[..]
Apologies for erasing the above text.
Moritz Lennert wrote:
Please don't do this ! I find the fact that GRASS does not provide a
default projection system, but forces the user to think about
projection from the start, one of its strengths, both for work and for
teaching.
Helena Mitasova wrote:
Moritz - that won't be lost, I agree that it is a very important
GRASS feature.
The concept of location/mapset remains, but they will have to start
thinking about it once they start GRASS,
not before they even open it. In my class when we go through starting
GRASS step-by-step, the students are fine (although there is
still the problem that some unzip apps create an extra directory when
unzipping the files automatically which people don't notice),
but when in workshops or when on their own I have seen users
struggling to get GRASS even open because they give it wrong path
(e.g., to the location rather than gis directory) or they don't know
what to select and whether to create a new location or new mapset.
I am not sure how well the approach of selecting the location from
within GRASS will work, but I am willing to give it a try.
I have just written the chapter on creating new locations and
reprojecting data in GRASS for the 4th edition of grassbook so I will
have to rewrite it,
but this also made me aware of how cumbersome the startup page may be
for new users.
On the other hand I just discovered how easy and elegant is to start
GRASS from command line and skip the startup screen altogether.
I am not pushing strongly for this change - we can try this first in
our lab on real-world projects with newcomers and experienced users
and see whether it would help or whether it would cause even more
confusion,
Dear Helena,
on Monday morning I tried to introduce, yet another time, a new-comer
in GRASS GIS (PhD student, determined to work with GRASS GIS).
The very basics of GRASS, in my very humble opinion are: Location,
Mapsets (PERMANENT vs User-Created ones), Extent and Resolution. As
usual, the Location versus the Mapset concept, is hard to grasp in the
beginning. As well, and perhaps more important, the Computational
Region, is difficult to understand.
My approach, so far, is to explain what makes a Location unique (its
reference system). And repeat it, again. And explain why this very
concept sets straight from the start GRASS to stand as a clear winner
among other GISes. Yes, there is a trade-off, but it's worth it.
Then, "what are Mapsets?" and how they (=multiple) are useful. Here, I
stress the identity of the PERMANENT Mapset. What purpose does it
serve?
Finally, the Region. Extent and Resolution.
For all of this, we need to provide clear and very, very, simple
visuals. Let the power visual communication do its work. One for each
concept.
Last March, I held a workshop in India. I feel that it went really
well, since all of the students followed up to the point where we did a
simple classification using QuickBird imagery. I did invest some time
in theory, before going down to the keyboard and the mouse. But,
through 2-3 repetitions for each of the basic concepts, the lab went
really well (most of the people had Zero idea about GIS and Remote
Sensing, coming from different disciplines).
I do feel that GRASS needs some simplifications. Especially in its GUI
section. IMHO, it needs some retouch. As it is, it is a bit "crowdy"
and noisy for a new-comer. Yes, I know, I don't have the magic wand to
retouch it, nor I have all of the required skills, the time or the money
to do it currently. But I feel it is a necessity if the project wants
to make it easier for newcomers to jump in and let part of average
GIS/RS users embrace it.
Just sharing a few thoughts. Warmest regards, Nikos
_______________________________________________
grass-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/grass-dev
