>> From the Library of
Congress in Washington, D.C.
>> I welcome everyone.
As a cartographer and a
GIS specialist I am pleased
to participate in this
unique program.
In the final segment I will present
two speakers whose research explores
the notion of cartography in the
future and the pursuit of geo-design
through architecture, cartographic
representation, and landscapes.
And following this there will
be a question and answer period
and in the final wrap-up Ralph
Aronberg will introduce the final
speaker of the day.
Our first guest is assistant
professor of landscape architecture
at the Harvard University Graduate
School of Design, Jill Desimini.
Desimini? Prior to joining the
full-time faculty Desimini was a
senior associate at Stoss
Landscape Urbanization in Boston.
She holds Masters of Landscape
Architecture and Master
of Architecture Degrees from
the University of Pennsylvania,
and a Bachelor of Arts in Urban
Studies from Brown University.
Her research focuses on cartographic
techniques, urban wilds,
and landscape strategies
for abandoned urban lands.
Please welcome Ms. Desimini.
[ Applause ]
>> Jill Desimini: It's
bright up here [inaudible].
Thank you.
I'm delighted to be here.
I consider myself a kind
of very lucky, very happy,
very interested interloper because
I neither make maps nor do I
study maps.
But in some ways I find
cartography and maps to be central
to almost everything that I do.
So I think you could consider me
perhaps a map user and a map abuser.
You can weigh in on what
you think about on that.
But for me cartography is an
operational tool to inform
and further the representational
capacities of my discipline
of landscape architecture and I
come from a very generalist field.
So please enjoy my
generalist tendencies.
This talk springs from an
[inaudible] exhibition I curated
in 2002, which has an
afterlife as a book project
with Princeton Architectural Press.
It will be an image-heavy
volume with 350 images due
to be released in fall of 2015.
It's going to be a busy summer but
please keep your eye out for it.
It includes the work of 50
designers and over 85 cartographers
across a range of mapping
and design practices.
The project has at its premise the
idea that conflating architecture
and cartography is a
productive practice.
It is critical of current trends
and representing landscape
that despite the widespread
availability
of data there's a looseness
to the drawings we create.
It is almost as if the readiness of
information has produced a laziness,
an obscuring of both our critical
eye and our critical hand.
For example, my students rarely
question the representational
choices of Google mapping
products or even recognize
that representational
choices are being made.
And while they are critical
of the aesthetics of GIS
as designers I think
that the aesthetics
of GIS is not often talked
about in my field at least.
So you could say that the
work is a critical response
to what Mark Monmonier called a
stylistic homogeneity yesterday.
Did it go forward?
In this exhibit and book project I
argue that the ascendants of mapping
and data visualization in design
culture, this kind of paradox
of globalized practices and
customized content that we heard
about yesterday, has changed the
architects, landscape architects,
and urban designers
communicate ideas about buildings
and landscapes, often privileging
abstract forces and flows
over the material conditions
of the site.
The collection reimagines
the protective potential
of cartographic practice
that afford greater proximity
to the ground itself.
The approaches presented here
seek to reconcile the precision
and instrumentality of the
plan with the geographic
and territorial scope of the map.
Cartographic grounds investigates
a range of surface conditions
and representational tools,
it follows the contour line
from its origins in early European
bathymetry to its terrestrial rival
and use in 18th Century Parisian
parks, to its projective qualities
in the contemporary work of
James Corner Field Operation,
designers of the highline
in New York.
And then on a completely other
end the stratigraphic column is
celebrated as a means to create
vibrantly colored geological maps,
and by extension to
depict subservice
and thick surface conditions
in design.
The work of Prussian geographer
Alexander von Humboldt demonstrates
the power of the section
with techniques
that provide the opportunity to
combine the physical characteristics
of the surface materials with
botanical survey information.
So in addition to the
section lines often deploys
to limit territory are used instead
to describe topographic morphology
and to explore interfaces
between surfaces and subsurface,
land and water, earth and sky.
I feel I don't need
to tell this audience
that there are no absolute standards
or conventions in cartography,
but from a designer's perspective
there are logics, systems,
and precise techniques
for describing the ground
that are capable of
transcending scales from the body
to the territory and materials,
from the aqueous to the terrestrial,
without losing fidelity to
the condition being depicted.
As design extends its
purview to larger
and larger areas it is time once
more to look closely at maps
and plans to immerse
ourselves in their beauty,
but also to uncover their ability
to describe their projective,
to be used to imagine.
We have an even greater
challenge now
as our landscape drawings are
required to read at numerous scales,
to be interactive, to
make sense of big data,
and to describe increasingly
complex systems.
So, as you can see, it's very
general, and given the broad nature
of this topic, the intersection
of the plan, and the map,
or to put it otherwise, the
role of cartography could play
in landscape architecture, as a
representational agent as well
as a geospatially specific practice,
the curatorial process was daunting.
So the exhibit, as you see, it's
kind of seen from these images,
is for the lobby, and it's
a space with severe climatic
and security issues for the
exhibition of artifacts,
especially valuable maps coming
from the libraries of Harvard.
And so thus, most of the
materials would be reproduced
and then a few could occupy
the cases that we designed.
So this made it even broader.
So the question became how to
limit and frame the content and how
to provide these provocative
marriages
between plan and map that I'm after.
So in doing the archival research I
was always jealous of those scholars
who came in with precise questions
and bracketed timeframes of inquiry.
I could never quite
articulate what I was after.
I had them pull thousands and
thousands of maps, but I had a kind
of visual sense of
what I was looking for.
And in the end the working ended up
embracing the notion that you have
to look back, and back fairly far to
project forward, and while it speaks
to contemporary practice and
design it recovers abandon methods
and works from past centuries.
So, there's the cases.
The "aha" moment came for me with
discovery of this French catalog
of maps from the IGN
from 1949 found deep
in an uncatalogued vertical
file of the Loeb Library
of the Harvard University
Graduate School of Design.
And it may be a common artifact
but for me it was novel and a novel
and compelling way to present a map.
So inside the maps are presented
like this with a legend on the left
and an excerpt of the map on
the right, given equal weight
to the conventional signs
which are usually relegated
to a small corner status,
or sometimes don't even show
up on the map itself and
the maps they produce.
So thus the map could both be read
through its isolated ingredients
and through the composite
expression.
So this power of the legend became
one idea behind the research.
As an explanation and [inaudible]
of symbols and conventions used
on the drawing the legend describes
these ingredients of making,
and as a person who teaches
people how to draw and how
to make it became crucial.
So the choice of symbolic
language defines the character
of the representation and
the content of the drawing.
By inclusion and omission
the determination of what
to include mirrors the decision
about what makes it
onto the drawing.
The legend requires upfront
consideration regardless
of whether it is set first
of extracted post-facto.
So the legend becomes
this embodiment
of cartographic convention.
And these conventions became the
lens through which the pairings
of plans and maps were viewed.
So it started out thinking
about the way in which you draw
and what you're drawing, so the
type of ground you're drawing,
and it evolved to think about
the modes of representation
and the way in which we draw.
The use of cartography in landscape
architecture is not a new idea.
In fact, the visualization
of geographic information has
been central to the development
of a methodology for addressing the
global scale of landscape practice.
The Harvard University
Laboratory for Computer Graphics
and Spatial Analysis
housed in Graduate School
of Design supported early
environmental planning projects,
and Jack Dangermond is a graduate in
landscape architecture of the GSD,
which I'm sure you're aware.
So Landscape Professor Carl
[inaudible] used geospatial data
to aggregate [inaudible] layers
in this 1967 [inaudible] studio.
The topography, land cover, and soil
maps were rendered with dot grids,
approximating a continuous landscape
that led him to question the need
for vector information
[inaudible] terrain.
This new data output rendered
an alternative landscape reading
as well as a method for
determining the suitability
of different development types.
Similarly, Ian McHarg, the Scottish
landscape architect and planner
in his seminal 1969 book, Design
with Nature devised a system
that depended heavily on
geospatial data to codify
and [inaudible] landscape.
The making of maps was
essential to McHarg's process.
He actually calculated the size
of the computer he might need
to be able to map the world
the way he wanted as a means
to reveal relationships and
determine locations, again,
appropriate for future development.
And his maps, like this
one here of Philadelphia,
changed the perception
of the landscape.
The city had never been
presented like this
through its ecological
layers to the discipline.
So if the projects of the 1960's
separated the map from the plan
and recognized the importance
of geographically specific data
for expanding the purview
of landscape practice,
the mapping agendas of the 1990's
divorced the map from the ground.
In The Agency of Mapping by
James Corner, an influential text
for the discipline, mapping is
freed from its close alliance
with the ground, allowing
for multiple spatial temporal
readings and contexts to emerge.
The mapmaker is given the
tools to construct context
and mapping is no longer thought
to be a tool of description
or representation but an instrument
to produce ideas and actions.
The explicit goal is not to
undermine cartographic precision
but to expand the methodological
potential or agency of the practice.
The results of these influential
works are both the ascendants
of geospatial information for
the discipline and mapmaking
and design culture, as well as the
distancing of the plan and the map,
the mapping, and the ground.
Leaving design practice with maps
that privilege abstract forces
and flows over material conditions
of sight where representations
of food webs, airline routes,
and social networks become
abstracted lines floating
above a white page,
connecting symbols and pictures.
Habitats are decontextualized with
images of stylized force animals
and mountains compressed
in space and time.
Airports are places where flow
lines cross and aggregate.
Popular destinations are peaks
in a data-scape indicating a
location of frequent check-in.
Thus, I argue, it is time
to look back to a moment
when the topographic
map was at the core
of landscape architecture training.
When understanding of the contour,
as John Charles Olmsted writes here
in a letter to a fellow colleague
and academic, he's the nephew
and partner of Frederick Law Omsted,
he describes the contour being
crucial for all acts of design.
And to the point when the
topographic map underlies the design
plan, here in the case of
Charles Elliott's open space plan
for metropolitan Boston, where
for the first time many layers
of information were overlaid onto
each other and the design discipline
in landscape architecture
before this each layer was very
much isolated.
And so this became a precursor to
the McHargan method that we looked
at earlier in the Philadelphia map
and to the GIS tools we use today.
So if the idea is to look
back it's also to move forward
to address design in a
contemporary and data rich context.
As I have stated the current
project argues for this realignment
of the plan and the map
within globalized practice
whereby the rigorous practices
of cartography and the precise
conventions of drawing allow complex
and coded data rich drawings
to read as representations
of the spatial qualities of the
terrain, the immersive depictions
of the ground plain,
the morphological material
characteristics are rediscovered
through the techniques
of cartography,
from the 1:25000 topographic
map of contours,
to the 1:5000 city plan
showing figure grounds,
to the 1:25000 walking
itinerary with its line symbols,
to the 1:300 ground cover planting
plan delineated with hatching.
So from cartography immerged
systems of representation
that have direct parallels and
design and allow the material
to be structured around 10
fundamental techniques or modes
of drawing likely very familiar
to this audience and less
so to landscape architects
in training.
Soundings mark the depth of
water at a place obtained
with a polar line weighted
by lead and noted
by number on a chart at a point.
While a spot elevation is a number
on a map that shows the position
and altitude, and is one
of the fundamental ways
that landscape designers
describe landform,
the way in which we can articulate
that space between the contours.
And in fact, Olmsted in his early
rendered drawings only used spots
to show topography.
In any point system the
relationship between measurement
and drawing is direct,
scaled translation.
The physical point of
measurement correlates precisely
to the drawn representational mark.
And I should say in
the next sequence
of drawings there will always
be a kind of cartographic image
on the left and a design
image on the right.
And the labels are
being cut off I guess.
I just noticed that.
Anyways. So where was I?
Points in space and time mark
physical and temporal locations
within the landscape and are
transferred on paper and screen
as points on a map or a chart.
The results in constellation
reflects both the system
of measurement and the
complexity of the landform
or surface being measured, in
this case, the depth of Cape Cod,
and the Arctic, respectively.
There we go.
The representations are alluring,
less for their clear depiction
of surfaces than for
their potential to refill
and inspire spatial relations.
Topography is hard to read
through point distribution alone
but the points do uncover
the intricacies of landscape,
the relationships between elements,
and the corresponding
methods of measurement.
The outputs of early
computer mapping
and contemporary point clouds
reveal the plasticity of the point
as a representational tool.
The density of dots is
both infinitely variable
and bears a direct relationship
to the characteristics
of the landscape.
The approximation of
terrain has improved
in accuracy through innovation.
Limited by the availability of data
and the sophistication of printers,
the early lab work
approximated topography.
One method was to create grid
cells, two miles by two miles,
in the case of [inaudible]
drawing shown earlier,
and use the average elevation
to generate a terrain reading.
Each cell was assigned a dot density
and the agglomeration revealed
a terrain of tiny pixels.
While not spot elevations the
results yield a field of high
and low cells and the representation
is one of points rather than lines.
The grid cells obfuscate
the fluidity of the terrain,
but the method tests the limitations
of the point as a field rather
than a gridded triangular
[inaudible] pattern.
With technological advances the
field distribution has reached new
representational heights
in point cloud scanning
and visualization seen
here of the Swiss landscape
by the landscape architect
Kristoff Sureau [assumed spelling].
The ability to extract
seemingly infinite points
from the landscape itself and
from a three-dimensional model,
which we create a lot, of
the ground is stunning.
So those are points.
Isolands are joining
points of equal value,
in this case vertical distance
above and below a datum
and the contour line, as a
member of this isoland family,
is the representational stalwart
of topographic description
and projection.
I spent an entire semester teaching,
grading, and contour drawing.
It couldn't be more fundamental.
And so here are two
classic examples.
The early Dutch [inaudible]
smoothly extracted from sandings
to create the impression of a soft
watery floor, lightly articulated
against the hard fortifications
of the adjacent built environment.
