Welcome to swarm-gis!

[Roger M. Burkhart]

Welcome to the swarm-gis mailing list!  Any messages sent to
address@hidden will automatically be sent to all list subscribers.

This list has been set up as a result of the breakout session on Swarm
and GIS at the '97 Swarm Fest that occurred on February 17.  The following
people who were at the breakout have automatically been added to the list:

  Roger Burkhart     address@hidden
  Bob Lusch          address@hidden
  Brian Haugh        address@hidden
  Miranda Warburton  address@hidden
  Tom Vest           address@hidden
  Bob Bell           address@hidden
  Fernando Escobar   address@hidden

In addition, the following people who also expressed interest have also
been added.

  Glen Ropella       address@hidden
  Chris Langton      address@hidden
  Peter Hraber       address@hidden
  Paul Box           address@hidden
  Brad Parks         address@hidden

swarm-gis is a majordomo mailing list just like all the other swarm
lists.  If you ever want to be removed from the mailing list or change
anything just use instructions for all Swarm mailing lists on the Swarm
web page (you don't have to be at your subscribed address to change
things).

Now for a brief summary of the initial meeting, and suggestions for where
we can go from here:

1. Lots of interest was expressed in links between Swarm and GIS.
   Many of the SwarmFest attendees are involved in models of ecology or
   archaeology or other applications that deal with real landscapes.

2. Three levels of potential links from Swarm to GIS were discussed:

  - Support for transfer of data to/from GIS systems and analysis tools (such
    as the "fragstats" package being used at UCLA).  The GRASS/Swarm link
    developed as an experiment by Jim Westervelt and Doug Briggs at Army
    CERL and University of Illinois is one example that reads and writes
    files in GRASS format.  Other available software packages have examples
    of other load/unload from other common GIS formats that could possibly
    be built into Swarm.  After the meeting, Peter Hraber also suggested
    load/unload to the standard image format of the Khoros system.

  - A better Swarm space library that could handle multiple layers of
    georeferenced spatial data and agents simulated within them.

  - An even more aggressive approach to spatial libraries in Swarm, in the
    form of a reasonably complete set of spatial object types along with
    typical spatial operations on them.  The emphasis would be not on the
    analysis and visualization of spatial data for which GIS and other
    tools already provide excellent support, but for supporting specific
    operations needed within a running simulation.

3. Methods of organizing a user community to help coordinate development
   and support of any of these capabilities were discussed.  This could
   range from informal coordination of individual projects to
   network-based collaboration of various sites to fully funded programs
   that would develop and support public libraries.

4. GRASS and its user community was discussed as an example of a user
   community that had relied on use of freely available tools.  Many
   of these projects involve modeling and might be a ready community for
   Swarm-based tools, typically in combination with their existing GIS
   systems.  This community is large enough that it might serve as a
   substantial source of additional support for wider use of Swarm.

5. Possible followup actions were discussed.  These included:

 - Obtain the source for the Swarm/GRASS link developed at University of
   Illinois, and evaluate it with the current release of Swarm.  Bob Bell
   of UCLA offered to try it out, and afterwards Paul Box of University of
   Florida also offered to help.

   Update: Jim Westervelt has now provided us with this source and we'll
   be making it available to Bob and Paul for an initial look.  We'll let
   the list know about any results.  The code is unrestricted so can also
   be made available to anyone else, but we'll try to get it converted to
   the current Swarm version first.

 - Consider organizing a Swarm and GIS workshop that had been discussed a
   year ago with Brad Parks of NOAA, Jim Westervelt, and others, following
   the conference on Environmental Modeling and GIS held in Santa Fe in
   January 1996.  Chris Langton has continued to discuss this possibility
   with Brad Parks.

 - Consider organizing a pool of potential participants, funding sources,
   and/or supporting institutions who could pool and coordinate resources
   to develop various Swarm/GIS links.  The first projects would probably
   consist of load/unload capabilities from various GIS formats along with
   new Swarm space libraries.  Further steps could go on to build a complete
   library of GIS spatial object and operations to use inside Swarm
   simulations.

   Bob Lusch of University of Oklahoma offered to look at possible business
   models for organizing some kind of funding consortium that could support
   such work on an ongoing basis.  Possible sources discussed include
   private foundations (including those that manage natural resources),
   companies that use GIS, public agencies (such as the National Forest or
   Park Services for which work is being done using Swarm), and other potential
   government users such as the Department of Defense.

Feel free to send any ideas, suggestions, or concrete developments that
could help make links from Swarm to GIS happen, to beter serve the needs
of current or future Swarm users.  This mailing list can be as inactive or
active as you want it to be (from announcements only to lively
discussion); we'll also be keeping Hypermail archives like the other Swarm
lists.  However quickly things get started, I look forward to building on
current Swarm foundations to fill the needs of applications that deal with
real space on earth!

Roger Burkhart