The Future of GIS
…if I really knew, would I still be here?
1
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Popular Mechanics, 1954???
2
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Lessons:
1. There are perils in predicting the future!
2. Never believe what you find on the Internet!
Picture submitted to an image modification competition in
2004, taken from an original photo found on U.S. Navy web
site of a submarine maneuvering room console mock-up at
the Smithsonian Institute in 2000
http://www.snopes.com/inboxer/hoaxes/computer.asp
3
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Uses of GIS: no change
The primary three:
• manage data
• analyze data
• communicate information*
BUT
• relative importance shifting
• implementation technology changing
*information=data which serves a purpose
4
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Changing Emphases:
From Data to Analysis
Spatial Analysis
Spatial 5% Analysis
10-15% Attribute Tagging
Attribute Tagging
75% Data Conversion
Data Conversion:
Past Present/Future
5
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Changing Emphases
From Description to Simulation & Modeling
Visual simulation &
virtual reality:
Picture worth a
thousand words: real time display of
how is, and how might be
maps & diagrams of -forest fire
how is, or how was -freeway traffic flow
Web portals serve static Web portals serve continuous
data sets sensor-derived data
Past Futureon logical
Symbolic models: based
Iconic models: scaled down
representations of the real thing relationships in mathematical or
statistical form
6
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
9:15 am 4:30 pm 10:15 pm
Population density (green is high) at different times during the
day tracked by cell phone data. Rome, Italy, July 10, 2006.
(note: cell phone location is constantly tracked by the network
to enable calls to be received.)
Applications: real time traffic information, transportation
planning, taxi-cab location, retail store location, etc., etc..
Source: The Economist, March 10-16, 2007 p. 20.
7
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Changing Emphases
from 2-D description to 4-D interaction
Past
• 2-D flat map displays
– User as observer
Future
• Effective 3-D visualization
– Via the merger of CAD and GIS?
– What is the data model?
• 4-D incorporation of time: “The time has come for time.”
– Via agent-based modeling / cellular automata? Or how?
• agents (e.g. vehicles, fires or people) interacting over time in a raster (cell)-based
environment according to established rules
• 5, 6 and 7-D incorporation of touch (pressure, texture, temperature), sound and
smell into modeling/simulation environment
– Aldous Huxley‘s Brave New World ―feelies‖ become reality?
• User as participant
– Users (researchers, professionals, the public) interact with the model
– Participatory GIS: the public as the planner
8
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Changing Emphases:
Out of this world
Past
– GIS applied to Planet Earth
Future
– GIS as a methodology for the analysis of spheres
• Other planets—Mars, Jupiter,
• The human brain
– One earth but many brains
– & visa-a-versa: does the brain use ―maps‖ for organization?
9
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Astrography
• A map of the nearby
Universe --- some
9300 galaxies each
represented by a dot.
The survey shows two
slices across the
sky plotted against the
redshift (or indirectly,
distance) of each
galaxy. Several large
structures (walls) can
be clearly seen as well
as large voids where
galaxies are relatively
absent.
10
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Astrography
• Surveys like the one above map the local
Universe by measuring the position in 3-
D space of galaxies up to a few hundred
Mpc away. The observable Universe has
a diameter of order 10 Gpc. To find and
observe galaxies at large distances (and
therefore in the early Universe) requires
large or special telescopes. The
formation of galaxies is not well
understood, although the field is
beginning to open up. The figure above
shows the start of the art at present, in
which galaxies have formed in a
computer simulation of the gravitational
collapse of the initial perturbations
which came from the big bang. The
simulations are difficult to compute
because the effects of gas dynamics and
star formation in the forming galaxies is
very poorly understood.
11
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Brain Mapping
12
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Technological Trends Underlying the Transition
Defense Conversion (and other) spin-offs Information Technology Evolution
• Location via GPS • Interoperability: easier sharing of data
– millimeter accuracy between users, and among vendor products
– available in every cellphone for E- – Metadata
911 – Spatial Data Transfer Standards
• super high capacity mass storage – OpenGIS
– pettabyte and more systems – Mash-ups
• high resolution ( many and varied
– Few users -> many
– Standalone -> web
– Fixed -> mobile 16
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Consequences: for GI Specialists
(the fifth “S” for GIS?)
geospatial information scientists, specialists (or students)
– appreciative of the broad ranging, integrative role geospatial data
can play
• location is the most common common denominator for all data
– highly knowledgeable with respect to the unique challenges of
geospatial data
• The challenges posed by a spherical earth (datums, projections,etc.)
• The complexity of spatial data representation in 2-D and 3-D
• The challenges of spatial statistics and modeling
– fully conversant with mainstream information technology
• and how to store and program spatial objects
– sufficient expertise in an application area (geology, local
government, marketing, etc.) to make a contribution.
