Design of Spatial Applications

Design of
Spatial Applications
Matthew Hockenberry
The Media Laboratory, MIT
hock@media.mit.edu

CHI 2007 Course Notes

Copyright is held by the author/owner(s).
CHI 2007, April 28-May 3, 2007, San Jose, California, USA.
ACM 07/0004.

Instructor Bio:

Matthew Hockenberry: The Media Laboratory, MIT

Matthew is a graduate of MIT’s Media Lab with an M.S. in Media Arts and
Science. He was previously at Carnegie Mellon where he studied Logic &
Computation and Human-Computer Interaction. He has been involved in
Academic Research for the past six years, and supervised projects for the
past four. He has experience in Computer Science, Design, Psychology, and
Experimental Design.

As part of his work at MIT he has developed location-based mapping
applications that focus on adding community and user-centricity to web-
based maps by utilizing artificial intelligence and data mining techniques. He
directs the PlaceMap Project, which is building place-based applications for
the MIT community. Matthew has published a variety of papers on spatial
applications and led seminars at MIT on this topic.

Introduction Design of Spatial Applications

Agenda:

Introduction 9:00am - 9:05am
Framing Lecture 1 9:05am - 9:50am
Case Study 9:50am - 10:00am
Design Exercise 10:00am - 10:30am

11:30am - 12:00pm
Design Exercise (cont.)
12:00pm - 12:15pm
Case Study
12:15pm - 12:55pm
Framing Lecture 2
12:55pm - 1:00pm
Concluding Remarks


Objectives of the Course:
• Introduce the idea of a "spatial application" an application that makes use of
spatial knowledge, awareness, or presentation in order to achieve its goals.

• Present the tradition of spatial representations from cartography, to geographic
information systems, to urban planning and art.

• Understand the psychology of spatial decision-making, and how our cognitive
maps and geographic common sense are different individually.

• Consider the social necessity of sharing spatial information and the impact this
has on design.

• Formulate design goals and approaches that can be employed successfully to
further tasks that rely on spatial knowledge, and demonstrate these in a group
design project.

• Review the state of the art for spatial technology.
• Approach new representations and uses of spatial knowledge and the impact of
these approaches and representations.


Abstract:
The design of spatial applications is intended to provide perspective on design issues in the
development of applications that incorporate spatial knowledge, representation, and
purpose. To this end, the course focuses on both traditional answers to these issues, as well
as exploring the background that lead to these answers.

This takes the form of understanding traditional implementations such as geographic
information systems, their purpose, and the role they have served in representing spatial
knowledge. Critical examinations reveal difficulties with these solutions, and the current
direction of web maps also provides new insight into the role of GIS in shaping spatial
application development. Related traditions, such as architecture and urban planning are
mined for their perspectives, often complimentary but with different focuses.

Background into spatial navigation, decision-making, action, interpretation, and
representation arise from cognitive psychology and neurophysiology. This background
produces understanding for certain directions in spatial application development, but also
reveals troubling contradiction between human cognitive representation and traditional
constructions.

This background serves to inspire new design perspectives that appreciate traditional
approaches but attempt to address human understanding and user behavior. Untraditional
approaches are also considered, including recent explorations into more human
commonsensical understanding of space, the relationship between spaces and places, and
the tension between the need for social sharing of spatial information and our own internal
representations.


Introductory Notes:

Questions: Ten minutes built in for questions each half, so ask them -
but I may defer until later in the lecture if appropriate.

Pace: Using those ten free minutes above I’m happy to focus on
important points for longer periods of time.

Interactivity: The main focus for interaction is during the case studies,
the design sessions, and the break (if you want).

Style: Generally speaking, the course focuses on ‘high level’ concepts
punctuated by examples and case studies, but questions are welcome
on more precise details (although generally, that’s what the references
are for).


Framing Lecture One

What is a Spatial Application?
Tradition of Spatial Representation
The State of the Art
Introducing Place
Psychology of Spatiality


Definition: An application that makes use of spatial knowledge,
awareness, or presentation in order to achieve its goals.

Design Point:
Don’t you mean mashup?
One might think that, considering that web map applications make up more
than 80% of mashups. Many mashups lack clear design goals and motivation,
something we should think about changing.

Design Design of Spatial Applications

Design Considerations:

Craigslist + Google Maps Craigslist


User Centered Design

Given a location for a user in a system, what can the system do?
If a user knows his or her location, what things do they want to do?

Location and space are strong limiting factors.
(yes, even today and in some respects especially today)

Until we invent the teleporter this is a fact of life.


Design Model

Focus on user experience and user centered design.

Strong emphasis on task(s) grounded in spatial reality.

Where you are matters - any application that cares where you are
should understand why that matters.

Design Point:
The question of information visualization will come up a lot -
it should, it’s important.

But information visualization removed from clear user
understanding and task goals is just information, not application


Tradition of Spatial Representation:

Spatial application arise from a strong tradition of the importance of
spatial information and techniques to represent our place in the world.

Tradition Design of Spatial Applications

The Tradition:
A Brief History of Maps:


Maps are:
Informative

Maps provide us with information about what is where. This is about
describing places, putting information in a spatial context, and
providing us with a rich world view of geo-information.


Maps are:
Directive

Maps tell us not only what is where, but how to get there. Anyone who
has driven a car to an unfamiliar place knows how important having a
good map can be. Getting from point A to point B has been one of the
fundamental purposes of maps.

Design Point:
In some regards we could think of direction as simply another
piece of information, but it has a special role in terms of task.
After ﬁnding out ‘what’ is there, the next question is almost
always, how do I get there?


Maps are:
Aesthetic

Maps have been works of art as much as they have been works of
information. Artists have explored issues of perspective, presentation
and information visualization within the borders of countries.

Design Point:
Maps have a rich history of artistic exploration. Just because spatial
applications are ‘serious’ doesn’t mean that exploration has to stop.


Maps are:
Lots of things...

...shared social perspectives, tools of political
power and expression, philosophical treaties about the world...


Maps are:
Changing?
What’s new in mapping?

Is it a little or a lot? A lot and a little?


Other Paradigms:
For Spatial Representation

Who’s out there: Things we say:
Geographic Information Systems Map, Cartogram, Geographic
Psychology of Space Information System, Mashup,
Urban Planning Painting, Directions...
Computer Science
Visual Design
Philosophy
Everyone...