And the French engineer and
parks director Alphand's work
at the Parc des Buttes-Chaumont
showing the contour plan as a means
to conquer and transform space.
Here the contour becomes a tool
to exploit technical mastery,
re-envision urban fabric, and
mobilize quick construction.
By including a contour plan in
the publicly distributed series
of promotional illustrations
called the Promenade de Pari,
Alphand demonstrated his skill
and precision underlying
the ambitious Parc des
Buttes-Chaumont project.
Paired with a rich
and dramatic engraving
of the park the contour plan showed
the before and after contours,
mathematically indicating the cut
and fill required for construction.
The abstract language may not
have been legible to the public
at the time, but the
impressive drawing bred faith
in the technical abilities
of government efforts.
It was the only piece
of color representation
in the entire volume,
these red contours.
And contour manipulation remains
the fundamental design tool
of the landscape architect.
The side-by-side comparison
of Dupantriel's [assumed spelling]
groundbreaking terrestrial contours
and James Corner Field
Operations plan of the University
of Puerto Rico Botanical Gardens
offers differing visual approaches
but shares this understanding of
the agency of the contour line.
Whereas the International
Hydrographic Organization
and the [inaudible] drawings
share a representational language
in coloration, showing
a clear influence
of cartographic convention
and design.
And the contour remains
design's most expressive tool
for representing topography,
here, describing the land forms
of a park proposal in Taiwan,
by Stoss Landscape Urbanism.
So we're going to move even
further back to hachures,
a definitely outdated
system of short lines
that follow the direction of
maximum slope to indicate relief.
And they're a close relative of
the hatch and design drawings
where lines or other
patterns mimic shading
and articulate surface variation.
And the hatch remains
integral to design practice,
and different materials
have standardized hatches,
but I would say the experimentation
with them is still somewhat limited.
So using lines to approximate
slope, shadow,
or texture has been traditionally
under explored as a means
of innovation or invention with a
few stunning exceptions that I'd
like to point out in these examples.
These contemporary
examples and this map I love
with unsystematic and messy lines.
So don't tell cartographic
terrain masters Emhoff or Lehman.
I think it yields a vivid
topography and really a sense
of the spatiality of that ground.
Compared with the fine
texture of the adjacent drawing
that emulates the feeling of
a forest that's being sought
after in this building and
represented in its plan drawing.
So I would argue at
least from my sphere
that the hatch may be coming back.
Where protagonists are
testing its potential
as seen in this hybrid drawing.
It's a plan section
isometric drawing
for a new public realm
in [inaudible].
So the hachure is replaced
by the shaded relief,
arguably the most vivid
and illustrative form
of topographic depiction.
It paints a complete picture
of the landscape using color
and tunnel [inaudible] to
blend varying elevations.
The terrain then is not
translated into a series
of where diminished white
space indicates deepness,
as with the hachures, not abstracted
into concentric horizontal slices
where they do not exist as
could be said of contours.
Instead the shades relief
captures the most intricate forms
in a fluid gradation
of tunnel variation.
And I find this pairing particularly
provocative with the Emhoff drawing
on the left in soft of a
highly influential cartographer
for this exhibit, and I think at
least for the design disciplines.
He was a purist with regards
to describing terrain.
He used color to define elevation,
optimizing a drawing's effect
through this color gradation.
In this portion of this famous
large-scale [inaudible] map he used
India, red, brown, gray, blue,
or deep green to describe ground
shadows and bright ochre's,
yellows, reds for sunny slopes.
The results highlight contrasts
blue-yellow and red-green
to present the intersection
of the Swiss lakes
and hills through blocks of color.
There are no lines
in the painting just
as there are no lines
in the landscape.
The architect Zaha Hadid's
blue [inaudible] painting
from the Hong Kong Peak
Leisure project is also rendered
as lineless blocks of color.
Hadid's proposal for a manmade
granite mountain of slabs set
against the intensity of
Hong Kong, between mountain
and harbor reads like a landscape.
You almost can't find
the building within it.
Again, tones of blue are
set against reddish browns
and pinks exclusively crafted with
hard edges to define color shifts.
In both cases tectonic
vision is clearly articulated
through surficial rendering and
carefully considered palate,
and this section on shaded
relief hopefully will include one
of the Hal Shelton maps
from the Library of Congress
that I'm very excited
to see tomorrow.
We depart from topographic
representation
to look at land use and color.
So if the ground morphology
is the first idea
of the landscape architecture
training this sort of vegetation
and the way we occupy the
land is also fundamental.
And so we look at the idea of
moving from a hand coloration
with watercolor to the extraction
and amplification of patterns
from aerial imagery, to the
potential overlap of these systems
to try to bring a textual
quality and a precision together.
Categorization and tolerance,
where to draw the line
between categories creates
the visual differentiation
of the ground plane, whether date
is gathered on foot by the surveyor
or through remote sensing
and satellite.
The classification is inherently
reductive, requiring delineation
within a continuous matrix.
The idea is to balance clarity and
description to find a hierarchy
that translates land
into clear taxonomies.
Once the categories are
determined choices must be made
as to how the map will be
drawn, flat or highly textured,
bright or muted, true
color or infrared.
Drawings that categorize
and delineate types
of land cover require
simplification and generalization.
These maps are deceivingly static,
masking the dynamic
process of occupation.
Landowners, land uses
change, vegetation appears
and disappears continuously.
The temporal capacity
of contemporary visualization offers
great potential to explore this,
something alluded to
and done in this project
that articulates the interface
between urban and ecological systems
as part of a proposal that linked
human habitation, waste processing,
and wetland regeneration
in New Orleans.
So among [inaudible] classification
the figure ground is probably the
most simple but yet
very provocative,
it's a binary coding
method that affords a clear
and powerful spatial
reading of the landscape.
We make figure grounds of
anything we can, I think.
The separation of object from
field underlies all cartographic
and design drawings.
While a figure can
represent anything,
the most fundamental figure
ground relationships in design
and cartography include three binary
oppositions, land mass versus water,
land form versus flat ground, and
built structure versus urban fabric.
[Inaudible] realm is perhaps
the most celebrated variant
of the figure ground
depiction of urban fabric.
Countering the dominance
of the bird's eye perspective
[inaudible] produce an exquisitely
surveyed and drawn icographic
[assumed spelling] map of the city.
The built form is shaded while
white represents public space,
both interior and exterior.
The map is not a traditional figure
ground of built and open space
but of public and private
occupation.
Building structure and vegetation
are filled leaving an unencumbered
ground between them.
The map and the plan
here are truly conflated
and the [inaudible] plan has long
been an inspiration to design
and in this case to urban designer
Juwan Buschettes [assumed spelling],
who adds a layer of
analysis in his drawing
of the fabric of Barcelona's
Old City.
His map includes temporal change
with archeological ruins
depicted beneath new buildings.
The buildings themselves
are classified typologically
and in coloration to the fills.
Side by side the [inaudible]
and Buschette's map
introduced the lasting power
of this representational innovation
and precision of the figure ground
as it has been adapted over time.
And here the figure ground becomes
a tool to articulate design,
a master plan proposal for the
extension of Monaco here as a series
of archipelagos into the sea.
So we're going to move further
afield to the stratigraphic column.
As a tool extracted from geologic
mapping that has relevance
to describe the depth and
layering of the landscape
that we've talked about earlier.
These simplified columnar
diagrams relating the succession
of named lithostratigraphic
units from a particular area
to the subdivisions of geologic time
that accompany geologic maps
can and have been coopted.
So here we're using this tool
to describe a thickened ground
and to describe the land use
in a sectional form in Paris.
Or here, as the basis of a project
conceived, through the manipulation
of lithospheric layers to
produce varied micro-climates
across a public park,
set in relationship
to this famous William Smith map.
Or, the sequence of
Comstock Mine Maps followed
by a Stoss landscape
urbanism drawing of a plaza
at Harvard described two
approaches to registering the depth
of infrastructure, each
taking subterranean slices
and representing the hidden networks
with vibrant, coded coloration.
Perhaps it has become clear
that geology has long
been a preoccupation
of landscape architectural practice,
from the work of Frederick Law
Omsted and Charles Elliott,
to contemporary proposals
by young designers.
This is [inaudible] proposal for
a new linear public space derived
through a reading of the
city's topographic geologic
and hydrologic layers,
set in relationship
to the patterns of urbanization.
These are the kind of maps we make.
The cross-section is
both a signature feature
of the geologic map
and a fundamental type
of orthographic drawing
in architecture
and its [inaudible] disciplines.
The sectional view compliments
the plan while affording an
understanding of material
and structural relationships,
overlaps, and wedges.
[Inaudible] Humboldt's widely
distributed, richly illustrated,
annotated, and exaggerated section
shows the vertical distribution
of plant life, subsurface,
surface, and atmosphere qualities.
Plants are shown heavily
at the base of the drawing,
thinning the upper
reaches of the volcano
with the cut away portion
listening the plant species
by elevation and region.
The image arouses the imaginary,
blending scientific material
with an evocative qualitative
landscape depiction designed not
only for the academic community
but also for all those interested
in natural sciences,
exploration, and pleasure.
Here it is paired with urban
designer Felipe Correa's reading
of that same volcanic corridor,
this time articulating topographic
manipulation showing the cuts
and [inaudible] required by
the urbanization of the region.
Barring the tool of the geographer
who used the linear transverse
as a means to understand new
territory, both as a line struck
across a map or line walked
through a physical terrain,
this set of sections describes
a walk across a design landscape
with a sectional length
corresponding to the view shed
as determined by topography, which
could segue to the next section
on lines, line types, and
lines as itinerate devices.
In the mapping of geographic
elements lines serve
to represent both entities
without dimension, boundaries,
property lines, ship and bus routes,
and those with dimension, roads,
rivers, drudge channels, rails.
With the latter changes in scale
reveal the areas behind the lines.
While the data layer may be
visualized as a line zooming
in reveals the thickened landscape.
On a map the road is a line but it
has a dimension and a right-of-way,
a spatial territory evaded by
the line symbol but included
within the classified dataset.
What is merely a hairline
on a map gains measurable
and legible dimension at
the scale of the body,
and is articulated thusly
in the architectural plan.
As shown by architect
Dimitris Pikionis's project
for a paved pedestrian route between
the Acropolis and Philopappos Hill,
as an orchestrated
procession articulated
through intricate stone surfacing.
As a drawing and a landscape
the textural variation
of the thickened line is
produced through simple elements.
Lines of material jointing
and topography.
So finally, at technique number 10
we return to the conventional signs
as the key to the graphic
integrity of a drawing,
and to charts where the navigational
symbol sit boldly atop a faint
and discontinuous terrain,
themselves making a type
of landscape that is
carefully calibrated spatially
to ease navigation.
They appear abstract with
annotation trumping other readings,
and yet they distill
and reveal key points
of spatial correlation
between map and ground.
The hierarchy and alignment
of the information is crucial.
In design the conventional sign
takes on an iconic-like quality
where the most successful
examples still have measure
and spatial fidelity.
So while they articulate
programmatic uses or land types,
in this case for the
Sultan Sea a series
of water purification strategies.
Or in this case as part of an atlas
presenting new energy alternatives
for the Taiwan strain.
So I want to end this section
where I began the talk,
with two for me particularly
compelling legends,
very different legends,
that speak to the diversity
of cartographic language in the
20th Century, and to its potential
as a transcendent language
that could inform the work
that we do moving forward.
And also to point to
some initial experiments
in translating this language, these
are experiments that were produced
as part of this project that try
to conflate rather
than pair and compare.
The first is a series that you
see here of exhibit graphics
where elements were extracted
from the maps and used
to draw designed landscapes.
So for example, the crosshatch
shadow line mapping process
developed recently by the
developer Pat Kennelly is used
to depict the intricate topography
of Alfonse Buchmann
[assumed spelling] project
in 19th Century Paris, radically
changing the traditional reading
of this landscape.
Or to the stratigraphic column
and hatch describing the layers
of Enric Miralles and Carme
Pinos' Igualada Cemetery outside
of Barcelona.
So these are sort of static, almost
dumb conflations of the idea.
And then I'm going to try
to play you an animation
that explored the representational
advantages and disadvantages
of the various techniques.
So in this case it's going to be
a video about drawing terrain.
There were sort of three that
we dealt with, one having to do
with figure grand, and the other
having to do with the peeling
and unpeeling of stratigraphic
layers.
For if anyone who's
interested contact me.
These are of the [inaudible]
area of Switzerland
with incredible GIS information
and they were produced
by a talented colleague
of mine, Robert Pietrusko.
So, we'll see.
This is where I come from the
terabyte side of the conference.
I'm going to attempt this --
it's only a half a gigabyte
video, but we'll try.
So as we watch a portion of
this video I will conclude,
through the application
of techniques,
and instrumental pairings of
images across the disciplines,
the power of merging cartographic
and design representation
is elaborated
through visual documentation.
The collection allows for the
extraction of similarities
and differences across
images, across eras,
and across ways of drawing.
This parallel between the map-making
and the plan-drawing is
not a new idea but one
that deserves contemporary
reconsideration
to bring together the
meticulous detail of cartography,
the prevalence of data,
and the ambition of design.
Through a revival of the craft
of drawing and an emphasis
on precision, specificity,
and invention found
across the disciplines there's a
potential to reengage the territory
and to influence the way in
which designs are enacted.
Thank you.
[ Applause ]
>> Thank you so much, Jill.
That was really very,
very interesting.
In introducing our next
speaker, John Hessler,
I'd like to congratulate him on
the planning and coordination
of this program and also
to include all the staff
who supported the behind the scenes
activities and manning the desk
up front, and providing all
the information to those
who have participated
and attended today.
John Hessler is a specialist
in modern cartography
and geographic information
science at the Library of Congress,
and a lecturer in mathematical
cartography and GIS
at the Johns Hopkins University.
