kurt@wrrc.ecn.purdue.edu (Kurt Buehler) (12/28/90)
(warning - a fairly long request for comments/opinions follows)
As part of a contract that we are currently working on, we were asked to make
a prediction about the software capabilities of geographical information
systems (GIS) on low-cost workstations in the year 2000. Realizing the power
of the vast expertise that folks that read these newsgroups, and our relative
ignorance of the subject, we decided to pose the question to you, the
developers and users of advanced hardware and software, and the insightful
problem-solvers of the world.
We realize that few people have really good idea about what a "low-cost
workstation" will be in the year 2000, but we are still interested in getting
comments about the software capabilities. We would be especially interested
in hearing from those of you working in the areas of image processing, pattern
recognition, and even computer vision, because these areas provide the
foundations for GIS.
You are all invited to give us your opinions concerning what the low-cost
workstation will look like from a hardware standpoint. But in case you don't
know (or would rather not speculate) you may consider that we are speaking of
a highly parallel machine (100's of processors (MIMD) or 1000's (SIMD)), with
gigabytes of disk, many megabytes (100+ ??) of memory per processor (assuming
a SIMD or MIMD distributed memory architecture) or many, many megabytes
(assuming a MIMD shared memory architecture), possibly other (highly-
specialized) chips for various tasks, 3-D graphics hardware, etc., etc..
You are all invited to predict what geographical information system software
will be possible on such machines. At least the following categories can be
considered with reference to raster and vector-based GIS:
1. Image processing
2. Pattern recognition
3. Environmental and engineering modeling
4. Map data input
5. Map data translation
6. User interface
For each software capability, it would be helpful for us if you could address
the following issues:
a. What hardware makes this software possible ?
b. What evidence suggests this software ?
c. Who would benefit from this software ?
Please consider this to be a serious request and that give your honest
opinions (hopefully based on some expertise in one or more of the areas
involved).
It is our desire *NOT* to begin a long dicussion in various newsgroups and
generally clog the net!!! So, *PLEASE* e-mail your responses directly to me at
the address below. We will (of course) summarize by preparing a synthesis of
the opinions given and make it available via ftp or by US Mail. Please
indicate in your response that you would like to know where to obtain the
summary and we will mail you the information at a later date.
Thanks for your time and interest!!!
--
Kurt Buehler
Indiana Water Resources Research Center,
School of Civil Engineering, Purdue University, W. Lafayette IN 47907
Internet: kurt@ecn.purdue.edu uucp: ...!{decwrl,gatech,ucbvax}!purdue!kurt