Other Paradigms:
For Spatial Representation


Changing Paradigms:
Forget Maps:

Hey! We just spent a lot of time thinking about maps!

That’s true but let’s think about maps in a different way:

Assume: Maps were really useful when we didn’t have a lot of ways of
using spatial knowledge and data and in certain circumstances.

Conclude: We should look at the justifications and goals of a map and
transport those to alternative domains and practices.

Sometimes we’ll need to bring a map along, sometimes not.


Changing Paradigms:
Forget Maps:

Old Idea: Map that shows you things you are interested in.

Goal: Get people to go to things that are interesting to them.
Limitation: Useful for planning but not everyone plans.
How can we address this goal in practical life.

New Idea: Phone that tells you when you are around interesting things.
Mechanism: Phone (with GPS) vibrates when you are near something
‘interesting.’ If you press any key it shows you more details.

Your preferences for ‘interesting’ may still be set on a web map, or your
phone could learn over time.


Maps & Spatial Applications:
Don’t start with this:

If you think your application needs a map, chances are good
that it will have one.

Not all spatial applications need a map representation to use
spatial information in effective ways. Some will - some won’t.

Start with this:

And design an application the way you normally would,
based on questions of user experience and task. If you need
a map, it will be obvious.

Design Point:
Actually in the current climate, almost every spatial application has a
map. It may be better to look at every other solution before deciding
to add a map. Of course, if a map makes sense - it makes sense.


Traditions of Spatial Representation
Geographic Information Systems

The Geographic Information Systems Approach Geographic
Information Systems (GIS) are tools and technologies used to view
and analyze information from within a geographic perspective.



The primary focus of these applications is to link information to location
and enable the visualization of large sets of spatial data.

Typically, a GIS application presents images that have been captured
by sensors, terrestrial cameras, and so on. It then supports the
manipulation of these images by zooming, panning and layering
additional sources of information.

More sophisticated applications represent this as vector information to
be rendered at run time. This allows the addition or removal of certain
parts of the geographic content independently (showing and hiding
roads, buildings, parks and so on).



The typical interaction in GIS applications is the query.

A user specifies a set of geographic information to serve as a base
structure and then layers supplemental geographic information on top.

For example, we might look at only the rivers in a geographic region
and then layer information such as presence and type of trees and soil
structure in order to predict riverbank erosion.

This kind of approach is very powerful especially when constructed
with modern design techniques.


Benefits of GIS:

There are a large number of benefits to the GIS approach.

It focuses on displaying accurate information, which is of absolute
necessity in certain kinds of applications.

The layer metaphor scales well and supports the view and
manipulation of large amounts of information that may or may not be
obviously related. In this respect, the GIS approach is very flexible.

Many aspects of the world can be captured in GIS; Spaces full of
discrete spatial objects, measures of the attributes and relations
between these objects, or even continuous measurement of several
different properties or themes within a concrete spatial region.


Limitations of GIS:

There are fundamental limitations to the GIS approach, and many
difficulties in implementing it successfully.

The most serious of these still remains the serious distance between
the system preconceptions, and the user’s understanding, of the goals
in interacting with geographic information.

This can often result in usability problems that are tied to failures in
interpretation and gaps between user task conception and GIS query
implementations.


Limitations of GIS:

These problems are well addressed by Traynor and Williams in their
survey of several GIS systems while attempting to understand how the
usability of these systems affected users.

They chose a selection of common tasks, such as opening a map and
analyzing multiple layers of spatial information.

They concluded that the GIS applications had three distinct problems
when used by non-specialists:

They often rely on technical terminology, they require a strong mental
model of the software architecture to be effective, and there is no
strong attachment between the final compound representations of
spatial information and how that information was generated.


Beyond GIS:
GIS applications are designed for individuals who possess expert skill
at dealing with the manipulation and organizing such data.

Beyond this is the idea of ‘thin client’ applications that need to be
constructed to present this information to casual end users.

This speaks to the fundamental limitation of GIS applications: They are
concerned about the precise output of large sets of data.

While this makes them well suited to data professionals, they limit the
end user population to only these professionals.

The use of layers to categorize disparate sets of information speaks to
the inability to establish deep meaningful relationships between this
information and an inability to tie it to the geographic display in more
than a very limited fashion.


Technology for Space and Place:
Rise of the Web Map

The base for new web-based mapping applications;
Offer increasingly powerful APIs (Application Programming Interfaces);
Enable outside developers to build their own maps (mashup).

Technology Design of Spatial Applications

The Nature of Web Maps

These maps often showcase widely disparate displays of spatial
information in a powerful web-based geographic display.

Web maps are very similar to traditional GIS applications with a few
key differences.

Google Maps is probably the best known example of these mapping
engines.


Web Maps: Differences from GIS

These applications dismiss the need for sorting through widely
disparate information within a single application and instead offer a
map based on the particular spatial information needs of the user.

If you need to see a map with all the cabs in New York City, go to this
address; if you are interested in a map with apartment listings from
Craig’s list, go to this address.

In a sense, each layer in a GIS application becomes a new instance of
a web map.

Web maps are in many respects the ‘thin user client’ GIS views that
we have been waiting for.


State of the Art in Web Mapping

Web maps can scour new sources of information from the web at run
time.

While earlier web-based maps were more clearly directive these maps
embrace the idea of a map based information display.

Anyone can display any kind of spatial information they would like.

This is a powerful approach and within months after the first next
generation web mapping engine launched hundreds of different maps
displaying all kinds of dynamic spatial information have become
available.


Criticisms of Web Mapping

In some respects, however, these maps are a step back.

They forego the complex layer based approach of GIS applications in favor of
tailored unique displays: This limits their scalability.

Programmers using these technologies must incorporate disparate spatial
information on their own, with only the capability of displaying that information on
these applications.

In short, these maps offer a powerful front end for the display of spatial information,
but not a mechanism for building relationships between that spatial information.
They fail to support the kind of complex relationship between geographic
information and supplemental spatial information that a developer might desire.

One can add “spatial information pins” to a map, but cannot change how the
underlying image is displayed based on differing spatial information.


The State of the Art:
A Sim City World?


