[GRASS-dev] adding lib_arraystats

[Dylan Beaudette]

On Feb 13, 2008 12:10 AM, Hamish <hamish_b at yahoo.com> wrote:
> Dylan Beaudette wrote:
> > > However-- this is starting to get into the realm of feature
> > > bloat...
> > > As neat as it would be to have R routines which could work directly
> > > on GRASS data structures, it might be overkill considering the
> > > possible amount of work. I'll keep checking around and post back
> > > other ideas.
>
>
> Well, people always seems to ask about kriging because they've heard of
> it.
>  http://lists.osgeo.org/pipermail/grass-user/2004-May/025709.html
>
> (see s.surf.krig in GRASS 5 and v.surf.krige on the wiki addons page)

This does seem to come up a lot. However...

<rant>
This is one piece of functionality I would rather see in a specialized
system-- i.e. implemented through gstat etc. Kriging is one of those
semi-mystical things that gets lumped into many GIS courses, and
through the magic of automated wizzards and such-- out pops a map.
Unlike RST / IDW, kriging requires several important steps and
decisions (variogram estimation, assessment of stationarity, etc..)
which the act of using something like gstat (partially) ensures.
</rant>

That said-- this is one of the cases where a strong tie to R is a
great thing. The use of an embedded R (per my previous comments) might
facilitate these more complex operations (kriging) in a way that is
not abstracted away from the wealth of good documentation (thinking
gstat here).

Without much trouble it would be possible to document a standard
kriging exercise with the spearfish data (or the new NC data)-- such
that the concern expressed in the linked thread
(http://lists.osgeo.org/pipermail/grass-user/2004-May/025709.html) can
be addressed. Here is one such (not necessarily the best) example:
http://casoilresource.lawr.ucdavis.edu/drupal/node/438

>
> > > On the other hand-- R has so many great plotting / analysis
> > > routines, it would be a shame to re-implement these in GRASS if
> > > they could be easily "plugged-into" via some kind of share library
> > > / SWIG interface.
>
> Yup, it would be nice to be able to provide modules with R as the
> backend versus writing/maintaining something new ourselves.
>

I suppose that these things really belong in a discussion on GRASS7.
I'll try and keep on topic!

Dylan

>
>       ____________________________________________________________________________________
> Looking for last minute shopping deals?
> Find them fast with Yahoo! Search.  http://tools.search.yahoo.com/newsearch/category.php?category=shopping
>
>