grass4.0 performance

[Kenn Gardels]

> 
> Very interesting mail.  Could you please tell me how to get these Arkansas
> Database tools?  We are beginning a project which will link databases to 
> GRASS.  If it has already been done, we may not have to write all the code.

I think I opened a can of worms with that comment.  A group at the
Center for Advanced Spatial Technologies in Fayetteville, in
cooperation with CERL and SCS, has been developing a set of tools
linking GRASS to COTS DBMS.  [COTS = commercial off the shelf]
My understanding is that selected sites have been testing beta
versions of this software, and it may be included in some form in
GRASS 4.1.  I'll leave it to them to be more specific to the list
(hint hint ... ).
> 
> Also, I have been told several times that GRASS is not good at working with
> raster maps.  This is in reference to computations, statistics, etc.  This 
> is apparently the opposite of what most people think.  Which is true?
> 
I havn't heard these comments, but it might be in reference to the
amount of flexibility (or lack thereof) in setting parameters or
procedures for certain functions.  Ie certain spatial functions like
resampling are done only one way, and the user can't change it
(without writing new code).  Too often, though, the user doesn't know
the method or understand its implication, so it becomes and education
and disclosure issue rather than a technical one.  

-kenn