Technology Targets: Frontend

Firefox plugin, RIA: AJAX application or Flash/Flex application,
Mashup (flash or js),Widget (javascript),Google Earth Integration


Technology Targets

Firefox plugin
(http://www.vinq.com/technology/greasemap/)

AJAX application
(http://api.local.yahoo.com/eb/)

Flash application
(http://www.neave.com/lab/flash_earth/)

Mashup (flash or js)
(http://www.housingmaps.com/)

Widget (javascript)
(http://widgets.yahoo.com/)

Google Earth Integration
(http://earth.google.com/)


Technology Targets: Backend
Web services, Spatial tagging,
Collaboration & sharing, Mediating data by space


Technology Targets

Web services
(http://research.yahoo.com/zonetag/)

Spatial tagging
(http://www.semapedia.org/)

Collaboration & sharing
(http://info.placesite.com/)

Mediating data by space
(http://dencity.konzeptrezept.de/)


Technology Targets: Hardware & More...

Mobile phone applications
(http://www.macromedia.com/mobile/gallery/)
Gps integration
Spatial videos
(http://theunseenvideo.com/)
Spatial web pages
(http://micro-info.blogspot.com/2005/03/
autodiscovery-and-location-aware-web.html)
Spatial art / installations


Quick Start Guide

Flash Python
(http://www.kirupa.com) (http://www.byteofpython.info/, also: http://web.media.mit.edu/~hugo/
conceptnet/)

Flash YMaps
Firefox plugin
(http://developer.yahoo.net/maps/flash/asGettingStarted.html)
(xul javascript) (http://roachfiend.com/archives/2004/12/08/
how-to-create-firefox-extensions/ , http://www.gmacker.com/web/
Javascript content/tutorial/firefox/
(http://www.w3schools.com/js/js_intro.asp) firefoxtutorial.htm)

AJAX Widgets
(http://dhtmlnirvana.com/ajax/ajax_tutorial/# , http://24ways.org/advent/ (http://widgets.yahoo.com/workshop/)
easy-ajax-with-prototype , http://www.yourhtmlsource.com/javascript/
ajax.html)
Google Earth
(http://www.keyhole.com/kml/kml_tut.html)

Google Maps
(http://www.econym.demon.co.uk/googlemaps/ ,
http://ruk.ca/wiki/Making_of_the_Charlottetown_Transit_Map)


New Assumptions

Things to assume: All of the data you ever want will be there.
Location information isolates people by distance.
Invasiveness is directly related to the usefulness of the invasion (with
caveats)

Things to not assume: All of that data will be easy to get, complete,
or nicely formatted.

If you build it, they will come.

After you stick data on a map your job is done.


Revisiting:
What is a Spatial Application

Anything that can instill a sense of place or making use
of where you and what that means.


Introducing Place:
Place vs. Space

Places are spatial locations given meaning by human experiences in them.

Place is distinguished from space by being socially constructed and local, rather
than quantitatively described and universal.

In other words, people make places out of space.


Introducing Place:
What is place?

In the physical world, a place is simply a space that is invested with
understandings of behavioral appropriateness, cultural expectations, and so
forth.

We are located in “space”, but we act in “place”. Furthermore, “places” are
spaces that are valued.

The distinction is rather like that between a ‘house’ and a ‘home’; a house might
keep out the wind and the rain, but a home is where we live.


Introducing Place:
Places are active.

Places provide a context for everyday action and a means for identification with
the surrounding environment.

They help inform our own sense of personal identity they make use identifiable
to others.

Behavior is linked to place.

Judgments of what is appropriate are based on the place of an act.
Meanings given to places are a fundamental component of social interaction.


Introducing Place:
Place as Social Construction

Place is both broader and more specific than space.

The same location— with few changes in its spatial organization or layout—may
function as a different place at a different time.

“An office might act, at different times, as a place for contemplation, meetings,
intimate conversation and sleep.” This suggests that a place may be more
specific than a space. “A space is always what it is, but a place is how it’s
used” (Harrison, 1996).

This meaning can change based on our social or cultural role.


Introducing Place:
Place as Social Construction

Humans rarely share spatial coordinates.


Introducing Place:
How can we understand place?

Who does the work?
Machines
Teach computers to do it (the ai approach)
Number of humans needed - little
Amount of software required - lots
Humans
Let humans do it (the wiki approach)
Number of humans needed - lots
Amount of software required - little

Both
Combination (human augmented ai)
Number of humans needed - some
Amount of software required - some


Design
Spatial Social Sharing:
Question: How hard is it to draw the country you live in?

Answer: Surprisingly hard.

Psychology of Spatiality:
How do we see space?


How do we see space: An anecdote

It’s complicated, but start with the fact that we don’t represent what we
see in three dimensions.

It’s more like 2.5D or 2D with elevation as an additional (and vague)
property.

Evidence: Ask people to estimate 2D distance, and then ask for the
same estimation over slope.
Result: People are surprisingly good at the first estimation, but
surprisingly bad at the second one.

Anecdote: The shortest distance between two points is a straight line,
and a flat one at that.


How do we see space place

Initial Experience

Place Construction

Communication & Translation

Interpretation & Assimilation


Platial Representations

?
Are maps the best representation?

Maps are useful because they are so global.

How do we really see space and represent place?

Effective application take advantage of this.
Ineffective applications will over-generalize.


Case Study One:
A Lesson in Reduction.

Case Study:
A Lesson in Reduction

The simplest way to achieve
simplicity is through thoughtful
reduction. (The Laws of Simplicity, John Maeda)

Consider two examples: Linedrive & Metrobot

Linedrive reduces spatial information to communicate driving
directions.

Metrobot reduces spatial information to communicate street
information.

Case Study Design of Spatial Applications

Case Study:
Linedrive (msn maps and directions)

What is linedrive?
Map visualization application focused on driving directions.


Case Study:
Linedrive

Why is linedrive compelling?

Design methodology

While it is an interesting exercise in some cool
algorithms, it also addresses that existing
representations and techniques don’t meet mental
expectations.

Exercise in simplicity: What is the most and least
amount of information to get from one place to
another effectively?


Case Study:
Metrobot

What is metrobot?

Business listing directory with a
unique spatial view.

Shows you the street, with
business listings with linking
information.


Case Study:
Metrobot

Observation: It translates really
well to a mobile platform.

How does it do this?

Over the web.

No special application, just scales
well and works nicely on web
enabled mobile devices.