17
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Consequences for GIS
Data is still at the heart
• Dominant IT (Information technology) issues:
– Hardware in the 1970s and 1980s
– Software in the 1980s and 1990s
– Data in the 2000s
• Not an issues of acquiring data, but of managing it
• Will we be short of it, or drowning in it?
• Will its availability be
– Plentiful and cheap
– in infinite detail, if you can afford it
– Severely curtailed by legal controls
to ensure personal privacy
18
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Consequences for GIS:
is a Tragedy of the Commons in the making?
• Invasions of privacy through detailed data collection and its
pervasive distribution produces a backlash of demand for privacy
– No call, no spam, no appraisal photos, no red light cameras (now
reversed): are they the beginning?
– Could geotagging with RFID devices become reality
• From pets to people
• for sex offenders, service personnel, employees, evacuees, everybody?
• The expense of data production, but the ease of re-production and
distribution, reduces the value of data to zero and chokes off its
availability
– Is public domain data the information age equivalent of the agricultural
commons?
What are the appropriate public policy
responses?
19
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Geotagging
Below: Which of
the three locations
should be
associated with the
resulting photo?
Above: Map showing photographer's view
( 59°19′39″N 18°04′21″E59.3275,
18.0725) of two buildings at a distance of
270 meters ( 59°19′42″N
18°04′38″E59.3284, 18.0772) and 1200
meters ( 59°19′45″N 18°05′35″E59.3291,
18.0931).
20
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Or is this the future?
A Self Generating System
Better data
More investment in More GIS use
Data/GIS
Better decisions
21
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Or this?
Microdrone $21,367
Base Station $19,424
Video Transmitter $1,545
Video Receiver $1,000
Daylight Video $1,545
Lowlight Video $3,100
GPS Hold $1,934
Complete Package $59,681
August 2007
Source: http://www.microdrones.com/
22
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Conclusion
• Our only model for the future is the past
• think back as many years as you are looking
forward
– change has been both revolutionary.. and glacial
• don‘t forget that the pace of change is
accelerating
– maybe just gearing-up for the information age,
not reaching its climax
• but remember,we have to get there from here
– can you envision a path?
• if we knew the future, we wouldn‘t be here!
23
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
GIS Futures
Some Detail on earlier concepts
• What GIS Technology Could Deliver
– Efficiency and Effectiveness
– Targeted Communication
– data as an institutional asset
• Operating Environments in which Delivered
– business and governments
– computing technology
– information architectures
– interoperability
24
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
What GIS Technology Could Deliver:
Efficiency and Effectiveness
• Cost efficient, quality service to customers: the key to
future business success
• mapping to manage potentially transforms organizations
• geography is the key to cost efficiency for pizza delivery or cellular
radio towers
• communications with citizens: the key to future public
sector success
• map based information is the key to intuitive information delivery
– travel directions (www.mapquest.com)
– natural or social environmental degradation (http://www.epa.gov/tri/)
– land ownership (www.dallascad.org)
– General public data (www.accu-source.com, www.publicdata.com,
www.openrecords.org)
25
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
What GIS Technology Could Deliver:
Targeted Communication
Displaying data differently for today’s target
constituency: *CC: Chamber of Commerce
DRT: Daughters of the Republic of Texas
• governor DAR: Daughters of the American Revolution
• recalcitrant group of legislators VFW: Veterans of Foreign Wars
CCNA: Canyon Creek Neighborhood Association
• activist (pissed-off) citizen posse
• CC, DRT, DAR, VFW, CCNA Tuesday luncheon*
• general public
• 6th grade class
Do-It-Yourself extraction from Societal Databases
• large, networked databases accessible to public at low/no cost
• free browser software (e.g ArcExplorer)
Content tailored for current location
• The mobile, handheld, interactive GI services revolution
26
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
What GIS Technology Could Deliver:
Data as an Institutional Asset
Data holdings, managed by GIS, as an asset on
private and public corporation balance sheets.
• no company does this today, yet
• billions spent on data acquisition and development
In the information age, information is an asset no
different from buildings, unsold inventory, and
machinery.
In 2000s, companies delivering information content
become dominant over hardware (1980s) and
software (1990s) companies.
27
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Operating Environments:
Business & Governments
• customers and citizens take charge
– requirements for service defined by customer/citizen not the provider
– demand more in shorter time at lower cost than you ever intended to deliver
– lower taxes/prices and more service
– Henry‘s ―any color you want as long as its black‖ no longer cuts it
• competition is relentless
– more people wanting to do what you do
– private sector assumes (or re-assumes) many gov. tasks.