He has written extensively
on the history of mathematics
and cartography, and is the
author of many books and articles,
including The Naming of America:
Martin Waldseemuller's
1507 World Map
and the Cosmographiae
Introductio .
And A Renaissance Globemaker's
Toolbox: Johannes Schoner
and the Revolution of
Modern Science, 1475-1550 .
Please welcome John.
Thank you.
[ Applause ]
>> John Hessler: Thank you, Jackie.
I guess we should have a PowerPoint.
Well, in the meantime, I just
want to thank you all for coming.
You will see once the PowerPoint
comes up, there were three of us
who were supposed to be in
this particular session.
Laura Kurgan was also
supposed to be in this session
but she had a conflict
and we had to move her up.
So, this session was purposely
planned to provide a bit
of a conflict between what I'm going
to say and what Jill said and then
with Laura kind of
being in the middle.
So just so you get a little bit
of the context of the planning,
that's the way it was
supposed to be.
Anyway, the title of my talk
has interested a lot of people,
the magpie and possum
part especially.
They're the names of cats,
and they're the names
of two particular cats,
cats that appear in a book
by the mathematician David Lewis.
And the book is called
Parts of Classes ,
and Parts of Classes is a book
on topomariology [assumed spelling]
and topomariology is something
that is sort of an active area
of research now that sits at the
base of sort of the foundations
of GIS, the real deep
spatial theoretical principles
that are being put to use
in those things in GIS.
And by the end of the talk you'll
have a little bit better idea
of the title.
The other part of the title, which
is somewhat purposely provocative,
"The End of Cartography."
Although Jill has provided a
perfect anecdote to the fact
that cartography is not ending and
that drawing is certainly alive
and well, I hope by the end of talk
that you'll see that a certain type
of cartography has advanced to
a point where we're beginning
to take things into account
that we never could really
take into account before.
And part of that Jill mentioned,
which is the idea of flow
and the idea of time, which
Doug Richardson will talk
about in his keynote address.
Now the title of this talk could
just as easily have been something
like "Chasing Tobler's Dream,"
or "In the Shadow or Tobler,"
or "In Tobler's Footsteps."
Because a lot of what I'm going
to talk about really comes
from the early work of Waldo Tobler.
I became a geographer or interested
in geography in the early 1980's.
I was perfectly happy teaching
mathematics at Harvard at the time,
and I came upon a paper that I
was going to use for my class,
which was his cellular
geography class,
and was immediately enthralled
with the possibilities.
But, so a lot of what you're going
to see today comes from
the work of Tobler.
My talk is going to bounce in and
out between the modern and the past
and the present, and there's many,
many videos embedded, so we're going
to keep -- we're going to hope
that the technology is working.
So we'll see what happens.
Now, really, in my estimation,
and of course these are opinions
that are mine, I'm not going to
attempt to go for completeness
for the historical material.
Most of the historical material
in the 20th Century is of course
under study at this point.
It is a field where the archives
and the materials are extremely
ephemeral, so when one is talking
about things like the history of GIS
or the history of cellular automata
in cartography, the source
materials are necessarily spotty
and we just don't have the
distance on a lot of this
to gain a great historical
perspective.
But, as far as the
future of cartography goes
to a certain extent
this is my opinion
and I'm not the only
one who shares it.
And really I believe the future
of the discipline needs to focus
on processes and not structures.
Not fixed items, not things
that sit fixed in time.
All geographic systems, from rivers,
to glaciers, to anything we want
to map, to anything we want
to study for the most part,
reject any notion of equilibrium.
The one thing about
the static paper map,
it is obsolete the minute it is
printed because normally everything
or some things have changed.
Now this is kind of
a new perspective.
I'm, like I said, not
the only person.
Michael Batty in London is
pushing these sort of ideas also.
But this is kind of where we're
going to be going with this talk,
towards the idea of processes,
towards the idea of disequilibrium
in cartography and how do we handle
the notion of disequilibrium.
Now, we can really talk about
kind of a hierarchy here.
We can talk about time, just the
flow of time, just as time ticks
on a clock, but that's not
really interesting necessarily.
To take snapshots of time is not
necessarily something that's all
that fascinating.
There is the idea of flow, the idea
that the flow of a river or the flow
of people or the flow of
transportation, the flow of energy.
Again, that's something
interesting to map
but in fact it only shows
us one particular dimension.
But if we're going to go a step
higher we have to start dealing
with notions of complexity and
really the science of complexity
and this is a brand-new
field for the most part,
maybe 30 years old, 40 years old.
But only really now beginning to be
able to do anything computationally.
It is only in the last 10 or
15 years that the structures
in complexity, the ideas of
emergence have really come
to the floor as a computational
possibility in geography,
and in other disciplines as well.
The question of emergence, when I
say emergence I'm talking about sort
of things that organize themselves
but don't necessarily
have causal properties.
Let's think for a moment
about crows flying around.
You watch the crows fly.
It's not one crow who has a plan,
it's all of them moving
in conjunction.
When you're sitting in the
chair here and you're deciding
that you're bored because
I'm going on too long,
or that you're desperate
for a cup of coffee,
it's not because one particular
neuron in your brain fires coffee,
it's a whole bunch of them
getting together as a chorus.
Ants, when they're building an ant
colony, there's not an algorithm
in each ant, it's the mix of all
of those things happening at once
without a plan that create order
out of what seems to be chaos.
And most of the geographic
phenomenon that we
as geographers are interested in
and what cartographers have
always wanted to grasp is
to a certain extent
that type of phenomenon.
Now, this really began early
on when people started thinking
about how is it to handle this
idea of flow, this idea of motion?
And really the first
one is something --
the first attempt is
really something called
the cartographitron.
And a cartographitron was
actually invented in Chicago
in 1959 to study traffic flow.
And they took millions of traffic
flow readings and entered them
into an interactive
cathode ray tube.
Now, they didn't call them flows
at the time, the actual line
in the Chicago report that shows
how this technique was done --
they're called lines of desire,
which is a much more poetic way
of talking about driving
home and sitting in traffic.
Your line of desire is
being interrupted by others.
But this is really one of the
first attempts to do this.
Now, other attempts were
later on looking at this idea
of flow, this idea of time.
One important one was a movie
that was created by Waldo Tobler
for the growth of the
Detroit region in 1970.
Now unfortunately, and if
any of you know anything
about where you might
be able to find a copy,
although Waldo Tobler's paper is
a seminal paper in looking at this
and one of the most groundbreaking
in this area, one of the first times
that time derivatives appear
explicitly in a mapping equation,
the movie is no longer extant.
Waldo doesn't have a copy
and all of the sources
that he's told me that
might, do not.
So what you're actually seeing
is the fact that because I wanted
to show a video of it, it didn't
exist so I reprogramed it and did it
with a modern Mathematica.
So, this is what Waldo's
program would have looked like,
although his was a
lot more primitive.
He had a whole bunch of surfaces
that showed the growth of Detroit.
Now one of the things
that's interesting
about this early attempt is not only
that it was an empirical
thing at first.
He actually looked at the
empirical growth of the Detroit.
But Waldo simulated what
the growth would have been.
And this notion of simulation in
Mark's talk, he sort of looked
at two poles at the end there,
one talking about in the
20th Century this notion
of primitive mapping
and I believe he said
as a social scientist he was
interested in that aspect.
And he left the simulation
part alone.
Well, I'm interested in the
simulation part and that's part
of where mapping in the
20th Century is moving.
We'll talk a little bit about
this a little bit later,
but we're beginning to get to
the point where we're being able
to start to predict, to look
at counterfactual situations,
to question what will happen if we
do this as opposed to looking back
and mapping what happens
already because we've done it.
A map -- the static paper
map, to a certain extent,
is almost always a historical
document even when it's made.
Whereas when we begin to look
at future simulations
things change a little bit.
Now when I'm talking about
modern flows I'm going
to talk about emergent flows.
And what you're about to see --
could you dim the lights
just a bit, John?
What you're about to see
here is basically millions
and millions of data points.
This is a day in the life
of the public transportation
of Great Britain.
This is a sort of modern cartography
that I'm going to be talking about,
and I'm going to be talking
about what got us to this point.
When we start looking at this, you
can see the time clicking up there,
we're at about 4:50 a.m.
right now, and we're beginning
to see London wake up
and the commuters begin
to get in their cars.
The pink part is the
subway waking up.
And now what you see is the
air flights have just begun
to take place.
Now when you look at
a map like this,
when you look at this occurring,
how can you think of a static map?
The energy flows are different.
What is the population
of a particular place
at a particular time?
How are these networks interacting?
Cellphones, all of that sort of
stuff, these are emergent properties
that are taking place
without prediction,
without necessarily
understanding causal rules.
But we want to be able to
study these, we want to be able
to understand how these actually
take place, how they interlock
with one another in order to plan
everything that urban planners
and geographers attempt to look at.
So when we look at something
like this it becomes difficult
to think of it as a static map.
You're almost looking at a place
where the actual topography
of the place itself is immaterial
to what's actually happening.
And so, this is what I'm talking
about when I'm talking about a flow.
And this is actually, like
I said, run on a very --
it's about 8.5 million data points
went into actually producing this.
Now, when we talk about
this kind of dynamics,
when we talk about this kind
of flow, we have to kind
of invert the way typically
geographers have talked
about location.
Okay? Now, places are not
locations where things happen.
A city is not a place
where something happens.
A city is there because it's a
place where things were happening.
In other words, we have to
invert this idea, that of place.
We have to start looking at
interaction and redefine the notion
of place as where things are
interacting, and not worry so much
about the actual underlying
topography.
And when we start thinking
about this in terms of internet,
globalization, economic
flows, banking systems,
place is almost immaterial, okay?
It's connection.
It's topology that
makes the difference.
And when we're talking about sort
of modern ideas in cartographic
and geographic analysis
sometimes London is a lot closer
to New York City than New Jersey is.
And really those things are
extremely, extremely important.
This sort of loss of
the notion of distance.
Now, we can talk about
change, and I'm not going to go
into a lot of equations here.
Anyone who's seen my
talks before knows
that there's always
an equation somewhere.
But we can start talking really
about the idea of spatial change.
How is it that things
actually change?
What is the drivers
of changes in space?
And we can think of events
that are random through time.
We can think of things
like historical accidents,
some physical constraint
on the area.
I mean you're not going to put a
golf course in Southern Algeria,
unless you really do something
that affects the water.
And we can begin to start
thinking of these things in terms
of functions of time, as
opposed to just events
that take place at a single moment.
And when we start doing this,
thinking of these things
as functions of time, we have to
begin questioning how do we model.
How do we model spatial
change through time?
How do we represent those flows?
And how do we understand what
we saw in that visualization?
It's one thing to visualize it.
It's one thing to take eight million
data points, put it into a computer,
throw it into some easy algorithm,
and show a visualization like that.
It takes 15 minutes.
But it's another thing
to try and figure
out what is going on
underneath there.
Because it's not only
that one network,
that's simply the public
transportation network.
When you start thinking about all
the other interlocking networks
that are part of that,
the underlying geography
and the underlying
things that are happening
in a geographic space
are much more complicated
than any static map could
ever be able to show.
Now, there was a lot of
experimentation early
on in this sort of idea,
and one of the attempts
at that was a thing called
American Graph Fleeting,
which many of you have
probably seen.
This was produced in the Harvard
Lab by Geoffrey Dutton and a number
of other people, and basically what
it's showing is it's showing the
population change from
colonial times all the way
to 1976 when this was produced.
And again, what they did is
they took this population data
and it was an extremely
complicated engineering feat
to actually do this.
You see Geoff Dutton
there mesmerized,
staring into the actual output,
which is actually a
holographic output.
And basically what you're looking
at here is a time-lapsed photograph
of single shots of a map that were
actually produced holographically.
And I'll show you a little
bit about that in a second.
But what you're actually
seeing is emergent phenomenon.
You're seeing the growth of
the copulation of the country,
which doesn't have causes,
it's just taking place,
but how do we understand
what's actually happening?
How do we understand those
underlying structures?
And when they produced
this particular video,
this particular movie everyone
who looks into it seems --
it must have been evil because
everyone who looks into it seems
to be totally mesmerized
and their brain seems
to have been taken totally away.
This is an advertising photo
that was done by the Harvard Lab
who actually produced it.
But this was actually done,
each little single change was
done on a holographic strip.
The holographic strip was then
mounted in a circular plastic oval,
or round, plastic circle
and when you walked
around it what would happen was each
of the holographic films would
then appear to you registered
and you would see this change.
And what they did is
simply filmed that movement
around this holographic piece.
But there were two key things
in the lead up to it that --
and because Nick Chrisman who was
at the Harvard Lab donated his
entire collection of papers
to the Library of Congress there
was some really interesting notes
that went with this
particular thing,
and I extracted a couple of those.
And basically they said that this
was an experimental four-dimensional
demographic map which attempts
to summarize the dynamics
of nearly two centuries
of population growth.
Attempts to summarize the dynamics.
In other words, the static
maps were not enough,
they wanted to have
some sort of dynamics.
The other thing was, and the
future of cartography lies
in going beyond the map
and approaching the real.
And I think that is somewhat what
the dream of the 20th Century is,
to approach some real
predictive, real modeling power
of geographic and complex
phenomenon.
Now, there's really a
couple of challenges here.
Now, this is a -- what
I've presented
so far is an extremely
positivist outlook.
I just want you to know that I
don't have that positivist outlook,
I'm not that confident in
our ability to do this.
There's been lots of
gimmicks before.
But to a certain extent this
time we seem to actually be able
to have a balance of
computer power, algorithms,
and vision that area allowing
us to push a little bit forward.
And really, there's
really three things
that we really need to
get under our belts.
One is the question of time.
How do things actually change?
And Doug is going to talk
about time and the integration,
space/time integration into
geography in the final keynote.
But then flows.
How do things interact?
And we saw those, the
transportation networks there.
