GRASS-DBMS interfacing

[kp5 at prism.gatech.edu]

A few days ago I posted a message/request to the GIS-L list requesting info
on DBMS and GIS interfacing.  Since I'm primarily interested in GRASS
rather than other GIS's, and since this audience is much more specific to
my area of interest, I'm repeating it here.

Hi,

Here at the Georgia Tech Research Institute, we're currently performing a
study into different strategies for interfacing Geographic Information
Systems and Database Management Systems.  For example, is it better to try
to build interface routines to let a commercial DBMS talk to a GIS, or
should you try to modify the GIS to use DBMS data directly?  What are some
of the problems involved in doing these types of links?  What real-world
experiences have other researchers had in working with extensible data-type
DBMS's (like Postgress)?

We're focusing on the GRASS GIS in particular, so I'm
most interested in hearing about experiences people have had
in linking the GRASS GIS to any type of external DBMS, or in 
integrating more powerful DBMS capabilites into GRASS.

This study is being performed under contract to CERL, the developers of
GRASS, and your inputs could potentially be used to design new features
into future versions of GRASS.

I'll be happy to share information and points of contact.

Thanks,

  --Kirk

   Kirk Pennywitt                       Georgia Tech Research Institute
   kp5 at hydra.gatech.edu                 347 Ferst Drive, ERB 173
                                        Atlanta, GA  30332
                                        Phone: 404-894-5248
                                        FAX:   404-894-3906