Case Study:
Metrobot
(original)

Metrobot introduces a very spatial,
but very simplified view of
information:

Goal: Show only the necessary
information effectively by giving a
strong task centered view that is
abstract but strongly orienting.

Comments: Metrobot seems very
‘zoomed in.’ It is somewhat difficult
to get a sense of context or navigate
far beyond the current location.

Good for information, difficult for
browsing and large scale search.


Case Study:
Metrobot New York, NY: Columbus Ave.

(redesign)
Metrobot does have a google map.
Unfortunately it is at the periphery of the
interface.

Here the google map becomes a strong
source of context in a new representation.

The real map is small, slightly skewed, and
contains overlays indicating the current
position and links.

It can be tied together with a little ajax magic
to the main representation.

Design Point:
Some general design decisions: the title has been
Metrobot
strengthened focus the current location and
branding emphasis has decreased. These aren’t
really ‘spatial’ design changes.


Design Exercise:
Objective: Develop the concept for an effective, interesting, and novel
spatial application

Requirements: Short summary of the application (abstract)
Sketches showing application summary + any additional features

User experience walkthrough (use caseish in nature, with any
appropriate sketches)

Answers to the following questions:
Who is the user?
What is the task?

What technologies are appropriate and why?
What is the role of spatial information and location?
How do we represent that information and why is that representation
effective?

Exercise Design of Spatial Applications

Design Exercise: Free Ideas
A representation that is focused on a particular spatial task, and is
unique for that.

A representation that incorporates time with place and space.

A representation that makes intelligent use of scale.

A representation that incorporates user interest goals.

A representation for planning, a representation for acting.

Absurdist ideas done well are ok too.

Domains:
Crime, coffee, health, meeting friends, making business connections,
finding pickup softball games (but only if you have a bat with you -
domain constraint).

Design Exercise:
When presenting:

What’s this idea.
What ideas led to this one.
What (if anything) did you learn.


Design Exercise:

Secret Objective: Hope you messed up.

We can learn a lot from that.


Case Study Two:
Relationship between representation and reality.

Case Study:
Relationship between representation and reality.

Consider two examples: Shoutwire & Housingmaps

Shoutwire is a social news site with an awareness of the locative
background of interaction.

Housingmaps is a mashup that helps search for real-estate listings.


Case Study:
Shoutwire
Shoutwire is a social news
site.

The comment page shows a
large map to indicate where
‘shouts’ are. Shouts serve as
an indication of approval or
interest.

Why is there a big map
here?

Design Point:
Shoutwire is an interesting site that is doing
something unique, but unique isn’t always good.


Case Study:
Shoutwire
(original)

What is this map?

Goal: Show national /
social background of
shouters.

Comments: Interesting
design decision,
encourages more global
community
acknowledgment and
perspectives.

This emphasis helps set
shoutwire apart from
other social news sites.


Case Study:
Shoutwire
(redesign)
Get rid of this map!

Goal: Show national /
social background of
shouters.

Comments: We can
show more information
with a tag cloud instead
of the map.

This doesn’t present as
striking an initial
impression, but takes up
less space and
Design Point:
communicates the same
There are lots of ways to redesign this. We could show
(or more) information in
numbers beside countries, add additional grouping (continent,
less space.
state). The names of individual shouters can be displayed by
ajax links, or more detailed map popups could be used.


Case Study:
Shoutwire

Of course, this is not the
only solution.

Point: There are lots of ways
to communicate spatial
information that is
meaningful and relevant.

Sticking things on google
maps is only one option.


Case Study:
Housingmaps
Housing maps is mashup that helps search for real-estate listings. It
combines Craigslist and Google Maps. All design, not much code!

Housing maps is a classic mashup, Google Maps + Craiglist, and it is
an effective one.


Case Study:
Housingmaps
Why is Housingmaps so compelling? It offers a great experience.

Simple controls that relate to the task, show the necessary amount of
information the user wants.

Multiple views of the information, traditional listings with simple clear
information and a spatial map view that is well connected.
Both serve as navigation tools depending on user need.

Everything works toward the task, nothing here seems like an afterthought.
One can actually imagine using this application.


Framing Lecture Two

Models for Spatial Representation
A Naive Geography
A Sophisticated Cartography
Building Blocks for Spatial Applications
Design Principles

Models for Spatial Representation:
Using and designing the world

We’ve seen some interesting alternative representations in the case
studies.

If people don’t see the world as satellite photos, and alternative
representations can be more useful that traditional ones in certain
circumstances, how do we design?

If we look at how people interpret the world, does that help? or are
there reasons behind traditional representations. In particular, what
kind of models make sense for our applications?


Natural Human Representation

This varies quite a bit.

Review: GIS Representation

GIS representation is very simple.

Data organization
Data is organized by sets of homogenous
information. This allows disparate
information to coexist.

Data visualization
Generally data is visualized as layers that
can be manipulated by the user. Often
times there are tools such as zooming and
magnification to help users.


User Centered Representation

Although not a
realistic
representation,

Saul Steinberg's
"View of the World
from 9th Avenue." is
very compelling.

Real examples are
limited (personalworldmap.org)


Task Centered Representation
Subway maps are strong examples:
Simple goal, get from point a to point b.

Limits user options, limited encoding in terms of direction and distance.
This can lead to some confusion in other contexts.


Playing with Representation

Distortion techniques can vary.


Playing with Representation

There is a balance between perspective and constraint.

How we manage this balance requires understanding which features are
important when - and why.

This comes from understanding user perspective, and consequently the
distinction between general human perspectives and personal ones.


Psychology of Space:
A Naïve Geography

Naive Geography (or common sense geography) is the body of
knowledge that people have about the surrounding geographic world.

Naive Geography captures and reflects the way humans think and
reason about geographic space and time.


A Naïve Geography

Tobler's "First Law of Geography":

Everything is related to everything else, but near things are more related than
distant things.

Design Point:
This statement speaks very clearly to why space
is so important in decision making, and why
spatial applications can be so important.


A Naïve Geography
Some anecdotal (though supported) elements of Naive Geography

-Naive Geographic Space is Two- -Geographic Space has Multiple Levels
Dimensional of Detail
-The Earth is Flat -Topology Matters, Metric Refines
-Maps are More Real Than Experience -People have Biases Toward North-
-Geographic Features are Ontologically South and East-West Directions
-Distances are Asymmetric
Different from Enlarged
-Table-Top Objects -Distance Inferences are Local, Not
-Geographic Space and Time are Global
-Distances Don't Add Up Easily
Tightly Coupled
-Geographic Information is Frequently
Incomplete
-People use Multiple
Conceptualizations of Geographic
Space


A Naïve Geography

Perhaps: 'Naive Geography' "may be a search for the principles,
schemata, and heuristics that allow people to find things in novel
environments."


