[Freegis-list] Re: GIS grant to help map cities worldwide

[Michael C Dietze]

I am relatively new to this list (been reading for a couple months and
this is my first post) but I have been continually suprised by the amount
of energy spent defending the _concept_ of free GIS, the current thread on
ESRI's investment (charity?) in the third world being just one example.
I would expect that people interested enough to join this list would be
sufficiently on board with the idea that such discussion would be
preaching to the choir.  This is not to say that we shouldn't extol the
virtues of Free (big F), free (little f), and assorted shades of
public liscence GIS, but that such a reaction has lead discussion, and
this discussion in particular, away from what I think would be the more
productive topics.

For example, Mr. Longhorn's contranst between OpenOffice and GRASS really
resonated with me and points to what I find to be one of the major
challenges of Free GIS, which is accessability.  I, like Adrian, am and
ecologist, and my university trains ~100 students a year in GIS as part of
a "masters of environmental management" program which has an incredible
job placement rate, including a substantial chunk into non-profit
conservation organizations.  They train exclusively in ESRI products.
How the non-profits these students go to work for are expected to afford
ARC is beyond me.  Even I, in my conceptual support of Free GIS do most of
my GIS work on Arc even though I do the vast majority of the rest of my
research from a Linux box...I was trained in Arc and haven't had the time
and energy to tackle what appears to be a very steep learning curve for
GRASS.  The question I have is how do we reach people like this, people
like me, people like Mr. Longhorn?  Telling them to hire a consultant
isn't an answer that resonates with me.

My vision is to see a Free GIS that is:

1) detectable - i.e. the various options and tools should be easy to find
by someone looking for them.  A new user shouldn't have to invest huge
amounts of research time into finding what they need.  While that time is
justified for a long-term large-scale project, it's not for the beginner
and there's no reason to make it hard for the advanced user.

2) intuitive - anyone who is familiar with using computer GUIs, and in
particular anyone trained in another GIS, should be able to make a basic
map without having to read the manual, without having to memorize hundreds
to thousands of commands, and without writing any code.

3) cross-platform

4) migratable - no one wants a piece of software that requires them to
throw out all their old work/data.

That is the bare minimum to get people into using the code...there are
many other things I would love to see as well (e.g. is there even anyone
out there THINKING about what it would take to implement Bayesian
Heirachical models in a GIS package?).  It is something where anyone can
do the quick and easy projects, but where the code is there for people to
extend and customize applications.  I strongly believe that we need the
first type of user to be able to build the support base for the latter
user to emerge in enough numbers.

In a nutshell, I think we need to be thinking about what we need to do to
COMPETE with ESRI and the like, rather than just preaching the virtues of
FreeGIS and expecting that to be sufficient to draw users.

Mike Dietze

p.s. Adrian, I think it's more than a little strong to say that "even the
best spatial ecologists don't understand <long list of stuff>".  I think
many (thought definately not most) of us, at least here at Duke, have been
thinking about these issues but that currently error propagation is damn
near impossible in Arc, and I suspect it's the same for other GIS
packages.  As for the use of traditional spatial stats like variograms and
krieging and the like, I think it's just part of the larger iceberg of how
ecologists (and probably most researchers in general) misuse and abuse
their stats, and the need to move beyond inappropriate frequentist
approaches.  You might want to take a look at our lab's paper in the new
(Jan '03) issue of Ecology.


____________________________________________________________

mcd7 at duke.edu	                  	Program in Ecology
http://www.duke.edu/~mcd7/lab		Dept. of Biology
lab: (919) 660-7403	          	Duke University
home: (919) 401-6594			Durham, NC 27708