But how do those urban
transportation networks --
one of things we're
working on now is combining
that with cellphone data,
with cellphone calls.
And you'll see between 4:00 and 6:30
in the evening in London the amount
of cellphone calls drop off
because so many people have now gone
into the subway and can't
use their cellphones.
And so you can see how all of
these networks are interacting,
and they're interacting in ways
that are not necessarily
obvious and predictable.
But to actually plan for these types
of things, for our future challenges
on both domestically
and when we're talking
about any networks worldwide,
understanding those
processes are important.
And scale, one of the catchwords in
all of this has to do with scale.
At what scale are we
looking at these things?
We found that there
are now scaling laws,
laws that show how size
distribution, for instance cities,
the distribution of cities are a
power law, which I won't get into
but even Paul Krugman the
economist has just written a paper
about the paradox of the city size
power law, that they only come
in these batches, they
don't seem to be random.
We don't understand
there's this empirical fact,
we don't understand why they're
not random, but they're not.
They follow this power law.
And so understanding
these kind of things,
these kind of dynamics is
really where I think the future
of cartography and
cartographic visualization,
not meant to [inaudible]
map are kind of moving to.
Now, to visualize time and
to understand how time flows,
early on there were a lot
of different attempts,
and many of them were
of limited success,
and some were of greater success.
One that's still with us
today is cellular automata,
and that's what we're really
going to talk about for kind
of the rest of our little talk here.
We're going to talk
about cellular automata,
and cellular automata were first
envisioned by John von Neumann,
and Neumann is one of the
great founders of computation.
Neumann had this idea and as Dr.
Strangelove as sounds as it is,
that through a computer and through
cellular automata he could reproduce
self-reproducing machines
and organisms.
He would be able to create machines
that would reproduce themselves.
And he came up with
this conceptual schema,
a very simple computer
program which was based
on what are now called
cellular automata.
Now, these cell-space models, and
basically we'll talk a little bit
about the details of them in
just a bit, really entered
into cartography really not so long
after von Neumann proposed them.
And there's a couple of
really important papers.
Hagerstrand's paper on diffusion
models using these nearest
neighbor cells.
But really in the 1960's
was really the first time
that people began looking at them
for simulations and
land development.
A lot of these things were done by
hand, they were calculated by hand.
The algorithms were so simple
you were able to sit down
and do the actual calculations
by hand.
And there were several theories that
through these we could actually --
the heterogeneous nature of space,
of geographic space would
be excluded from the models
and we would get some pure notion
of what the mechanisms
underlying phenomenon were.
Now that is of course impossible,
it was a pipedream, but it was one
of the things that sort
of started people thinking
about cellular automata
as a possible way
to simulate geographic phenomenon.
Now really the first person
to really be able to do this,
to come up with a method that
seemed usable was Waldo Tobler.
In a very important paper, in
a seminal paper I would say,
called "Cellular Geography," a
paper that I'm sure is not read
by very many geographers now,
it's a very technical paper,
interestingly enough it was
published in a book by Gale
and Olsson called Philosophy
in Geography .
So we've got this rather
complex cellular automata model
which is appearing in a book
on philosophy in geography.
Philosophy, the cellular automata --
the philosophical import of
cellular automata is quite bold
or quite active today.
People like Daniel Dennett,
and things like that,
looking at cellular automata to
mimic the way the actual mind works.
And also people believing
that the universe --
Stephen Wolfram in the new
kind of science thinking
that the universe itself is
actually a cellular automata.
So there is a lot of
philosophical talk going
around these sort of
things right now.
But at this time this
was a fairly new idea.
Now, Tobler introduced it and at the
time when he wrote this paper he was
at the University of Michigan,
and there was also a man
named Arthur Burks there
who was studying automata
in what was called the Logic
of Computers Research Group,
which started in 1956.
And Tobler seems to have been
influenced by some of the work
that was going on, although he
has told me Burks was a student
of von Neumann, published
posthumously actually von Neumann's
papers on cellular automata.
But in speaking to Waldo he said
that he had never read a paper
by von Neumann, that there
were other influences
that moved him towards the
cellular automata paper.
He wrote the paper in Vienna
in an institute that was
at that time being directed
by Howard Rifa [assumed
spelling] the economist,
and he also worked a little
bit on Bayesian Decision Theory
at the period, and Waldo has told
me that there's a Decision Tree,
which appears in his paper,
was basically inspired
by his contact with Howard Rifa.
Now, at the same time that
Tobler was writing this,
during this period there
were a lot of people,
more in the conventional
computer topography world
who were also trying
to look at time.
This was produced, this is a fairly
famous video of a Symap video
which was the main program
before ODYSSEY that was produced
by the Harvard Lab for Computer
Graphics that Jill mentioned.
This particular one was done
by Allan Schmidt filming a
whole bunch Symap printouts
to show the development
of Lansing, Michigan.
This particular one is like watching
paint dry, so I'm not going to --
I'm really not going to
torture you with much of it.
But you can see here one of
the things that took place
in the early development of
GIS, in the early development
of computer cartography was
this continual going back
to how do we make this look
like a function of time?
How do we inject time into this?
You know, they had
all the map printouts,
they had all the population
printouts
for the American Graph Fleeting,
but yet time was something
that was highly important.
Now, we're going to
talk a little bit
about Tobler's cellular geography
and then I'm going to actually --
oops, sorry about that --
and then I'm actually going
to show you a couple of cellular
automata that I'm working
on that are actually about,
you know, modern simulation.
But first we need a
little bit of ground.
So the article was published
in 1979, and Waldo has told me
that what he was in
writing it was an article
in Scientific American
by Martin Gardner,
and of course Martin Gardner was
the great meta-mathematical themes
author who published in Scientific
American for many, many years.
And John Conway, who was a
mathematician had just invented this
thing called the Game of Life,
and we'll talk a little bit
about that in just a second.
And this thing took off.
It went in the -- in 1979 terms,
went viral with basically
no one being able
to program this particular game into
a computer yet, but people doing it
by hand and then sending.
John Conway had put
a contest out that
if anyone could produce a sequence
of algorithm or a sequence of steps
in the Game of Life that produced
stable evolutionary real life
that evolved or life inside
the computer that evolved
that he would pay them a
certain amount of money,
and this of course
started a whole craze
where telegraph were
paying back and forth.
Waldo himself actually
sent several of those in,
which he's actually sent me
Xerox's off which he's kept.
The other thing that Waldo -- that
influenced him was this series
of papers by Stan Ulam, who was
a colleague of von Neumann's
and wrote a whole bunch of
papers on random processes.
So, really, however, it
was Conway's Game of Life.
Now Waldo gave five different
models, five different ideas
in his cellular automata paper, five
different ways he could envision --
sorry, five different ways he
could envision things taking place.
And he produced this diagram
which is in his paper.
And what the five models are,
and really we only have to care
about the last one, are
basically how the neighbors
of a particular place interact.
So if you're living in a particular
spot, and his thing focused
on land use, what happens around
you influences what you do
with your piece of land.
And basically the assumption
was that in the fifth model,
the geographical model, land use
at some location is dependent
on the land use at
neighboring locations.
And this is really an instantiation
of Tobler's first law of geography
which is that, you know,
everything affects everything else,
but near things affect
things more than others.
And so basically there was
this nearest neighbor idea,
that neighbor -- what happened next
to you, what happened in a land,
in an area around you affected
what happened to your piece of land
or any other piece of land.
So this was kind of a
notion of if you take
that as a given you can
look at how does it evolve?
So if what happens next to me
affects what happens to my piece
of land how does it evolve?
And so basically he
came up with these ideas
of nearest neighbor models and
really there's been several of these
that have taken place,
several different models
of what a neighbor is.
A cellular automata divides the
world up into a grid and each one
of those grids has
a particular value,
a particular discreet value to it.
And what you take as a
neighbor, for instance,
in the one here you're taking
the whole square around you
as your neighbors, the red,
and in the other one below
you're only taking the ones
at the cardinal compass
points as your neighbor.
You can also do action
at a distance,
saying that your neighbor
is 10 things away.
But for the most part a neighbor
is what's right next to you.
Now, this is actually an
output of a cellular automata
for the disappearance of glaciers.
This is a prediction that
I've been working on.
What you're seeing is
the red areas are places
where glacial melt-back is expected.
The blue areas are stable areas, and
the intermediate areas are places
that are basically land,
the purple and the other,
which don't have any
melting potential whatsoever.
And so this the kind
of thing you would get.
Now you don't have to do
it in only two dimensions,
you can do it in three, and you
can do it any number of dimensions
but obviously you can only
visualize it in three.
But really the way it works is
the cells, each one of those grids
in some dimensional
space has some --
is adjacent to a whole
bunch of other cells.
Each cell can only take on
one value at a given time
from some set of possible values.
The state of any cell depends
on the states of the other cells
in the neighborhood, Tobler's First
Law, and then the key point is
that you have to define a
set of transition rules.
In other words, as time goes on
each one of the cells, for instance,
in Conway's Game of Life, if
you're surrounded by three cells
that are alive, well, then you
die because of overpopulation.
If you're surrounded by just --
if two cells are alive,
well then you're okay.
If you're only surrounded by one
cell of a person that's alive,
well then you die from loneliness.
And you can -- the patterns that
are generated from these very,
very simple rules -- and you can do
this for land use and you can come
up with all sorts of various
things, are extremely complicated.
This is a three-dimensional
output of a cellular automata
which basically is
about segregation.
This is an agent-based
model, I'm not going to go
into agent-based models just
because I don't have a lot of time.
But this is a typical
three-dimensional output,
it doesn't have to be
just two-dimensional.
But this is actually a mapping over,
in this case, a geographic
space, so.
Now, when we talk about simulations,
and I'll finish up in just a second
with an actual cellular automata,
we really are talking about kind
of three different types,
and we tend to classify
simulations by algorithm.
There are simulations
which basically we take
differential equations
and we transform them
into discrete variables.
These are things where
we have sort of causal
or we can solve the differential
equations in a way that makes sense.
There are things called Monte
Carlo methods, and this is sort
of where we inject sort of
randomness into processes.
And then even if there's no
underlying sort of indeterminism
in the system but we inject
a certain randomness in order
to perturb it, in order to kind
of then simulate what's happening
by perturbing the system randomly.
And the other one is the cellular
automata which for the work
that I'm doing which
I'll show in a second,
really gives the best results
because really I don't know any
of the underlying equations
which are driving the system.
I can have ideas.
I can look at data.
I can derive suggestions.
But what cellular automata allow me
to do is basically
play with the system.
Now, this particular one is
an area that I'm interested
in as the Kislak curator.
This is a simulation of the entire
history of Mayan civilization
and basically, it's to
simulate the Maya collapse.
So what we see here is
we see a geographic area
and this is the cellular
automata running.
So each of these cells
and there are many,
many layers of different
networks on top of here.
This happens to be population
and we can see it increasing
and decreasing.
This happens to be
forest and land cover.
So we can see and those things that
you see those networks building,
once the population reaches a
threshold, it begins to migrate.
Those are migration
lines between cities.
You can see the population
growing and now fading.
And what's amazing about
cellular automata is the idea
that you can keep track of
particular types of functionalities
so the graphs that you see on the
right here are basically the one
down in the lower right
hand corner is forest cover,
what kind of trees
are being cut down.
But all of this needs to be
nested in these transition rules.
And so really, as far as I can
see the future of cartography,
simulation is going to be probably
one of the most important aspects.
It's still in its birth pangs.
We are still in the very
early years of being
to actually do this effectively.
But the tools are being placed
in more and more people's hands.
Even things like CrisisMappers,
the online group of people
who come together to map refugee
crises and things like that,
are actually using
agent-based models
to predict refugee movement
and that sort of thing.
There are some open questions
and I'm sure maybe you have some
open questions and thank you.
[ Applause ]
>> Mesmerizing, John.
Thank you.
We have a few minutes here.
I was wondering for Jill and John,
if anyone has any questions
of desire for them?
So please --
>> John Hessler: No desire lines?
>> There you go [laughter].
Here, we have a question down here?
>> This is for John.
The only question I really have is
how much is what you're doing rely
upon the accuracy and
consistency of data over time?
>> John Hessler: Oh, it
relies importantly --
well, that's a good question.
The things are not
necessarily empirical.
>> Go to the microphone.
>> John Hessler: Yeah, the things
are not empirical in that sense.
We're not trying to, with
cellular automata; we're not trying
to prove a theoretical concept.
What you're doing is
you're attempting
to make simple transition rules
which give you a sort of idea
of what the mechanisms at work are.
You then run this thing,
see happens.
Does it suggest anything
to what the real data are?
You then go back and tweak it.
You then maybe change the
underlying transition rule.
The one we looked at, the Mayan
one, takes into account rainfall,
population, trade but all of those
things are these nested networks,
these coupled networks which
don't necessarily have any reality
to them whatsoever.
The quote I started out about the
fact that we're in age of simulation
where we're using the non-real
to simulate the real is
really what these are about.
They'll get better with
time as algorithms increase.
I think the most interesting part of
this sort of research now is really
in network coupling and how
these things interact together.
You know, taking two
different types of networks
and tweaking them and
see what happens.
But the key now is that
we've got the computer power
to use all this huge data and
I think that is going to open
up cartography and geographic
analysis easily as much as GIS has.
Mark?
>> John, you started out by making
the point that static maps are --
well, they've sort
of done their time.
They're on their way out.
But I was thinking about
the situation, let's say,
with meteorological forecasting
ensemble modeling where you're not
so much concerned with random
probabilities as you are
with a variety of different inputs.
And so I mean, in that situation,
you actually run multiple models
and you see how consistent the
results are and then you can come
up then with one output map
where you have a single map
that shows a consistency
or reliability.
And I would imagine that
for the kinds of phenomena
that you're talking about,
something similar might be involved.
>> John Hessler: That's
exactly what happens.