A Naïve Geography: A Story

Finding things in first world economic systems


Naïve Geography in Practice?


From early man to the renaissance man:
A Sophisticated Cartography

The maps and other representations we are familiar with don’t seem to
be directly based on the common sense understanding of geography.

Cartography, from data collection to presentation, comes from a very
different background with its own traditions and techniques.


Elements of Cartography:
Cartographic Definition of a Map

What is a map?
“A graphic depiction of all or part of a geographic realm in which the
real-world features have been replaced by symbols in their correct
spatial location at a reduced scale.”


Cartographic Definition of a Map

Map Functions
Information Storage To be effective,
Communication must be correctly designed
Tool for Analysis and constructed
Final Presentation


Parts of a Map

Legend, Scale, Credits, North Arrow, Place, Inset, Ground,
Figure, Neat line, Border, Title


Elements of Cartography

Elements of Cartography

Medium, Figure, Ground, Grid, North arrow/Compass,
References / Sources / Credits, Point/Line/Area symbols,
Border, Neatline, Insets, Text and Labels, Title, Scale,
Metadata, Coordinates, Projection, Legend...
Graticule/

There is a lot that (could be) going on here.


Lessons from Tradition

Traditional Cartography offers guidelines for successful design.

Difficulties arise when the design focuses on novel elements.

Even within this tradition, however, there is a lot of flexibility (and room
for error).


Example: Map Title

Consider one element: the title
Varying our language here can significantly alter meaning.

Distribution of Employment by State 1996
USA: Employment Distribution 1996
U.S. Employment: 1996 Distribution
America at Work
Where the Jobs are Today

Design Point:
The point here is that detail can have a large impact. What
traditional cartography doesn’t have to deal with (but we do)
is the difﬁculty of constraining dynamic information.


Kinds of Maps
Map Types
There are many different established map types and guidelines
for their construction.

Point Data: Area Data: Volume Data:
Reference, Choropleth, [Isopleth,
Topographic, Area qualitative, Stepped
Dot, Picture Stepped Surface,
Symbol, surface, Hypsometric],
Graduated Hypsometric, Gridded fishnet,
Symbol Dasymetric, Realistic
Line Data: Reference perspective, Hill-
Network, Flow, shaded, Image
Isopleth, map
Reference

How do we choose?
Look at the data, look at the dimensions, look at scale...
This becomes harder for more complex maps (dynamic
data, elements of time, nonstandard distortion)


Cartographic Design

Why we need design:
A map has a visual grammar or structure that must be understood and used
to get the best map.
We should reflect cartographic knowledge and convention when it makes
sense (e.g. forests should be green) but it won’t tell us everything.

Focus.
Good design will draw focus to the elements that are important... and away
from the elements that are not as important.
It’s all about focusing attention.

General Design Tools
We can use traditional design tools like like visual balance, color, contrast,
text and patterns.


Cartographic Design Elements

We can rely on traditional design elements:

Visual Balance Balance & Alignment Elements of Contrast
A more holistic measure. Visual Create Visual Levels More contrast = stronger figure
balance is affected by: the Not just the darker element
"weight" of the symbols the visual Color & Contrast
hierarchy of the symbols and Color can be useful in Contour
elements the location of the emphasizing and focusing Sharper contour (edge) = stronger
elements with respect to each information. Humans, however, figure
other the visual center of the map are bad at coding complex color
associations. Color can also be a Closure
Visual Center simple method of adding contrast Closed element = stronger figure
A little off of the true center. to a visual image, but at the same
People will start looking around time it can decrease contrast. For Enclosure
this point. This should be the example, saturation and Intensity More enclosed = stronger figure
perceptual center of design. 5% of map better onto values than hue. Without it, can’t distinguish
height elements.
Dimensions of Color
Hue Saturation Intensity

But we need to acknowledge how they may change spatial representations.


Example: Cartographic Contrast

All Income Levels Highlighted Level

We need to acknowledge how they may change spatial representations.


Epistemology of the Spatial World:
Spatial Data


Spatial Data

x,y { }
Getting Spatial Data
Keeping Spatial Data
Using Spatial Data


How we get it (an example):
How do we collect spatial data?
What kind of spatial data do we need to collect?

IKE:
A New Zealand company called Surveylab, has licensed technology developed
by the U.S. Army to produce an all-in-one mapping tool.

The device, originally called HAMMER and rebranded IKE, for Hand-held
Apparatus for Mobile Mapping and Expedited Reporting combines a Global
Positioning System (GPS) receiver with a hand-held iPaq computer, a digital
camera, compass, laser distance meter, inclinometer and Geographic
Information System (GIS) software in one portable device.

Simpler Methods: Compass, Pencil, Stars...

Spatial Data: Who gets it?

Professional Surveyors
You’ve probably seen them around.
Goal is to gather, update, and maintain the data.

Anyone
Open submission and access to data.

Is government control a concern?
If it is then we (the people) should do that job.

And keep it open for all of us.
Also we might want data outside the norm.


The Degree Confluence Project

The goal of the project is to visit each of the latitude and longitude
integer degree intersections in the world, and to take pictures at each
location. The pictures and stories will then be posted.

OpenStreetMap: Wiki style world

OpenStreetMap is a free editable map of the whole world. It is made by
people like you.

OpenStreetMap allows you to view, edit and use geographical data in a
collaborative way from anywhere on Earth.

Spatial Data: Methodologies

Deconstruction
Accounts of place are reduced to spatial data.

Advantage: No bias, pure data, lots of uses.
Disadvantage: Lacks understanding, requires a lot of work to get back
to the initial place sense.

Translation
Accounts of place are transmitted directly to us, with encoded spatial
data.

Design Point:
This isn’t to say that deconstructed data is bad, or
even harder to use. Sometimes it can be signiﬁcantly
easier to work with than translated data.


Spatial Data: Methodologies

Example spatial data - elevation.

