets that won't fit into memory.

Cheers,
Dylan


>  I wouldn't worry about GRASS - it's great for hardcore spatial
> analysis (buffer this, overlay with that, compute watersheds, clip to
> river buffers, count population in areas,...) but if you need to go
> outside the boxes given you're going to have to start writing in C...
> It's nice to know the power of GRASS is there, but I try and avoid
> using it!
>
> Barry
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>