You run a whole bunch of
different types of simulations.
You see if they're converging
into some sort of trend.
And then attempt to figure
out whether there's some sort
of underlying causality
that you can understand.
For meteorological maps,
causality is not as important
as predictability but yeah,
it's the same sort of thing.
Whether you actually
need to print the map
out though is another
thing altogether.
I think that's really the -- one
of the things I found interesting,
someone and I guess it must have
been Jim Ackerman talked about
or someone asked a question
about the road map and said
that you don't see on a
GPS what's around you,
whereas a road map shows
you what's around you.
And I found that interesting because
when I've got my iPad in the car,
I can see a lot more
about what's around me
than I can on a static road map.
And I think the map thing, the
paper map and the static map thing,
whereas I know other speakers
weren't willing to kind
of push the cart over, I am kind of
willing to push it over [laughter].
And really just because I think
we've come so far and if you look
at major map producers, USGS,
Swisstopo, all those sort of stuff,
the paper map is you know, print
on demand but how long is it going
to be before the print on
demand is no longer necessary?
I think in real solid geographical
research, I think the dynamic is
where we're headed, the
real time and I'm not going
to stamp on Doug's entire talk.
So I'll let him deal with the
entire space time integration.
>> John, over here?
Oh here.
>> I'll go, I'll go.
John, over here.
Here I am.
>> John Hessler: The
lights are like --
>> Yes, exactly.
Question, it would seem that this
cellular automata would be very
useful in things like
predicting water levels rising due
to climate change at
various locations
which is actually a fairly
complex issue, I would imagine.
>> John Hessler: All that
sort of stuff is really --
cellular automata are being used for
all of that kind of modeling now.
Everything from glacial melting
to water rising to anything
where the difficulties
with the other two types
of simulations are kind of apparent.
It's something that a
certain extent is a toy.
I mean, you really can think of it
that way because there are a lot
of them out there which
are in fact, toys.
But the ability to generate
complexity out of nothing,
out of simple, simple
rules, the rules that go
into these are incredibly simple.
But then when you go through
several thousand iterations,
you'd begin to generate
stable patterns.
You begin to generate equilibriums.
You begin to generate things
which aren't predictable.
The one thing that's interesting
about a cellular automata
because there's no equation,
you actually have to run it.
You can't say, "Oh, this is the
way it's going to come out."
You actually have to run the
thing to see how it will come out.
And the spatial distribution
of the beginning points
is so critical to it.
I have a sensitive
condition, sensitive dependence
on initial conditions that
lives deeply in these things.
And they so much mimic the
complexity of the world
that they are a powerful
tool for a lot of that.
>> John, I was thinking you
were going to bring in Ptolemy.
You know, it just -- we've
been waiting for Ptolemy.
>> John Hessler: I guess I can.
Okay, okay.
The Ptolemy was -- yes, the last
line in my abstract was the end
of the Ptolemaic project.
And really, what I meant
by that and I probably --
you noticed I skipped the slide
because I was running
low on time there.
What I really meant by that
was the fact that the idea
of producing an accurate map of the
surface of the Earth is completed.
We can pretty much do it at will
at any scale and I think as far
as that aspect of the cartographic
enterprise, we are tweaking
but that's really what
I was going with that.
And then I was going to
talk about this concept
of vagueness in these models.
But that was the Ptolemaic point.
>> So that's it.
Is this just another iteration
of the great predilection
for watching things that move?
Why don't you put this in the
context of decades and decades
of animations and simulations
and world's fair things
and so on and so forth?
>> John Hessler: Well, I
think to some extent --
>> And specifically, one thing
I remember from Waldo about that
in the early days of computers
with arrays of vacuum tubes,
the fascinating thing of
just watching the pattern
of glowing lights shift
and change which was in it,
in itself really meaningless
but it was fascinating.
And everybody would go to
the back of the console
and looked at the vacuum tubes.
Isn't that really what
you're doing here?
>> John Hessler: Well, I
think to a certain extent,
if you think about human
predilection for motion, yes.
But that's not what
I'm talking about here.
One can do a sociological study
on one wants on the concept
of moving mental maps perhaps as
opposed to static mental maps.
But that's not what I'm doing.
The historic stuff I think
is really interesting.
And as you well know, having
worked extensively in this period,
there's still so much to be done.
We haven't touched the surface of
the archives, touched the surface
of the sources, the ephemera, even
the stuff from the Harvard Lab is
so ephemeral and the
stuff that underlies this,
interpreting it is just
a fascinating thing.
It's one of the dangers, I
think, of actually talking
about 20th century cartography to
an audience of the history of people
who are in the history of
cartography because the methodology
that has served us so well in
the past, in the Renaissance
and the 19th century is not
applicable always directly
to the 20th century.
The 20th century is far more
technical, far more mathematical
than some of the earlier periods.
And there is just no secondary
sources with which to rely
on context for the most part.
Those things are being produced.
The articles that were in Mark's
book, the thousands of them, I mean,
a great many of them or
most of them were based
on primary source research
that every one
of those authors had to dig into.
So I think the 20th
century right now,
modern cartography is a special case
which our methods are
not quite as clear.
>> We have just one
more question here.
We've run out of time, right
here actually, right now.
>> John -- I think I had
the microphone first.
Sorry [laughter].
Since gee, I was going to pass over
what you had to say about paper.
But John asked my question about
Ptolemy so I'll go back to paper
because I think they
relate a little bit.
I think, well, first of all,
you are sounding more and more
like a positivist the deeper
you go into this conversation.
>> John Hessler: Yeah, it's true.
>> But -- and one of the arguments
against it and certainly as a way
of model of looking at cartographic
history is that one of the ways
in which complexities of
society, modeling complexities
of society presents a challenge
for historians of cartography
of modeling the complexities of
mapping is that people operate
at different functional levels.
And I think, I guess the best
analogy for that that I have
for this very dumb analogy
is the advent of air travel
or rocket travel did not limit
the necessity of the wheel.
And the jury is certainly out in
technologies that probably neither
of you or I can visualize at this
point about what is going to happen
with the way of paper maps.
But I think one of the --
>> John Hessler: So just to answer
your -- just to then you can --
to answer that point
but the cellphone certainly
eliminated the payphone.
The computer on your desk certainly
eliminated your adding machine
and your calculator.
There are things that disappeared
no matter what individual
functionalities are.
I do believe we, as
far as paper maps go,
I don't think it will ever
stop being produced but --
and I'm not talking about a
map that someone has, you know,
whenever you go to a new city,
you're going to want a little map
because you probably aren't
hooked into the Wi-Fi network yet.
But if you were hooked into
a universal Wi-Fi network,
you probably wouldn't pick
up the map at the desk.
You would simply look at your phone.
Certainly, it might
not be advantageous
for them to print them anymore.
So I guess there is a point where
I think there is a tipping point.
And we're not there yet but maybe
five years, we'll come back,
have another 20th century conference
and we'll see where we are.
So [laughter].
>> This is a much longer
conversation
than we could probably have on this.
>> John Hessler: It is [laughter].
>> I beg to disagree but --
>> John Hessler: Okay.
>> Thank you so much, John and Jill.
Ralph will introduce the
final speaker, our keynote.
[ Applause ]
>> I have one point on this.
Those of us who attended the
Washington Map Society meeting last
night heard the Director of
the Mapping Unit of the NGA,
the National Geospatial
Information Agency.
He's dealing with the same issue
and the pushback he's getting from,
as he terms it, the warfighters
on the frontline in Afghanistan is
that a map with a bullet hole
through it can still be used.
But a bullet hole -- an iPad with
a bullet hole through it cannot.
So they're dealing
with the same issue
and maybe they'll come out,
find a way to solve it.
Mark Monmonier opened our
conference with an overview
of 20th century cartography examined
in particular overhead imaging
and dynamic cartography.
Our final keynote speaker,
Douglas Richardson,
will close the conference
with a look at the present
and future geography
and cartography.
Dr. Richardson is ideally suited
for this assignment as he has been
at the forefront of new developments
in these two fields
for the past 30 years.
He is known to most of us
as the Executive Director
of the Association of
American Geographers
where he has led a highly
successful organizational renewal
of the association and has built
strong academic research publishing
and financial foundations
for its future.
But you may not be aware that prior
to joining the association some 10
years ago, Dr. Richardson founded
and was president of
Geo Research Inc.,
a private sector scientific
research company specializing
in geographic research
and technology
and first real-time
interactive GPS, GIS mapping
and data collection
technology which he patented
and is the world's first patent
for this technology which are now
at the heart of a wide array of
real-time interactive mapping
and large-scale operations
management and applications
in most major industries
and governments.
Please join me in welcoming
Dr. Richardson.
[ Applause ]
>> Douglas Richardson:
Thank you so much, Ralph.
It's a pleasure to be here and John,
you've organized a
fabulous conference.
I found it really stimulating.
I found it enjoyable.
And those are the two
characteristics that you like to see
in a conference so a
pleasure to be here.
Let me make sure I've
got my gadgets here.
Okay, I want to talk today
about the concept of real-time,
space- time integration in
geography and cartography.
As we've seen, space time
integration has long been the topic
of study and speculation
in geography.
However, in recent years,
an entirely new form
of space time integration has
become possible in geography
and geographic information
science, that is, real-time,
space-time integration
and interaction.
I'm going to talk a
little bit about the --
be a bit indulgent because this is
a historically oriented conference.
I'm going to talk a bit about
the origins of real-time,
space-time integration in geography,
geoscience and cartography.
The pivotal development enabling
real-time, space-time integration
in GI science, et cetera was
the invention and development
of real-time interactive GPS,
GIS functionality in systems.
This occurred in the
late 1980's and 1990's.
As we have seen, time is
a multifaceted concept
with deeply personal and
subjective meanings as well
as formally constructed but
still arbitrary meanings
and the ontologies for time in
GIS science can be as complex
as they are for space and geography.
However, at its simplest
and in the practice
of geographic information science
research, it is still useful
to think about GI science
research challenges
in three basic temporal domains.
Those are past time which focuses
on historical GIS or documenting
and interpreting historical
events, places,
processes, maps and so forth.
There's future time, we've seen
a little bit about modeling
and simulation, projecting
future spatial scenarios.
And then there's real time,
understanding and engaging
with events, people,
places and spatial processes
as they are emerging,
evolving and interacting
in a dynamic and mobile world.
It is this latter form
of space time integration
that I will address
today in this talk.
While of course interrelated,
each of these simple categories has
its own set of special challenges
in gathering, understanding
and representing space
time integrated data.
Historical time and
geographic data especially
from the more distant past is
highly constrained by pre-existing
and frequently indefinite
or inconsistent time
and space categories
of bygone years.
Historical GIS, for example, cannot
ignore and must struggle to work
within and reconcile the profusion
of its received categories
of time and of space.
Now future, on the other hand,
future time GI science work is
less restricted in the sense
that its temporal and spatial
constructs are not necessarily
constrained by past
categories or practices.
So it can more freely be defined as
the researcher or a modeler wishes.
Its delineations of space or time
need not be constrained by those
of a world that is or
already has occurred.
The possibility, moving to real
time, the possibility of autonomous
and continuous real-time,
space-time geographic data creation,
integration and use as one travels
through space is a relatively new
and thoroughly transformative
development for geography,
GIS and GI science, as well as for
science and society more broadly.
The real-time interactive
GPS-GIS functionality at the core
of this transformation is
also largely responsible
for the explosion of spatial
temporal data being generated today
in great proliferation.
And it is the underlying
breakthrough enabling many new
geographic research
initiatives and myriad real-time,
space-time integrated
geospatial applications
in governments, businesses
and society.
By the mid-1980's, again back on
the origins, by the mid-1980's,
geographic information
systems had developed
into very useful cartographic
tools for computerized storage,
manipulation and graphic display
of multiple thematic layers
of location data and
attribute information,
as well as some analytical
functions.
It greatly improved the efficiency
of making and updating maps compared
with non-digital processes.
However, GIS alone was
a post facto system.
Its locational feature and attribute
data were laboriously gathered
and integrated within the system
well after it was collected.
Neither new or existing GIS data
was fused with or interacting
in real time with spatial
temporal data
or with the dynamic
real world around it.
In the mid-1980's also, GPS had
come along in the early 1980's
and my firm was a very early
adopter of that technology.
But GPS alone provided location,
not surrounding context.
We've heard an awful lot
about surrounding context
in this conference.
And I agree.
It is crucial.
It is, in fact, the crucial
component of location
if you don't know the
context around you.
And if you don't know the
context around you in real time,
it's very difficult
to do a lot of things.
As Ron Abler noted in 1983 at one
of the conferences I organized
on this topic, GPS
equipment will tell me
where I am with great precision.
But knowing precisely where
I am may not be very helpful.
Location, no matter how
precisely specified,
is sterile in and of itself.
Context determines whether
knowledge of location is invaluable.
Contextual knowledge
immensely enriches the value
of locational information.
But the real-time integration,
real-time GPS/GIS integration
which I'll refer to henceforth
as RTI GPS/GIS enabled
for the first time a
continuous and mobile creation
of highly accurate real time
and tightly fused geospatial
location attribute
and temporal data.
And its simultaneously
interaction with GIS maps and data
for interactive real-time use,
for example, research mapping
and societal applications
immediately as one moves
through experiences and
interacts with the world
in one's own environment.
The early conceptual
and technical challenges
to achieving this real-time,
space-time interactive
functionality were formidable.
Today, this capability
is ubiquitous.
It's in every cellphone.
It's in every navigation
device that you might use.
It's in a lot of our
consumer products.
But just like so many things, at
one point in time, it did not exist
and it had to be created.
It had to be generated and
thought through and put together.
The fundamental conceptual, the
time line on that is as follows.
The fundamental conceptual,
ontological, scientific
and technological research
necessary for the development
of the first RTI GPS/GIS systems was
largely accomplished during the mid
and late 1980's.