Deconstruction: Elevation map and table of elevation measurement
(FASL, AMSL, HAAT)

Translation: Here translation is very context dependent. In many
translation this information wouldn’t come up at all. An example where
is would come up might be in a discussion between two mountain
bikers:

“The trail is pretty tame, except a quarter mile after the bridge where it
drops to a sharp incline for about an eight of a mile.”


Spatial Data: Storage

Database Semi-structured Other...
The data is stored in a XML, other loose Entry
large database. (often) hierarchical Organized
data structure. Automated
These databases Manual
tend to be difficult to Concerns of Emergent
organize, but expand scalability. Wiki-style
well as a method of ...
data storing.


Spatial Data: Location Awareness

The user’s location is often the principle spatial data item of interest.

If we know this, get this, and choose to use this it can radically alter
the design and structure of our application.

How do we get it?

Either the user tells us. Or we guess.


From my world to ours:
Minding the Gap

This tradition descends from the need for a social spatial view in a
fixed form.

With our new dynamics, we can push beyond this, but we need to
push in the right way.

There is no formula but there is a language.


Building Blocks of Spatial Applications


Building Blocks of Spatial Applications:
Location Awareness

Location awareness usually refers to approaches that understand where a user is,
either through network monitoring, special hardware such as GPS, or combinations
of these approaches with user input.

The precision of these techniques is rapidly increasing.

Technologies such as wireless triangulation and wireless positioning are rapidly
becoming able to approach these levels of precision without the need for external
sensors

Exemplar: Skyhook Wireless offers a service called Loki that exists as a plugin for
the Firefox web browser. This relies on access to wireless access information.
Comparing signal strengths and system conditions with observed database trends
of user behavior can be very precise.

Soon it will be able to precisely identify almost any location. However, the
necessary granularity for most tasks comes down to place – not to a number of
meters.


Web Maps
Web mapping APIs are the direct decedents of GIS style approaches to spatial
representation.

There are a number of key differences, however, which separate them from GPS to
some degree and make them attractive as possible building blocks for applications.

The main areas of interest are the lightness of the web maps when compared to
traditional GIS and the ease with which varied and diverse information sources can
be incorporated and realized, the result of which being the so-called mashup.

Exemplar: Google Maps are perhaps the best known of the web mapping APIs and
offers a very diverse set of features. Google Maps can be deployed on any web site
(given a Google approved API key) and can incorporate information from any
source. Additional functionality, such as seamless navigation, spatial interaction,
and drawing capabilities are also provided.

Web maps such as those offered by Google provide a rich foundation for the
display of spatial information but these web maps don't provide the capabilities for
aggregating outside data or interpreting it.


The Geo-semantic Web
The Geosemantic Web is an attempt to incorporate geographic and spatial
information in a semantically meaningful markup for the web. This is related to the
general conceptions of the semantic web. Specifically meaningful semantic geodata
and metadata are structured into web documents with the intent that they are
human readable, but also with direction for them to be machine-readable.

Exemplar: The Open Guide network is a geosemantically structured set of city
guides. these encode rich semantic markup in the form of RDF or XML. The Open
Guides represent a project within this approach that serves a practical purpose (city
information) is of a significant size (covering over ten major cities - mostly in the
United Kingdom - by contribution of altruistic individuals) and is well-structured
practical semantic markup with direct human representation and machine
instruction.

One could imagine a world where all of the information related to spaces and
places were carefully associated with correct geosemantic meaning.

It would, however, be a much more perfect world than today. Markup remains
limited by the insights and interests of the user base.


‘Smart’ GIS
GIS is focused on concrete data collection with an emphasis on objective spatial
data. This usually involves methods of data acquisition involving human agents with
specialized devices, but these are giving way to mobile data acquisition and
satellite photo analysis (remote sensing)

Exemplar: Environmental Systems Research Institute, Inc., commonly known as
ESRI, has emerged as the premier GIS solution in the commercial sector. Their
solutions, such as ArcGIS, offer support for numerous kinds of data sources,
manipulation capabilities, and advanced queries. This allows expert users to make
significant research efforts into geographical problems. Recently trends in web
mapping have resulted in sharing capabilities that offer interactions similar to those
found in Google Maps. This allows a full cycle of data collection, interpretation,
analysis, and sharing.

There have also been recent trends towards ‘smarter’ GIS systems that offer
models of behavior that have preserved some existing human interpretations of
geography. ArcGIS has begun to embrace these, but support remains very limited.

Good GIS systems such as ArcGIS are good for a particular kind of user, the expert
user. In general, the system does not attempt to understand the data itself.


Artificial Space
Ironically, perhaps some of the most interesting work in understand place comes
from research into artificial space. In the realm of computer supported cooperative
work and complex data visualizations, spatial metaphors have been useful for
communication and presentation of large amounts of data. To that end, significant
effort has gone towards understanding the role of place construction with an eye
towards practical investment of platial knowledge.

Exemplar: The work done by Dourish and Harrison is significant, as is the work in
the Data Mountain project. Here spatial memory is utilized for organizing
documents and there is clear observable place construction in resultant user
behavior. These insights offer predictive power for the developers of such systems.
Place construction is a key component in the virtual world, as well as the physical,
and designing with this understanding creates systems that are able to support
larger amounts of data, increased efficiency, and support of communication.

However, the focus is on how this will be designed for, not how to identify this and
make use of it within the system. There is not an active role in the system for place
identification and subsequent utilization of this information. These would be
systems that actively capture palatial determinations with the goal of reincorporating
them into the system.


Spacial Reasoning in Non-humans
Significant work has been done with regard to spatial understanding in systems less
vocal (and presumably less intelligent) than humans. From robots, seeking to
navigate unfamiliar environments with limited sensors, to rats moving through
mazes, the history of these efforts is rich. The focus here is usually on small-scale
space and (almost exclusively) on navigation. There is a strong focus in studying
information search that is relatively simplistic (such as pure retrieval for rats in a
maze) or where it can be clearly encoded (for robots).

Exemplar: Projects such as those proposed by Werner include navigating
wheelchairs and robot office navigation. These devices employ interesting
algorithms for the identification of features (corners, obstacles etc.) and serve as
useful aids in navigation and identification of basic spatial features that form the
core of visualizing small-scale spaces such as rooms or even buildings.

This kind of spatial work is interesting, and deserves consideration simply because
of the significant amount of time and effort that has been invested in it. However,
the differences between small-scale space and larger geographic space are poorly
understood and may be more profound than originally offered.


