TERMINOLOGY

Manual pages, help use:
raster map (most), raster file (some), 
for imagery or some modules use also raster layer or just layer

Book uses
raster layer, raster map layer
map is used for graphical output that may include 
several raster and vector layers as well as scale, north and legend

lets use RASTER MAP

   UPDATE Dec 13, 2006: I have changed entire GRASS from
                        raster/vector file to raster/vector map
Book:
GRASS data models, 

Manuals
GRASS data type -> sounds confusing

Book and manuals?
raster data type (CELL, FCELL)

r.colors
why do we have the same name of col=rules
and then the option is also named rules=
after some thinking it looks logical, but at first
it was really confusing - the option rules
should be better explained in the man page.

v.out.svg
option precision is not explained very well -
what does precision=0 mean?
 - m3: have asked the author

d.legend
at=bottom,top,left,right
Screen coordinates to place the legend (as percentage)
Options: 0-100
this is confusing

r.region 
man page talks about raster map region and boundary - it should be
consistent raster map boundary or spatial extent - as it is now
it is somewhat confusing

r.recode 
says is for raster categorical maps - can we use it for e.g. elevation
to recode to categorical map?

r.mapcalc 
manual says first that division by 0 leads to 0, then that it leads to NULL
FIXED by Glynn

r.to.vect
under bugs it says
For feature=line the input raster map MUST be thinned by r.thin; if not, r.line may crash.
it should probably be r.to.vect instead of r.line

r.digit
confusing parameter bgcmd, left button does not work correctly on Mac

r.circle
it is not clear what the parameter mult does (I have completely misrepresented it)
also if I give it to start at 200, it actually starts at 209 (that is on 10m res. region)

r.random.surface
uses old terminology for category value -> cell value, category label - > category value

r.statistics
refers to old limitation to integers 
This is nice to avoid the GRASS limitation to interger in raster maps because 
using category values floating point numbers can be stored.

uses layer instead of raster map
The output-layer is a reclassified version of the base-layer with identical category values. The results of the calculations are stored in the category labels of the output-layer.

r.coin 
has messed up formatting

r.surf.area
does not say what units is the result in