Proven technologies embodying RTI
GPS/GIS functionality for real-time,
space-time geographies
were firmly demonstrated
by the late 1980's primarily
by my firm, Geo Research.
Initial commercial systems came
into increasingly widespread use
for research and a wide
array of government, business
and educational applications
during the early 1990's.
I'm not sure you can see this.
Can you see this reasonably well?
I hope so.
So the early research
focused on a lot of challenges
in putting together a core real-time
GPS/GIS system that could handle all
of these inputs and all
of the complex mapping
and cartographic issues as
well as real-time map updates,
data flowing from multiple sensors
and integration of that data,
both spatial and temporal data with
current GI systems as they existed
and the challenges of doing that in
the field and in real time with all
of the equipment involved.
The core GPS/GIS system involved GPS
maps and data interaction, creating
and updating GPS and GIS
databases in real time,
real-time spatial temporal
data creation,
real-time interactive user
interfaces, active displays,
feature attributes, ontologies,
project design, set-up issues,
data analysis functions,
differential correction issues,
wireless integration which was quite
a challenge at that point in time,
GPS and GIS QA/QC structures
and procedures,
and sensory inputs both
automated and otherwise.
The basic inputs were GPS
supplemented by dead reckoning,
cellular and laser range finding
offsets, all of which we do a lot.
The time inputs were from GPS so
GPS provides both location and time.
And then the mobile user
attribute information inputs
that were supplementing
the automated creation
of the digital maps,
the lines, points,
and polygons of a GIS automatically.
We also not only created the
locational features and structures
of GIS in the field in
real time but we allowed
for extensive attribute
data entry as well
and that attribute data was
also involved both spatial
and temporal data locations.
So it was fully integrated
into, directly into the GIS.
And that was everything from
manual keyboard, pen, touchscreen,
voice command, barcode
and RFID inputs.
We also developed a unique
module that was called the XDS
or external data source which was
designed to be an all-purpose module
for taking in automated electronic
sensor data from a whole wide range
of radiation instruments.
And those included everything
from air quality instruments
to radiation detection
equipment, biomedical sensors,
RF signal strength sensors.
Digital cameras were the
first to integrate all
of those sensors into
a GPS/GIS system.
Digital video bathymetry
and so forth,
those systems allowed essentially
then great amounts of data
about the areas being mapped
which and those sensors,
attribute data were automatically
and precisely georeferenced
and time stamped.
Going on the other side,
there was a very --
a lot of development
around active map functions.
In other words, real-time map
update and display, GIS maps
and attribute data linked images
such as digital photos, foreground
and background maps and interactions
between those, continuous updating
of existing background maps,
for example, as well as creation
of new maps as one
moved through space.
Moving map orientation,
lots of options.
There are aerial photos and
satellite imagery integration,
both as background maps
and as surfaces upon
which to create new maps.
Aerial photos, satellite imagery,
and then a whole raft of dynamic
and this was the -- there was
a quite complex whole range
of dynamic scale, map
projection data
and coordinate system
adjustments that needed to be made
and could be made in real time.
Coming out of that, then
subsequently was a whole range
of vehicle tracking roving systems
and fleet headquarter systems
with complex interactive
capabilities.
And then, so those are
the basic early systems.
All of this was in place
by the early 1990's
and it's been reinvented
and reinvented and copied
and so forth ever since that time.
But this functionality was
developed by the early 1990's.
Coming out of this is a whole range
of what I call geographic management
systems in terms of applications
and this is the whole world
of what I call either geographic
management systems or mapping
to manage and these are all of the
sorts of organizational activities
that enterprises now do,
work order processing,
peer-to-peer communication,
real-time map data
updates and so forth.
And what's happened and what
it really hasn't been addressed
in this meeting is the fact
that now, we're not just mapping
to as we used to with the post
facto data collection for creation
of a map, even if it's
within a GIS system.
And we're not only
mapping for decision making
which has been a big function that
has been much appreciated and used.
But even beyond mapping for
decision making, what we're seeing
with these technologies
is day-to-day operations
and management of organizations.
So not just post facto
decision making
but real-time interactive management
of operations of large companies,
government agencies,
health researchers and a lot
of research programs and projects.
So it's this real-time,
day-to-day management that is going
to be what you see an awful lot
more of in the coming decades.
It's quite; it's already
quite established
in the last decade or so.
But it's something that we've
been working on with many,
many clients for many years.
And on the other side, we'll talk --
so I'm going to talk about some
of these management systems first
and then I'll talk about some
geographic research outcomes.
So new research frontiers
generated by RTI GPS/GIS
and geographic management
and applications.
These early on were patented by
our firm and we developed patents
which we then licensed to some
of the GPS companies and so forth
with international patents.
The title of this was interactive
automated mapping system.
So I urge any of you who want to
follow this more to really go ahead
and take a look at this patent.
It's quite broad.
It's quite deep.
It's quite well thought out
and it covers a whole range
of what's been done over
the last 30 years now.
And now, it's everywhere.
Real-time GPS/GIS is
interacting with the world
in so many different technologies
that it's almost become invisible.
The other aspect of this I'd
like to discuss is what is the
compelling experiential immediacy
of these technologies.
It's also increasingly --
they're increasingly mediating
and changing how individuals
interact with
and experience the
world and each other
than one can say for
better or worse.
But the fact is, it
is making changes
and those changes have
opportunities for some
and they represent barriers
and filters for others.
So it's not a clear cut system
whether this is useful or not
but it is in fact what has come
out of this particular technology.
The [inaudible] from a cartography
perspective that we could now,
based on this real-time
interactive GPS/GIS technology,
we could map directly on the
face of the Earth in real time
with this technology is to me a
profound innovation in the history
of cartography and I don't
understand why it has not been part
of generic cartographic thinking in
terms of changes that have occurred.
For me, I can't think of
a more far-reaching change
at how maps are produced.
Part of that has to do with
something we will talk about later
and that is the degree of innovation
that now occurs in geography
and GI science within
the private sector.
And for many years, geography
did not have a private sector
research component.
Almost in fact, when I went
to grad school, geography
and academic geography were
practically congruent categories.
In other words, there was not a lot
of high-end research in the field
of geography in the
private sector like there is
in so many other disciplines,
for example, computer science
or chemistry with Dow Chemical and
DuPont or so that form the genetics
with all of the high-tech
genetic firms and so forth
where they are supplemented
and interactive
with a very strong
private sector research.
That's only the private
sector research component
of geography has really only come
into its fore from about 1980 on.
And when I founded my company
in 1980, I was the only company
who had a Geo-something name to it.
Now everything is Geo
fill in the blank.
But I think that's something we
need to think about in cartography.
We need to think about as we want
to advance our research frontiers,
how we can interact with
university, private sector
and public sector more productively.
That sort of integration is involved
in all of these kinds of things now
and this particular graphic you
see not only all of these roads
which were mapped by real-time
GPS/GIS processes as people drove
down the roads and took in
the addresses and took in all
of the other attribute interests
there as well as all kinds
of activities from
plants and so forth.
So then this began in the early
1990's or late 1980's for us
but to be put more into a
productized thing rather
than the research software and
systems that we had developed
that were wildly popular
during that period of time.
We started to productize that
into a system which we called --
we used the term, oops, we used
the term GeoLink for the system
and the product that we developed.
And it's pretty much like you see
here and this is a sort of stripped
down version of some
of the functionality.
It began to be adopted greatly.
This is for maps.
This is a mapping of highways
in Yellowstone Park very early
in the 19 -- late 1980's,
lots of press and so forth.
Highway maintenance goes
high tech, GeoLink software
and then GeoLink unites GPS
and GIS and I don't know
who that guy is but [laughter].
>> Yeah.
>> Douglas Richardson: I
don't recognize him now.
So but there are other
kinds of press
and then early 1990's,
building maps on demand.
Seldom in our lifetime
does a product come along
that revolutionizes
the way we do our jobs.
And then another -- as we move
forward, it became clear to us
that this was greatly democratizing
and spreading the ability of people
to map and so one of our
slogans was GeoLink moves mapping
into a whole new field, yours.
And this is part of
not only the citizen
and participatory GIS
activities that are much discussed
but a key development here
is that those with domain
or substantive expertise
in a particular area
that is being mapped could now do
the mapping directly themselves.
So for example, if you're a
botanist and you are trying to map
out endangered plant species, you
didn't need to bring a mapper along
as well as yourself to
identify where these were.
You could do field work
directly in the field and map
out your own endangered plant
species or other kinds of things.
Same is true of a utility
lineman who knows what's
on the poles and so forth.
And we did a lot of large
projects for utilities
but we trained the linemen
eventually to do it themselves.
And they knew more than we did
about what was on the pole.
They could accurately [inaudible]
about those 33 attributes
that we were mapping out for some
of these large utility companies.
Same thing with transportation.
Transportation specialists could
map out all of the thing on the road
from the side rails to all of the
stuff that had names on it that none
of us really know or
quite understand.
So moving this mapping into
the domain, into the realm
of domain specialists was
a very big step, I think,
and one that we haven't
acknowledged much in terms
of the way cartography is going.
Now that's a threat in some ways
to professional mappers
and cartographers.
It's certainly a threat to
surveyors and they've been trying
to every way they can to sort
of maintain some kind
of monopoly on mapping.
But it's just not going to happen.
It's not their concern.
We've had to fight those battles.
Also, so the applications that
we generated were just everywhere
around the world and in every
industry you could imagine
from agriculture.
This reminds me also of a system
they gave for Westinghouse to map
out all the radiation concentrations
at the Hanford Nuclear site
with the XDS and the automated input
of radiation detection equipment.
It was wrapped up on sensors and
an array of sensors similar to this
but this is a precision GPS
agriculture application that we did.
Lots of people are already
on and these are all
from the very early 1990, 1991
era and you can see the kinds
of companies that were using
it, Woodward Clyde Environmental
and Engineering Consultants.
This is, by the way, a
system that we worked with --
this is something that it
was a Motorola GPS receiver
that we did all the software
and the integration GPS/GIS
integration development work
early on.
So I'll just click
through these very quickly.
U.S. Department of Interior, Bureau
of Land Management was using it
for forest fire activities
as well as lots
of other activities
early on 1990's, 1991-92.
California Office of Emergency
Services, this is Dave Kehrlein
who was quite an innovator.
Then we developed additional
structures here
around particularly this whole
concept of vehicle tracking.
So we integrated all of the data
collection and mapping into sort
of movable peer-to-peer
systems that could be organized
for data collection, data
base updating in real time.
And then management of operations
from those updated
systems in real time.
And so we had fleet
headquarters systems.
We had vehicle operator
and individual raft
of individual vehicles in the fleet.
It could be managed and it
allowed them to talk back
and forth plus they all relied
on the same set of functionality
for updating their maps and
gathering data, creating new maps,
creating new attributes, creating
new line and points and polygons
for streets or whatever it may be.
Florida Power and Light, you
know, all kinds of people and part
of the great points is that people
don't realize now is how much this
technology has made mapping faster,
easier and less expensive
than ever before.
And that's why so much more
mapping is now going on.
Lots and lots of environmental
and disaster response activities.
We were the only one
that FEMA was calling
on for many years including during
the Mississippi floods of '93
when we mapped out with helicopters
all of the people who were stranded
on rooftops and so forth.
And then shortly thereafter
when the floodwaters receded,
we were hired to drive all
through all those streets.
They were very muddy
and smelled bad.
Mapping out the extent of
damage on all these structures
so that they could get a
quick, accurate inventory
of what the extent of damage was.
And then thirdly, after that
was fixed up, we went through
and were hired to map the
thresholds of all of the structures
that were there so that the flood
FEMA could understand whether there
should be flood mapping.
Well, flood mapping was done
but so they could understand
where their flood insurance
was feasible or not,
or rebuilding was feasible or not
in some of these neighborhoods.
Early on, these kinds of automated
things were the digital photos
automatically attached
to the household location
which was determined
not only from driving
but there was a laser range finder
that were more sophisticated
than this where from the vehicle,
we could take an automatic offset
of the distance from the point,
the GPS point in the vehicle
to the distance to the house and
its bearing and even its elevation.
All of this an automated process.
Utilities, the Army Corps
of Engineers were using
this really early on.
They still do.
We did, this is an article from
1990 in which we counted some
of the international projects that
we're doing that range from all
through the ASEAN region, from
Kathmandu to Peru and Argentina
to Soviet Union and throughout
Europe and parts of Africa.
We had projects in every
continent doing this kind of thing.
Mexico, Canada.
We also developed software that
worked for the precision grade GPS.
Most of it you could -- most of you
know there are two different grades
of GPS, the civilian grade which
is a little bit less accurate.
There's a military or
Defense Department grade
which is much more precise.
So we were hired to do the
software and develop software
for these so-called plugger units.
So those were [inaudible] kinds of
systems that we developed for those.
Here's an example of a project
in the Philippines in Subic Bay.
When the Navy abandoned
the Subic Bay Military Base
in the Philippines, they
gave everything back
in pristine conditions.
So we were hired to go in
and inventory everything
on the base including the equestrian
facilities, the horses, everything
and to the government's credit, if
anything was broken that we found,
they went back and fixed it
up before we gave it over.
Of course now, we're thinking
about maybe going back to that base
and of course [laughter], I don't
think they would be given back
to the U.S. government so
there's probably another price
to pay if that were to happen.
We were in the early '90's, this is
what, '90, a little maybe a couple
of years later, mapping at
Red Square with our systems
and nobody knew what we were --
we weren't working for the CIA
but we were collecting data.
We were hired to collect data
around Chernobyl but we did a lot
of sightseeing and mapping
around Moscow and other places.
And at that point in time,
people had no idea what it
was and yet we were there.
We were, that's one of the
timeline when we were employees.
We also were early adopters of RFID
which now is being
rediscovered every year.
