Helena
(cc to GRASS 5, sorry for so much traffic today)
On Fri, Nov 02, 2001 at 11:24:12AM -0500, Helena Mitasova wrote:
Markus
I meant d.leg.thin or d.legend, because d.leg.thin helps when you have
continuous data (in fact Bill did it when I started to do more continuous
data stuff with s.surf.tps and r.flow) but it does not resolve the problem
of placing the legends for category data and for continuous data it is
useless for data with non-linear values (such as flowaccumulation - just
try to draw a legend for the dsout output of r.flow). All of this was
solved with legends for SG3d (you can chose discrete or continuous, you
can place the legend with mouse both for discrete and continous and you
can define the values for which the color is shown for both discrete and
continuous and you could even chose the size of the font - the
capabilities are described on the web), however nobody seems to be
interested to include it into grass5 although that means that for many
maps you cannot really display your data in a meaningful way.
I fully agree - the legends problem is severe.
For the others: The SG3d code is here:
grass_exp/src.contrib/GMSL/SG3d
grass_exp/src.contrib/GMSL/sg4d
-> experimental tree, not stable branch.
Especially sg4d is great software, see
http://www2.gis.uiuc.edu:2280/modviz/viz/vol1.html
I am writing
this long mail to urge you to put the legends issue on top of the agenda
for the developers meeting, it shoudl take an experienced programer few
days to implement it when using Bill's SG3d code.
[...]
People just now expect to get the legend along
with their map as ArcView does it and parctiacally all of on-line mapping
services do too. So not being able to get the legend when looking at your
map is a big turn off for people who may want to use GRASS.
... very true...
For spearfish when you do the legend for geology you get either huge
squares and letters, or if you run it as if it was a continuous map so
that you can get the legend smaller then you do not get the cat labels for
all colors. The easy workaround how to get the legend along the map is to
stretch the window to make space for it (rather than using frames which is
how you are supposed to do it, but it is really inpractical - it is used
in d.display).
[...]
Helena
All open for comments...
Markus
just a note - the reason I cannot get cat labels for all categories when I use
the workaround
with the smooth gradient option is that the smooth option always writes only 5
values/labels,
so I get nice legend for soils.ph which has 5 categories and I am out of luck
with geology
which has about 10. I am working with spearfish data set and I guess I must be
quite frustrated
given the number of misspellings I managed to make when writing to Markus.
Sorry for that,
helena
Helena
(cc to GRASS 5, sorry for so much traffic today)
On Fri, Nov 02, 2001 at 11:24:12AM -0500, Helena Mitasova wrote:
> Markus
>
> I meant d.leg.thin or d.legend, because d.leg.thin helps when you have
> continuous data (in fact Bill did it when I started to do more continuous
> data stuff with s.surf.tps and r.flow) but it does not resolve the problem
> of placing the legends for category data and for continuous data it is
> useless for data with non-linear values (such as flowaccumulation - just
> try to draw a legend for the dsout output of r.flow). All of this was
> solved with legends for SG3d (you can chose discrete or continuous, you
> can place the legend with mouse both for discrete and continous and you
> can define the values for which the color is shown for both discrete and
> continuous and you could even chose the size of the font - the
> capabilities are described on the web), however nobody seems to be
> interested to include it into grass5 although that means that for many
> maps you cannot really display your data in a meaningful way.
I fully agree - the legends problem is severe.
For the others: The SG3d code is here:
grass_exp/src.contrib/GMSL/SG3d
grass_exp/src.contrib/GMSL/sg4d
-> experimental tree, not stable branch.
Especially sg4d is great software, see
http://www2.gis.uiuc.edu:2280/modviz/viz/vol1.html
> I am writing
> this long mail to urge you to put the legends issue on top of the agenda
> for the developers meeting, it shoudl take an experienced programer few
> days to implement it when using Bill's SG3d code.
[...]
> People just now expect to get the legend along
> with their map as ArcView does it and parctiacally all of on-line mapping
> services do too. So not being able to get the legend when looking at your
> map is a big turn off for people who may want to use GRASS.
... very true...
> For spearfish when you do the legend for geology you get either huge
> squares and letters, or if you run it as if it was a continuous map so
> that you can get the legend smaller then you do not get the cat labels for
> all colors. The easy workaround how to get the legend along the map is to
> stretch the window to make space for it (rather than using frames which is
> how you are supposed to do it, but it is really inpractical - it is used
> in d.display).
[...]
> Helena
All open for comments...
Markus
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5
On Fri, Nov 02, 2001 at 11:53:41AM -0500, Helena Mitasova wrote:
just a note - the reason I cannot get cat labels for all categories when I
use the workaround with the smooth gradient option is that the smooth
option always writes only 5 values/labels, so I get nice legend for
soils.ph which has 5 categories and I am out of luck with geology which
has about 10.
Hope here:
I recently implemented a "steps" parameter (get from CVS):
d.leg.thin help
Usage:
d.leg.thin [-nsm] map=name [color=name] [lines=value] [thin=value]
[labelnum=value]
[...]
labelnum Number of labels for legend description
options: 2-100
default: 5
The new labelnum allows to overcome this problem at least partly.
Maybe someone want to implement a mechanism to define the minimum
value to be displayed?
I am working with spearfish data set and I guess I must be quite
frustrated given the number of misspellings I managed to make when writing
to Markus. Sorry for that,
Sorry Helena, but I was hesitating to modify your text.
My apologize (I was just to lasy to write it again with own words).
Ciao
Markus