RE: Re: [GRASS5] National Center for Open Source Policy and Research

GRASS GIS GRASS GIS developer
,
1

There is some karma responding to this statement, on this particular
list... :-]

I recall when I supported GIS software vendors for Sun Microsystems in
the early 90's, and at one point knew root passwords for machines
containing source code for ESRI, Autodesk, Oracle, Sybase, Informix,
and Bentley. I also participated in the winning Sun-Hughes bid for
EOSDIS.

Jack Dandgermmond was making the rounds in DC, railing against GRASS
because "my tax dollars are funding my competition", and with the
Forest Service (601?) contract, ESRI surpassed Intergraph in GIS market
share. The GRASS user's group, a hardscrabble collection of
individuals, to whom I used to sneak compilers and helped get a machine
or two, had the foresight to change their charter from GRASS advocacy
to interoperability. NSDI, Gore's 'Digital Earth' speech, and the rest
of OpenGIS history are part of the lore.

Whether or not OpenGIS, and national GIS policy in general, diffuse
into irrelevance by virtue of trying to standardize all of enterprise
IT (anyone remember GILS? etc) remains to be seen, but it is difficult
to say that the OpenGIS WMS and WFS protocols are not fully embraced by
the federal government, and international institutions in general.

Being here near the center of the current portal *Earth wars, I think
the real challenge for the government is to find relevance when folks
like Google, Yahoo and Microsoft are subsidizing their GIS offerings to
the tune of tens of millions of dollars, just so people will be
addicted to them.

As for analytical capability, I think the real challenge will be a
Darwinian struggle as to what is the most sustainable platform for
service delivery. If you can make a case for running your shop better
with GRASS, can find suitable talent, and get it supported, and can
genreally do mission critical stuff, why not.

But ESRI is ESRI and Intergraph is Intergraph, software, professional
services, retired generals and all. But what can be accomplished on
open source is not to be discounted. In time of strapped cash at all
levels of government, the cost of Oracle+SDE+ArcIMS+per-seat licenses
versus PostGIS+MMS is not trivial. For example, GlobeXplorer is
*completely* built with open source and custom glue, running databases
on >20 CPUs. Users don't care, so long as it is reliable and has the
content they need.

I'll just note that there is fairly high-level support for the general
concept of creating a "guest" co-location facility directly at EROS
datacenter, whereby people could run whatever (GRASS, ERDAS, etc)
directly against huge datasets on a managed subnet, and indeed realize
the first "A" in DAAC. (distributed ACTIVE archive center).

And at this point I'll also make a plug for my hobby project:
http://iserver1.ciat.cgiar.org/mms/CGIAR_prop.html .

Chris

There is no U.S. Federal policy on GIS. At best, GIS policy exists on

an

agency-by-agency basis (sometimes at the Dept. level, e.g. maybe at
Interior). I can say unequivocally thet there is no Dept. of Commerce

or

NOAA or NWS policy on GIS.

Regards,
Tom

2

Chris,

Thanks for the detailed appraisal of the current 'state' of government GIS.
What I read from this for GRASS users is that, if those using the software
in federal settings find that it is a better solution for their needs and
can make a case to that effect, there may be increasing opportunities to use
open source solutions. GRASS may or may not be the 'right' solution. But it
is good if it can be considered among potential ones (especially given its
government past). I don't know how the new NCOSPR figures into to this. I
think making GRASS a resource supported by the center is something that will
need to come from federal users.

Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton

From: <cgnicholas@alamedanet.net>
Date: Wed, 26 Oct 2005 08:12:11 -0700
To: Thomas Adams <Thomas.Adams@noaa.gov>
Cc: <roger@spinn.net>, <grass5@grass.itc.it>
Subject: RE: Re: [GRASS5] National Center for Open Source Policy and Research
Resent-From: <root@grass.itc.it>
Resent-To: <grass5@grass.itc.it>
Resent-Date: Fri, 28 Oct 2005 14:17:34 +0200

There is some karma responding to this statement, on this particular
list... :-]

I recall when I supported GIS software vendors for Sun Microsystems in
the early 90's, and at one point knew root passwords for machines
containing source code for ESRI, Autodesk, Oracle, Sybase, Informix,
and Bentley. I also participated in the winning Sun-Hughes bid for
EOSDIS.

Jack Dandgermmond was making the rounds in DC, railing against GRASS
because "my tax dollars are funding my competition", and with the
Forest Service (601?) contract, ESRI surpassed Intergraph in GIS market
share. The GRASS user's group, a hardscrabble collection of
individuals, to whom I used to sneak compilers and helped get a machine
or two, had the foresight to change their charter from GRASS advocacy
to interoperability. NSDI, Gore's 'Digital Earth' speech, and the rest
of OpenGIS history are part of the lore.

Whether or not OpenGIS, and national GIS policy in general, diffuse
into irrelevance by virtue of trying to standardize all of enterprise
IT (anyone remember GILS? etc) remains to be seen, but it is difficult
to say that the OpenGIS WMS and WFS protocols are not fully embraced by
the federal government, and international institutions in general.

Being here near the center of the current portal *Earth wars, I think
the real challenge for the government is to find relevance when folks
like Google, Yahoo and Microsoft are subsidizing their GIS offerings to
the tune of tens of millions of dollars, just so people will be
addicted to them.

As for analytical capability, I think the real challenge will be a
Darwinian struggle as to what is the most sustainable platform for
service delivery. If you can make a case for running your shop better
with GRASS, can find suitable talent, and get it supported, and can
genreally do mission critical stuff, why not.

But ESRI is ESRI and Intergraph is Intergraph, software, professional
services, retired generals and all. But what can be accomplished on
open source is not to be discounted. In time of strapped cash at all
levels of government, the cost of Oracle+SDE+ArcIMS+per-seat licenses
versus PostGIS+MMS is not trivial. For example, GlobeXplorer is
*completely* built with open source and custom glue, running databases
on >20 CPUs. Users don't care, so long as it is reliable and has the
content they need.

I'll just note that there is fairly high-level support for the general
concept of creating a "guest" co-location facility directly at EROS
datacenter, whereby people could run whatever (GRASS, ERDAS, etc)
directly against huge datasets on a managed subnet, and indeed realize
the first "A" in DAAC. (distributed ACTIVE archive center).

And at this point I'll also make a plug for my hobby project:
http://iserver1.ciat.cgiar.org/mms/CGIAR_prop.html .

Chris

There is no U.S. Federal policy on GIS. At best, GIS policy exists on

an

agency-by-agency basis (sometimes at the Dept. level, e.g. maybe at
Interior). I can say unequivocally thet there is no Dept. of Commerce

or

NOAA or NWS policy on GIS.

Regards,
Tom