We integrated these systems with
these RFID systems for cattle,
for agricultural research
systems and for the fed --
the USDA agricultural
research service.
And we had sensors for
weather and we did a lot
of climate air quality stuff
already and we integrated a lot
of those instruments but we also
then integrated the ear tags
for the RFID tags and we
had automated water monitors
and automated sensors on the
weight and all that sort of thing
and then we radioed all
the information back in.
So those RFID integrations occurred
as well in the early 1990's.
But my point is, I guess,
not to talk about what we did
but it's always so -- you know,
if you listened for example
to Google Talk, Google whom I have
lots of friends there but you know,
their view of the world is
that nothing happened before
six years ago when they got
into the mapping business.
And of course, all of their maps
were generated on the ground
by companies that used
real-time interactive GPS/GIS
to create those databases
in the first place
and they're now using the same
thing to drive through streets
with the automated
cameras and so forth.
It's amusing to hear
them go on about this.
We did a lot of education
and conference work.
I'll just point on this
as a conference thing.
Some of these people,
many of you would know
so maybe I can just step up here.
Well anyway, they include,
boy, it's hard to read this.
But anyway, they included
Jack Dangermond
and Dan Kotter who's still kicking
the old -- let's see, anyway,
I won't go now through
all of the names.
There's Ron Abler, of course and --
but quite a few folks who many of
you worked with many long years
in mostly federal agencies.
We did a large [inaudible]
with Joe Morrison who's here
for the Census Bureau to help them
tool up from a research perspective
for conducting the census with GPS.
So mapping to manage was the concept
that we had for these technologies
that creates maps and so forth
that we had both research focus
and a kind of an applications
focus and I think that the --
it's the management of operations
that has become so pervasive now
in the use of these technologies,
both personal consumer applications
and these company applications.
So coming back to this, we talked
earlier just showing you a set
of applications flowing out of these
technologies that are in this realm
of mapping geographic
management systems.
And now, I'd like to talk
just for a minute about some
of the applications and new research
frontiers that have also flowed
out of these kinds of functionality.
And I'll give this an example,
one that's probably
less familiar to you.
You're probably familiar with
the environmental research,
with transportation research,
with a lot of socioeconomic data
collection that's going on now.
Social sciences are all very
much picking up on RTI GPS.
But we're doing a lot of work
at and I'm doing personally,
and [inaudible] doing a lot of work
now on health research with NIH
and CDC and many other
organizations.
And this also depends on
the fundamental premise
that we've always has driven
our research which was
that research agenda systematically
incorporates spatial data
and analysis into global
health research,
hold extraordinary potential
for creating new discovery
pathways in science.
This came from an article
that I published with a number
of institute directors from NIH and
some leading GI scientists recently.
More on that in a minute.
So strategic initiatives
that we've been engaged
in over the past decade at the AAG.
Since I've been doing research my
whole life, when I came to the AAG,
instead of treating it like a
normal scholarly association
and doing meetings and publications,
I started getting a lot
of research projects and now
that keeps us all interested
in staying there but it's something
that we do well as an organization.
And so especially on
cross-cutting issues
that benefit the discipline
as a whole.
So we did develop an NIH initiative
for an NIH-wide GIS infrastructure.
When we saw how fragmented
and partial they were
and the odd thing is that the world
of health research
involves biomedical research
who are very smart.
They're very intelligent people
and they own those statistics
so they're not afraid of numbers.
And yet, it's the one domain,
one large domain that is least --
been least penetrated
by GIS and GI Science
and the only reason I can think
that that is so is that for so long,
the gold standard in health research
has been the longitudinal study.
It's been changed over time
and to get the idea that change
through space and change over --
you know, spatial change is
equally important in terms
of understanding the ideology
or the cause of diseases.
It's their treatment, the
diffusion of disease and so on.
It's been a difficult task but
now we've, I think succeeded
because we're getting dozens
of instead of trying to knock
down doors and explain that this is
important, we're now getting lots
of our fees coming out
of lots of people coming
to us saying, "What do we do?
What do we need to do on this
big study or that big study?"
So we did this with the
Cancer Institute and NIDA.
We did another workshop called
NIH-Wide Geospatial Infrastructure
with them.
We did publish a report
that's here called
"Establishing NIH-Wide Geospatial
Infrastructure for Medical Research,
Opportunities, Challenges
and Next Steps."
Geospatial Frontiers in Health
and Social Environments,
a project that I was the PI on it.
It just wrapped up.
Other participants are named there.
And coming out of these things, the
workshops that they really wanted
and the workshops that turned out
the topics that rose to the top
of the pile were spatial temporal
analysis for health research.
And again, that was the whole
response to this plethora
of new spatial temporal
data that relates to health
that can now be used to understand
disease better and look at all sorts
of interactive environmental health
kinds of issues that are out there.
Practically, every
institute at NIH can do this.
And of course, there's a lot of
need in terms of infrastructure
for integrating existing large
studies as well as small studies
that are going on so that health
researchers can have access
to other data and other information
pertaining to health, that is,
and geospatially referenced.
And we got a lot of
things in the topics
around geography, GI
science and health.
We're all looking at new research
particularly in social environments.
There's a great deal of interest
by NIDA, the National Institute
of Health's National
Institute on Drug Abuse.
Nora Volkow who's the director
of that and I are good friends
and she's -- we've done
a lot of work together
and she's also coincidentally the
granddaughter of Leon Trotsky.
But it has nothing to do
with her current work.
And so to that, we did a science
article called, "The Spatial Turn
in Health Research,"
that I authored together
with Nora Volkow whom I just
mentioned and Mei-Po Kwan,
one of the leading GI scientists.
In fact, in my mind, perhaps the
most transformative geographer
practicing today.
Robert Kaplan who's
the director of OBSSR,
the Office of Basic
Behavioral and Social Research.
Mike Goodchild who most of
you know, leading GI scientist
and Robert Croyle, who's
the number two person
at the National Cancer Institute.
So this has caused a great deal of
all medical, biomedical researchers
around the world who
read Science magazine
and this article has generated
a great deal of interest
for geographers and GI science about
the potential for using these kinds
of analytical approaches,
methods and data generation.
And the key areas that come
out of all these again,
that they touch back on is the
spatiotemporal data explosion
in health research,
real-time interactive GPS
and real-time space-time
interaction in GIS and GI science
as fundamental underpinnings
of the work that they're doing.
That work includes work on
gene environment interactions.
As you may know, just about every
institute at NIH does a lot of work
with the human genome now.
And my daughter just wrote a
very incisive, critical review
of the fallacies of using
genomics carelessly.
And I think you're going to
find the next big sort of error
of discrimination is going to
be genetic or genome-based.
Laura, take note of this, but
it's something that will, I think,
you'll find something that we
have to look at very carefully
because there are all these
epigenetic and of course,
environmental interactions
with genes
and this has brought back the whole
human gene interaction debate.
The title of my daughter's book,
by the way, was called Sex Itself:
The Search for Male and
Female in the Human Genome .
And she's looked at an
awful lot of muddled science
and incorrect assumptions
and research that's been
derailed as part of that.
Exposures research is another big
area we're working with them on
and that includes looking at these
-- at individual scale exposures.
So these are GPS that this
dimension on the diagram is time
and these are movements
over time of people.
This is a base map
and then these are --
these different planes above
it represent the spatial extent
of risk factor concentrations.
So risk factor concentration
over one and this could be
of various pollutant or risk
factor concentration over two
which can be either a later
date of this one or it can be
as a different type of exposure,
whether it's a social
environment exposure
or whether it's a physical
environment exposure.
So this kind of work
relies on real-time GPS/GIS
and is in fact, enabled by it.
It's also driving in terms
of scale, there are --
I don't know if there are
laws of scale but there are --
right now there's a great deal
of new enabling of work --
research through the great plethora
and detail of geospatial data to go
from -- go down to the individual
level scale in health exposure
and then up to the -- say, the
citywide or population level.
So I'm doing work particularly on
how you can scale back up and down.
You come down to the individual and
then how do you generalize from that
and go up to the population, and
working with MIT on the citywide
and working with scholars
like [inaudible]
on the individual exposure
level is fascinating.
Health disparities, really
important area, and I did a book
with [inaudible] and it's called
[inaudible] Drug Addiction
that looked at the relationships
of this using a lot of work on GIS
as well as there's other
aspects of geography.
Finally there's data
confidentiality, I have a project
with NSF looking at what are
the unique confidentiality
characteristics of geospatial data.
And it is different
from say, social data --
social science or other data.
In other words if you visualize
social science data you end
up with a chart or a
graph or a histogram,
something like this that's the way
in which you visualize that data.
Well if you visualized
geospatial data, you have a map
and that map is a unique
visualization
that is much more difficult to deal
with in terms of confidentiality,
because most of us if we take a
look at a map of Washington DC,
even if there are no road names,
even if it's been distorted
to some extent and masked and so
forth, we're still likely to be able
to find patterns, recognize the city
and recognize certain
features on that.
So I'm working together with
ICPSR, The Institute for Social
and Political Research at University
of Michigan, and the PI [inaudible]
in Georgia, also who
heads up that group
as a Co-PI as well as others here.
Really interesting, I also run
something called the National
Geospatial Advisory Committee and I
foolishly took on the chairmanship
of the Geospatial Privacy
sub-committee and if any
of you have seen the amount
of material courses --
an issue I have been concerned
about and you know, following ever
since I got involved in this,
because it's inherent in what I do
or have done, and what
we all do now.
So -- but the Obama administration
just put out two large reports
on privacy, if you haven't
seen them yet you will,
because it'll be circulating around.
One was called the Podesta
report which looks at a lot
of policy issues related
to big data and privacy.
Geospatial data is inherently
big data for the most part,
and then the second one was put out
by PCAST, the President's Council
on Science and Technology,
and it's all
about the technological
approaches to trying to deal
with privacy related to
particularly locational privacy
which concerns all of us here, but
other types of privacy as well.
So, so again I just want
to [inaudible] not only do these
technologies have great implications
all across the border for societal
organizations, institutions,
universities, companies, government
agencies, they also have a lot
of input for research
and advancing research.
They create so many new research
frontiers for us in geography
as well as in other disciplines
that is hard to ignore.
So we're looking now at the
team we've got together looking
at geographic context, health
research infrastructure,
geospatial data confidentiality,
scaling up work, and there's people
from MIT and people from University
of Illinois, people from Harvard
and the University of Miami as well
as others that many of you know.
So this is a broader approach
that we're taking to these issues,
so in any case I'd like to then
just take a few minutes and talk
about what I call real-time
space time fault lines
in geography and GIScience.
And from my own personal experience
I've seen the impact that RTI
in real-time -- the impact that GPS
GIS has on the discipline as well
as in the way we conduct our
research, the way we do our work
and the way that others
interact with us.
There's now a huge realm of
geographic practice in the world,
in society, in government agencies
and in private sectors and so forth,
and the generalization
of that is one
of these where I call fault line.
So I think it's created completely
new possibilities within the realms
of GIScience and GIS for
interacting with and representing,
and making decisions about, and
negotiating the world around us,
as it is encountered and
experienced in real-time.
These fault line signal
these continuities
between possibilities prior to and
after the introduction of disruptive
and transformation -- transformative
technology such as RTI GPS.
In the case of the
development and introduction
of RTI GPS GRS functionality, what
are the kinds of transformations
that have occurred under
disciplines of geography,
cartography and GIScience?
And we've talked about many of
them so I won't repeat those,
but certainly one is that these
functionalities transform mapping
and cartographic methods
thoroughly worldwide,
by enabling the collection and use
of geographic information that's
far more detailed timely, accurate,
immediate and specific to
a particular application
than was possible before, it's
the way maps are now made largely.
The other major use -- another
major innovation area of course is
in remote sensing and mapping,
but it's also largely responsible
for the explosion in the amount
of highly detailed specific
and updated feature and attribute
in spatial temporal data
that's now available
for GIS applications
and GIScience research.
It's also enabled the
real-time integration into GIS
of mobile electronic sensors, such
as environmental pollutant monitors,
digital cameras, bathymetric
instruments, noise monitors,
biometric health sensors which
now lots of people are running
around with for better or worse and
so forth, resulting in vast amounts
of new georeferenced and
time coded sensor data.
And those all rely on real-time
interactive GPS GIS as the front --
the core functionality in which this
-- the signals are entered into.
And it's also a core component
in the management of day
to day real-time operations
as I mentioned
that most governmental
agencies and large businesses,
here's a way for geographers who --
you know, I tried to get my tenure
at the AG, I was trying to
get Jaguar to stop bemoaning
about how misunderstood we are
and people don't understand us
and they're closing down
departments and so forth.
My answer to that is, you
know, if they don't understand,
it's up to us to tell them
first of all, but secondly,
these technologies, new technologies
that are revolutionary like this,
when they're integrated with
geography's traditional strengths,
of inter disciplinarity
of integrated science,
of [inaudible] focus, when you
integrate those new technologies
with our traditional strengths,
you have an explosively powerful
explanatory set of tools available,
and for so long many people have had
to push one or more of these away.
And I've really made this a
major focus of my time at the AG,
two things, one to really try
to reconcile the new changes
with the traditional approaches,
feeling that there's synergy there
and that both benefit by that.
Second major area is trying to
reconcile in terms of research
and teaching and everything
else, the two --
probably the major innovations
in the field of geography
over the last few decades, those are
the revolutionary new technologies,
and secondly, they're critical
theory and critical approaches
to research questions and to our
practice of geography and so forth.
In my opinion those are also highly
synergistic and we're starting
to see instead of the opposite poles
approach that existed a couple --
you know that [inaudible] so
we're starting an awful lot
of integration, an awful lot of work
between GIScience people
and critical theory.
In fact critical theory is informing
the questions and assumptions
and concepts of data and space
and time in a relativistic element
and uncertainties as well
as the social context
in which these are all carried out.
so I think it's very
important that those two --
we see those two as reconciling and
-- because they certainly do inform
and benefit from one another.