Common Sense Collection
While not intuitively obvious, common sense knowledge systems provide insight
into a new kind of approach. These systems attempt to capture common sense
facts about the world, similarly to how one might capture common sense
understandings of place.

Exemplar: Open Mind Common Sense is a system that depends on web-based
entry of structured common sense statements. These can be statements like “it is
cloudy when it rains.” While these statements are not always true, they often are (or
are often perceived casually by humans to be).

While systems like open mind offer an interesting approach, they rely on altruistic
data entry. They also tend to be less specifically focused on accounts of place (they
are usually more general, with specific persons or places rarely identified). Some
systems tend to be significantly more structured as well, relying on data input from
knowledge engineering rather than casual use. The primary focus should not be on
a special ‘place knowledge data entry’ but on a more flexible approach that can be
embedded in general spatial applications.

Here the focus becomes on implicit inference, and not data entry and collection.


Designing Principles:
Design Principles


Design Principles:
1. Think about place, not space.

Latitude and longitude can be precise to inches -
but what distinction matters?

Consider New York City directions and directions in the country.


Design Principles:
2. Get a room with a view.

Consider what the appropriate granularity is for information.

Can clustering and grouping show us more with less?

Use intelligent location awareness.


Design Principles:
3. Simplify the world, don't recreate it.

Focus the goal of the application.

We already live in the world -
we need to see less, not more, in our digital view of it.

If we could notice everything that was going on,
why would we need to look at a computer?


Design Principles:
4. Avoid the tyranny of the majority.

Spatial Application != Web Map + GUI (necessarily)

Choose representations that make sense in and of themselves and
which further the goal of the application.

Avoid the tyranny of the majority.


Design Principles:
5. Linking the virtual and the real.

There is a lot of information out there, but an incredible amount is
grounded in space.

Even when this is not a direct mapping, spatial relationships can
produce interesting associations.

Where do you blog?


Design Principles:
6. Some assumptions are inevitable.

The baseline of spatial data and location awareness is rising rapidly.

Assuming that all of the spatial data you ever want will be available is
not (so) unreasonable.


Design Principles:
7. Follow the tradition, Don't follow tradition.

Learn from the goals and methodologies of the tradition.
Understand the goals and background that create success.

And repeat.


Design Principles:
8. Transcend spatial limitations.

Space is a limiting factor, a good application should transcend this
limitation.

You don’t always need to be there to be there.


Design Principles:
9. Balance perspectives.

Spatial applications need to successfully balance a number of
perspectives - human cognitive perspective, individual perspectives,
and shared social perspectives.


Design Principles:
10. Spatial Applications are just applications.

The same rules of usability and design haven’t disappeared just
because we’re talking about space.


Concluding Remarks
& Questions

hock@media.mit.edu
www.spatialapp.com/

References:
1. Aggarwal, C.C., Yu, P.S. Data 12. Chen, G., Kotz, A Survey of 35. Hackos, J.A.T., Redish, J.C. User
Mining Techniques for Personalization. Context-Aware Mobile Computing 22. Egenhofer, M., Mark, D. Naive and Task Analysis for Interface Design.
IEEE Data Engineering Bulletin. Research. Dartmouth Computer Geography. In Frank, A.U. and Kuhn, IEEE Transactions on Professional
(2000). Science Technical Report. (2000). W., (Eds.) Spatial Information Theory: Communications. (1999).
A Theoretical Basis for GIS. Lecture
2. Agrawala, M., Stolte, C. Rendering 13. Cheverst, K., Davies, N., Mitchell, Notes in Computer Sciences (1995). 36. Hardt, S. Naive Physics. In
Effective Route Maps: Improving K., Friday, A. Experiences of Shapiro, S. (Ed) Encyclopedia of
Usability Through Generalization. developing and deploying a context- 23. Egenhofer, M., Frank, A.U. Object- Artificial Intelligence. (1992).
International Conference on Computer aware tourist guide: the GUIDE Oriented Modeling for GIS. URISA
Graphics and Interactive Techniques. project. In Proc. International Journal. (1992). 37. Harley, J.B., Laxton, P., Andrews,
(2001). Conference on Mobile Computing and H.J. The New Nature of Maps: Essays
Networking. (2000a). 24. Egenhofer, M., Freksa, C., Miller, in the History of Cartography. (2002).
3. Aime, A., Bonfati, F., Monari, P.D. H.J. (Eds.) Formalizing User Actions
Making GIS closer to end users of 14. Cheverst, K., Davies, N., Mitchell, for Ontologies. (2004). 38. Harrison, S., Dourish, P. Re-Place-
urban environment data. ACM GIS’99. K., Smith, P. Providing Tailored ing Space: The Roles of Place and
(1999). (Context-Aware) Information to City 25. Elias, N. The Civilizing Process. Space in Collaborative Systems. In
Visitors. In Proc. Adaptive Hypermedia The History of Manners. (1978). Proc. ACM Conference on Computer
4. Alexander, C., Ishikawa, S., and Adaptive Web-Based Systems. Supported Cooperative Work. (1996).
Silverstein, M. A Pattern Language. (2000b). 26. Entrikin, J. The Betweenness of
New York: Oxford University Press. Place - Towards a Geography of 39. Hayes, P. Naive Physics
(1977). 15. Churcher, N. Applications of Modernity. (1991). Manifesto. In Michie, D. (Ed.) Expert
Distortion-Oriented Presentation Systems in the Microelectronic Age.
5. Bleecker, J. A Design Approach for Techniques in GIS. In Proc. New 27. ESRI. The GIS Software Leader. (1978).
the Geospatial Web. Where 2.0. Zealand Conference on Geographical http://www.esri.com (2006).
(2005). Information Systems and Spatial 40. Hayward, W.G., Tarr, M.J.,
Information System Research. (1995). 28. Foresman, T. The History of GIS Wallace, W.P., Stewart H.L. Spatial
6. Bouquet, P., Warglien, M. Mental (Geographic Information Systems). language and spatial representation.
models and local models semantics: 16. Cohn, A.G. Qualitative spatial Prentice Hall Series in Geographic Cognition 55. (1995).
the problem of information integration. representation and reasoning Information Science. (1997).
European Conference on Cognitive techniques. Advances in Artificial 41. Heidegger, M. Poetry, Language,
Science. (1999). Intelligence. (1997). 29. Furnas, G.W., Morristown, N.J. Thought. Translated by Hofstadter, A.
Generalized fisheye views. In Proc. (1971).
7. Bourdieu, P. Outline of a Theory of 17. Couclelis, H., Golledge, R. G., SIGCHI conference on Human Factors
Practice. (1977) Gale, N., and Tobler, W. Exploring the in Computing Systems. (1986). 42. Hernandez, D. Qualitative
Anchor Point Hypothesis of Spatial Representation of Spatial Knowledge.
8. Bugayevskiy, L.M., Snyder, J.P. Map Cognition. Journal of Experimental 30. Giddens, A. The Constitution of Lecture Notes in Computer Science.
Projections. London: Taylor & Francis. Psychology. (1987). Society. (1984). (1994).
(1995).
18. Curry, M.R. The Work in the World 31. Goguen, J. An Introduction to 43. Hockenberry, M., Hoff, J., Selker, T.
9. Burrough, P. A. and Frank, A. U. - Geographic Practice and the Written Algebraic Semiotics, with Applications A Mindset for User-Centered Spatial
(Eds.) Geographic Objects with Word. (1996). to User Interface Design. In Nehaniv, Applications. Intl. Conference on
Indeterminate Boundaries. London: C. (Ed.) Computation for Metaphors, Spatial Cognition. (2006).
Taylor and Francis. (1996). 19. Cresswell, T. In Place / Out of Analogy and Agents. Lecture Notes in
Place. (1996). Artificial Intelligence. (1999). 44. Hockenberry, M., Selker, T. A
10. Burrough, P.A.; van Gaans, P. F. Sense of Spatial Semantics.
M.; Hootsmans, R. Continuous 20. Dudek, G., Jenkin, M.R.M., Milios, 32. Goffman, E. The Presentations of Conference on Computer Human
classification in soil survey: spatial E.E., Wilkes, D. Map Validation and Self in Everyday Life. (1959). Interaction (2006).
correlation, confusion and boundaries. Self-location in a Graph-like World.
Geoderma. (1996). International Joint Conferences on 33. Google. Google Maps. http:// 45. Hockenberry, M., Gens, R.
Artificial Intelligence. (1993). maps.google.com (2006). PlaceMap: Building Community
11. Camara, G., Souza, R.C.M., through Active Context Mapping.
Freitas, U.M. Garrido, J. Spring: 21. Egenhofer, M., Burns, H. Visual 34. Google Maps Mania. http:// Siggraph. (2005)
integrating remote sensing and GIS by Map Algebra: a direct- manipulation googlemapsmania.blogspot.com
object-oriented data modeling. user interface for GIS. In Proc. (2005). 46. Horowitz, B. Creators,
Computers & Graphics. (1996). Working Conference on Visual Synthesizers, and Consumers.
Database Systems. (1995). Elatable.com / Yahoo Inc. (2006).