• change is constant
– Government evolves from driver to consumer of technology
– Commercial Off-The-Shelf (COTS) software rather than custom designed
• decentralization to the individual
– Much new technology is not new: its been around for a while, so what‘s new?
• Computers, video recorders, fax, wireless
– Dramatic price drops make it as available to individuals as to organizations
28
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Operating Environments:
Computing Technology: Predictions from Spring 96
• 2,000AD desktop machine
– 1,000 MIPS/MHertz
– 8 Gigabyte RAM (permits data in memory)
– 100+ mb/sec comm. line (to the home?)
• Windows95/NT merge challenges UNIX
• Java /Netscape challenges Microsoft
(prediction as of Spring 96)
29
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Operating Environments:
Computing Technology: Predictions from Fall 99
• Desktop loses its dominance
– Variety of computing appliances: palmtops on up
• Computers act rather than just process
– Old model: human enters data, computer processes it, human receives and acts
– New model: data from sensors & transducers, computer processes, computer
acts to get job done
• Data and computation become real time
– Old model: processing archived data to guide future action
– New Model: processing current data to control current action
• Information and its processing at any and every place
– Mobile and wireless dominate over fixed and wired
– Info access no matter where I am, where data resides, what its format is
Concepts derived from Tennenhouse, Director, DARPA, May 1998
30
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Operating Environments: Information Architecture
(more 1999 predictions which have come to pass)
• server and client clearly separated
– GIS vendors specialize in client and/or applications server
– mainstream database vendors provide data server(s)
(traditionally, GIS vendors used proprietary, not mainstream, databases)
• client systems primarily comprise browsers which receive data combined
with software as ―applets‖ or objects from the applications server
– e.g. receive data and zoom/view software
– potentially simplifies software management
– big software systems are decomposed into components (―objects‖), which are then
re-assembled by user (or developer) as desired.
• mainstream data base vendors meet GIS/AM/FM needs for:
– long transaction processing
• current DBMS systems assume immediate transaction processing, but
• a drawing or map takes time to be modified
– replication of data files
• considered a no-no in current DBMS, but
• two people need to work on the same map area
31
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Operating Environments:
Interoperability
• Spatial Data Interchange Formats • Between GIS Systems and other
– VPF (Vector Product Format): US applications
Military for map products; directly – COM/OLE (Common Object Model/Object Linking
useable by applications and Embedding): originated by Microsoft
– CORBA (Common Object Request Broker
– DXF (Data Exchange Format): originated Architecture)
by AutoCAD Two competing standards for object-based technology.
– SDTS (Spatial Data Transfer Standard): • Between GIS Systems Themselves (for data)
currently required for Federal Agencies – SQL3 --SQL extended to support spatial (and other
Except for VPF, involve translation (from multimedia) data queries
internal private to external public – ISO TC211--International Standards Organization
format), therefore inefficient. Tech. Committee on spatial data
– ANSI X3L1 (American National Standards Institute,
• Between Standard Data Base Systems GIS Committee)--US view for ISO
– SQL (Structured Query Language) – OGC (Open GIS Consortium)--Open Geodata
Interoperability Specification (OGIS)
• standardized (supposedly) user-oriented
(supposedly) data request • Metadata--data describing data
– ODBC (Open Database Connectivity) – datum and projection, accuracy and lineage, etc.
– FGDC (Federal Geographic Data Committee)
• standardized (by Microsoft) programming Metadata standard
(‗call‘) level interface to databases
32
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals
Next Generation Satellites
• NGA (National Geospatial-Intelligence Agency) signed NextView contracts for
development of next generation of commercial satellites, with DOD being given
priority access in times of need
• Digitalglobe contract in fall 2003, focused on
– Higher resolution
– Delivery time to customer
• 3 hours now (Iraq war)
• Future: 90 minutes standard, 20 minutes ―rush jobs‖
– WorldView-1 launched September 2007
• .5 m panchromatic
• 1.7 days revisit
– WorldView-2 launch 2008
• .5 panchromatic and 1.8 multispectral (4)
• 1 day revisit
– Supplier for Google Earth
• Orbimage contract in fall 2004
– OrbView 5 satellite to launch early 2007 (now early 2008)
– 0.41 m panchromatic, 1.64 m multispectral (4 bands)
– 3 m. position accuracy
– 3 day revisit
– Downlink imagery real time to ground station
Note: the award of this contract to Orbimage resulted in their acquisition of Space Imaging
(which failed to get the contract) in January 2006 and renaming of the combined entities as
GeoEye. OrbView 5 now called GeoEye-1
http://www.geoeye.com/corporate/constellation.htm
33
11/19/2007 Dr. Stuart Murchison, UTDallas GISC 6381 GIS Fundamentals