So I'm going to go on really quickly
now, we're getting close to my time.
Of course another fault line I won't
dwell on, but it's the introduction
of this -- and this technology
is really the enabling technology
for broadening mapping participation
to include not only
those directly involved
in the subject matter
that's being mapped.
The botanists, the environmental
scientists, the lineman,
the archaeologist et cetera,
or community activists,
human rights organizations,
I chair the AAAS, Science
and Human Rights Coalition,
and human rights organizations
around the world are very interested
in what these technologies
and what geography can do
to help provide evidence
of human rights abuses, because
of course whenever they occur,
everyone denies that it existed.
So we're working with
remote sensing and with
on the ground GPS GIS mapping
to try to document some of these
and then also doing a lot of
bibliographic work in that area,
but anyway there are lots and
lots of ordinary citizens involved
in this that are -- who are
doing great work in Detroit --
from Detroit to Bangladesh.
And I think they are involved
in some of those projects
and finally there are --
with the flows of data
that were now seeing there's the
so called volunteer geographic
information and crowd sourcing kinds
of applications that we're
seeing grow much more rapidly.
And those also depend
on our RTIG PSI GS.
So future Trans and research
challenges, I'm not really going
to go through this, I'll leave a
couple of handouts at the back.
One is a science article I
referenced on the Spatial Turn
in Health Research, the other is
an article that addresses some
of these issues in a
little more detail and some
of these research challenges
are listed there.
So I would encourage you
to take a look at that.
I'll just maybe pick out one
or two and this explosion
of real time data is
often integrated
with distributed environmental
sensor systems, remote sensing
and crowd sourcing
in the development
of more sophisticated tools and
methods for analyzing, modeling
and visualizing spatial-temporal
data at the scales ranging
from the everyday to the
life course and spatially
from the micros scale to the global.
It's all happening and we've got --
we don't have smart people working
on these, we need more
of you working on it.
We don't have enough
horses to pull the cart
but there's a tremendous
amount of work to be done
in these areas that's very fruitful
and could be very beneficial
for society at large, of
course to other people as well.
So future trans -- continuous
advances we're seeing
in computing mobile
wireless technologies
and these have all made changes but
they don't -- so they enable us.
So anyway temporal scale
I'll just mention --
we should address temporal scale
with the same degree of attention
that we pay to spatial
scale in geography.
We've got to seek integration across
domains in terms of our models.
With these frameworks,
ontology-based frameworks
they are likely to become
in my opinion key contributions
to knowledge about the practice
of spatial-temporal analysis.
Transforming geographic data
into meaningful information I
think that speaks for itself.
Ethics human rights and
human subject protection,
this is an area that important.
GIScience and geography
is an international
and interdisciplinary enterprise.
When we look at impacts on human
subjects we can't only be focused
on a national level such
as the IRBs and sort
of health related generated
civic protections.
Examining the impacts
of our research
on others is a necessary
component of research in our field.
Concepts of human subject protection
vary greatly from country to country
and among academic disciplines.
User interfaces, briefly we
should enhance our real time user
interfaces to match those
available for computer games.
Analytical tools a lot of questions
and a lot of research needs
to be done here but time constraint
decision making imposes less time
for analytical processing and
so we need other kinds of tools
like predictive analytics for
really super-fast integration
of real time activities.
Indoor extensions of GPS
GIS, there's a proliferation
of new techniques, now we
need to focus in on that.
Some of which we used Laser
Range Finders and RFID early
on for some indoors stuff but
it's more sophisticated now.
Computing large data -- so it
sets the whole big data issue.
I don't think anybody
wants to hear that now.
Policy and institutional
considerations again we must
confront and address
locational privacy
and data confidentiality issues
related to real time space-time data
from which much of these
privacy concerns result.
So these are key issues we need to
address, figure out how to address,
how to live with, how to
restrict or not restrict.
Find ways to deal with these
very real and important concerns
because I'm confident
that we will overtime,
not to everyone's satisfaction but
there's a lot of interesting things
that are happening in ways of
functioning in these areas.
Just a moment on -- so the
burden of this discussion
from these geospatial technologies
to what I call transformational
research
in geography science
technology creativity discovery.
I'd just like to make a few comments
on that based on my own experience
and feel free to totally ignore and
reject it but these new technology
and scientific innovations we all
know they're embedded in society
and their [inaudible]
relationships and so forth.
And -- but at the same time
they have very much transformed
geography, cartography and GIScience
and will continue to drive research
and applications well beyond
these fields, we see it already.
In here within such transformational
changes are interesting applications
and perhaps lessons regarding how
innovation in geography can occur
and how it might be understood,
natured and fostered in the future.
The process of innovation are
now intertwined disciplines
of cartography, geography, I
really hesitate to differentiate
between geography, GIScience,
cartography and so forth,
they are different endeavors but
they've -- there's a Venn diagram,
they all overlap so much that the
peripheral parts are more connected
to the center than they
are to the periphery.
But what is the actual role
of theory and its interaction
within the practice and
applications of GIScience?
These are questions that I
think many people I've seen come
into this general concept of
doing research with a very --
something I learned in high
school that you get the theory,
somehow the theory comes
from nowhere and then you get
from that applications and
then technologies and then
within technologies
and that applications.
And so in my experience that's
a far more integrated practice.
Somebody used the term mangled,
that's actually not a term
that Trevor Barnes
coined, a sociologist did
but there is a mangled --
there is a really mixing of day
to day experience, iteration
theory in the problem solving
that results ultimately
in the advancement
of science in my opinion.
So what are key -- how
is theory generated
in GIScience which is a new science?
What are key theories and
in what ways do they serve
or hinder innovation and discovery?
I think that it's the tendency
to decouple technologies
and applications from theory and
GI Science is not served as well.
I think that the research that takes
a closer look at the actual practice
of science, the ways in
which science and technology
and application driven research
interact and the actual practice
of GIScience, this has been done
in almost every other field.
It's long overdue in our conception
of what we think we
are doing in GIScience.
I think we also need to examine
past and potential future linkages
and synergies among public, private
university sectors in real time,
space time and innovation research.
It's the case -- it's a fact
that the research innovation,
the private sector in this area
has been long more advanced
than it has outstripped
that in the universities.
Certainly that's the case up
until maybe 15 years ago or so.
I think now it's more of
an equal contributions
to real cutting edge
research in terms of GIScience
in the private sector but
there's a tremendous amount going
on at the private sector that many
academics either know nothing about
or have a very hazy kind of
concerned idea about but a lot
of that innovation both needs
to be critically appraised
but it also needs to be taken
into account if one is trying
to move the science forward.
After all the term GIScience
was coined until 1992.
And the major focus of its early
decade was trying to describe
and attempting to understand
what had already happened
in the development of core GIS
and just geospatial technology
functionality and then to begin
to ascend this -- assemble
this into a science.
I think this is good.
I think it's useful but I think
it's in oddity that we need to look
at not through -- we need
to look at it objectively
in terms of what it's doing.
Which raises the question;
how did all this early
development actually happen prior
to a formal GIScience that now
aspires to guide it and further?
What drives what, the GI
Science or the GI Technology
and how do they interact?
And how did the unnamed
science work then?
How does GIScience work now?
What is its dynamic?
What flows from what?
How much of the innovation
flows from science?
How much it flows from the
technology that's been generated
really nearly in so many
different corners of the world
or is the reciprocal
interaction between the technology
and the science actually the real
dynamic which I think is the case?
It's perhaps also time to examine
current interactions and boundaries
between GIScience and
geography themselves as sort
of separate disciplines
or integrated disciplines.
In what ways are these boundaries
fixed or rigid, permeable or mobile?
What is the intellectual trade
across these boundaries look like?
Is the flow balanced or
dependent in which direction?
Are the two fields
diverging or converging?
These are all interesting questions
that I think we need to think about.
There's a conceptual thinking.
I prefer instead of
theory I prefer conceptual
as a term but nonetheless.
But how does conceptual thinking
as a part of the iterative process
of overcoming research obstacles
generate innovation in GIScience?
In which social structures
of GIS will best enable the
next conceptual breakthroughs
and new methods and what social
values will shape the future
of our coupled scientific and
technological research program?
This is incredibly
important to think about.
And I think that in
terms of the practice
of science the social structures are
certainly very influential and need
to be taken into account
and need to be thought
about as we plan our
structures in our organizations,
in our research programs.
How does -- how do curiosity
driven research and applications
or where research actually
differ in the practice
and intersect those
drivers of GIScience
and how mutually exclusive are
these categories in any case?
I don't think they are
that mutually exclusive.
So just to leave that on this note,
a lot of work that I think addresses
these issues from other disciplines
and from the general fields of
philosophy and history of science,
there's a lot to offer and I think
we should be focusing in some
of these works that Andrew Pickering
who in fact developed
the term mangled in terms
of his book The Mangle
of Practice .
Helen Longino a feminist
science philosopher The Fate
of Knowledge , and
several other books,
Peter Galison who wrote just wrote
a really great book a few years ago
called Einstein's
Clocks, Poincare's Maps .
All of these look at
the social context
within which science is conducted
and I think they offer illuminating
and very productive ways that we
might begin to think about this
if we really want to move
this whole thing forward.
Okay, and I'll just confuse
you with this last quote
because we've been talking about
time so much and we understand
that time can be subjective as
well as structured arbitrarily
so I'll close with this quote.
Thank you very much.
Okay, let me just add at the end.
There are archival
documentation available.
It's online for our RTIG PSI GS
and also hopefully we might have
a much more extensive archives
on this topic available at the
Library of Congress Geography
and Map Division sometime soon.
That's my hope.
Questions please.
>> This is Marlon.
Doug I want to thank you for
a really enlightening talk.
You've given us enough
ideas probably
for about a two week long symposium.
I was fascinated by -- well I
guess one sort of minor thought,
current developments in
unmanned aerial vehicles.
I mean I think that that's
going to be another core arena
in which real-time integration is
going to be very heavily involved.
But it was interesting
that you showed that slide
with patent number
5214757, and -- is this on?
>> Yeah, it is on.
>> It is on, okay.
I appreciated how the fact
that you had the slide
for patent number 5214757, and they
took up the advantage of wireless
to plug that into Google Scholar.
And curious as to whether or not
there would be any academic articles
that would be siding it, well
I didn't have time to scroll
through the -- I [inaudible] 381
citations that are listed but most
of them seemed to be
patents -- other patents.
And it strikes me in a
sense that there are sort
of two parallel literatures,
there is the parallel literature
of the inventor who uses
the patenting process
and other inventors obviously are
very familiar with that literature.
Generally speaking
people in academia aren't,
and but the same token -- when they
say parallel literature sometimes
inventors, and in fact patent
examiners are not terribly --
are familiar with the
general technical,
academic literature
including the examiners
who apparently approved a
patent, I think around 1990
for what basically works out to
be a [inaudible] map which proved
to be somewhat useful at
least for the time being
for some patent trolls who then
undertook to initiate law suits
against 20 geospatial firms.
There seems to be though, I
mean this kind of disconnect
between the inventor community
that you refer to let's say
as the private sector and academia.
Do you sees much of a
convergence happening?
>> Douglas Richardson: Well they are
two different types of literature
and writing a patent is a very
structured process and it's informed
by -- has to be informed by all
research because if it's shown
that somebody else is
doing that already,
even if the patent is then issued,
the patent can be challenged
that there's other prior
art that involves this
and sometimes patents slip through,
that are trivial or
not very meaningful.
And I can tell you in
the case of our patent,
it was thoroughly researched
by both --
both by us and by the whole
sequence of patent examiners.
And they raised the all -- they
asked all kinds of questions
where we'll provide answers and
they did all kinds of research.
It was a very comprehensive process
because it is such a broad and core
and important new are of technology.
So it got reviewed very well, but
I don't think -- I would urge --
in think that people -- I know
a lot of people in academic area
in private companies do read
academic articles if they bare --
tend to bare on their
particular area of interest.
They don't tend to write very
many however, and that --
probably all the academic
articles are written have been
in the last 12 years or so.
But we should be mounting up now,
if I had started earlier I would
have had quite a publication record.
But none the less it's - I
think that there's great value
and I would urge anyone
interested in any topic
to consider doing research,
in fact in always,
not only in the patent literature.
In fact one frustration I
have is somebody once said to,
I was talking about something
we had done and he said,
"Well where is it in
the literature?"
And I said, "Well here
are 30 different accounts
of people doing and using it."
But because it hadn't been
published in an academic journal,
they didn't know about it
or we're not able to kind
of grasp what was going on.
So I think that -- I would urge
researchers in any of these topics
to both do look at
patents because it's a --
it's sort of a cutting edge
of innovation and creation
and new ideas and so forth.
Now not all of them are
going to be in the topics
that someone is interested in.
But I also would urge people to --
particular people who want to
write histories of geography
to not ignore all the other
literature that's out there
that may not -- that are relevant
to this particular history
if the y want to come up with an
accurate history of what occurred
because it'd be kind of like a
historian not going out and looking
at -- in other fields not going
out and looking at actual artifacts
and letters that had been written
and basic archival material
when they do a history.
And I think that most
people doing histories
and geography, they don't do that.
They are missing -- there's
a blind spot in everything
that occurred that's not
published in an academic article.
And that seems to me to be not a
very thorough way of doing a history
and I think I would urge those
doing histories of the discipline
to look both at the much broader
literature, now that it can be done
with Google it's easy to
get that sort of thing.
Of course the articles written
within the discipline are important
but if you want to understand
these kinds of things you do have
to go beyond what's
already been written.
It's -- yeah?
[ Inaudible Speaker ]
You bet. Thank you all.
I really appreciate it.
[ Applause ]
>>This has been a presentation
of the Library of Congress.
Visit us at loc.gov.