awk, site files, etc

mike camann camann at pick.uga.edu
Tue Jan 25 18:37:47 EST 1994


Chris Skelly recently wrote:

>Yes, but the point I think the others were making was that if there
>are two ways to do something lets go the route best for the user, not
>the programmer. Having said that...if I had to make a decision on
where
>to put GRASS programmer time at this point it would _not_ be on
>"desk top" GUIs. It would 98% on analysis functionality. If I want a
>neato-jiffy point click and gee look at the diolog box I will buy
>a PC commercial product, eg MAPINFO. GRASS is not IMHO designed to
>impress computerphobic managers, its for researchers and people who
>know enough about what they want that they might want to tinker with
>the code.

Unfortunately, the original issue-- the one that started this
discussion-- wasn't about fancy GUI's or other bells and whistles.  It
was about whether to expand the sites file format to include elevation
and possibly other ancillary data.  I think that this indeed add
analysis functionality, but I agree with Bill Hargrove in the sense
that users should understand the inner workings of the file
manipulations their analyses are performing, and *nothing* fosters more
knowledge of grease-monkey file manipulation than using the built-in
Unix file processing tools to do the job.

Sometimes though, the temptation to cobble something together with
shell scripts should be avoided.  It's a great way to prove a concept
or work out a method, but in the long run GRASS functionality won't be
extended with subdirectories full of shell scripts.  As GRASS's
functionality is extended with new tools it is reasonable to expect its
underlying file structures to change as well so they can exploit the
possibilities inherent in those tools.

One final thought on the site file format discussion: why not add a tag
to each field, i.e.

north=83401.65|east=4578.9|elevation=1638.0|label=burial_mound  etc.?

Then sites programs could parse each line in the file and extract
*whatever* data they need.  Users building site lists could put any and
all data they want into the site description.

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Michael Camann                          camann at dial.pick.uga.edu
Department of Entomology                camann at phoenix.cs.uga.edu
University of Georgia                   (706) 542-1388
Athens, GA 30602                        (706) 542-2276

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%



More information about the grass-user mailing list