References Design of Spatial Applications

References:
47. Janecek, P., Pu, P. A Framework 57. Kuhn, W. Ontologies in support of 69. Mark, D. Spatial Metaphors for 78. Neisser, U. Five Kinds of Self-
for Designing Fisheye Views to activities in geographical space. Human-Computer Interaction. Knowledge. Philosophical Psychology.
Support Multiple Semantic Contexts. International Journal of Geographical International symposium on spatial (1998).
Conference on Advanced Visual Information Science. (2001). data handling. (1992).
Interfaces. (2002). 79. Norman, D.A., Draper, S.W. User
58. Kuipers, B. Modeling Spatial 70. Mark, D. Counter-Intuitive Centered System Design; New
48. Janecek, P., Pu, P. Visual Knowledge. Cognitive Science. (1978). Geographic "Facts:" Clues for Spatial Perspectives on Human-Computer
interfaces for opportunistic information Reasoning at Geographic Scales. In Interaction. Lawrence Erlbaum
seeking. In Proc. of the International 59. Kuipers, B., Levitt, T. Navigation Frank, A. Campari, I., Fiormentini, U. Associates. (1986).
Conference on Human Computer and Mapping in Large Scale Space. AI (Eds.) Theories and methods of spatio-
Interaction. (2003). Magazine. (1988). temporal reasoning in geographic 80. Open Guides. OpenGuides: the
space. Lecture notes in Computer guides made by you. http://
49. Jankowski, P., Andrienko, N., 60. Lakoff, G., Johnson, M. Metaphors Science. (1992b). openguides.org (2006).
Andrienko, G. Map-Centered We Live By. (1980).
Exploratory Approach to Multiple 71. Mark, D., Comas, D., Egenhofer, 81. Pederson, E. Geographic and
Criteria Spatial Decision Making. 61. Lanter, D.P., Essenger, R. User- M., Freundshuh, S. Gould, M., Nunes, Manipulable Space in Two Tamil
International Journal Geographical Centered Graphical User Interface J. Evaluating and Refining Linguistic Systems. Spatial Information
Information Science. (2001). Design for GIS. Technical Report for Computational Models of Spatial Theory: A Theoretical Basis for GIS.
the National Center for Geographic Relations Through Cross-Linguistic Lecture Notes in Computer Science.
50. Jordan, T., Raubal, M., Gartrell, B., Information and Analysis. (1991). Human-Subjects Testing. COSIT. (1993).
Egenhofer, M. An Affordance- Based Lecture Notes in Computer Science.
Model of Place in GIS. In Proc. 62. Leung, Y.K., Apperley, M.D. A (1995). 82. Peng, C. E-Service as a New
International Symposium on Spatial review and taxonomy of distortion- Paradigm for Interactive
Data Handling. (1998). oriented presentation techniques. ACM 72. Markosian, L., Kowalski, M.A., Multidimensional City Modeling. E-
Transactions on Computer-Human Goldstein, D., Trychin, S.J. Real- time Technology, e-Commerce and e-
51. Kahana, M. The Neurophysiology Interaction. (1994). nonphotorealistic rendering. In Proc. Service. (2004).
of Human Spatial Navigation. Computer graphics and interactive
Psychology Science Convention. 63. Liu, H., Singh, P. ConceptNet: a techniques. (1997). 83. Prado, A.B., Baranauskas, M.C.C.,
(2004). practical commonsense reasoning Medeiros, C.M.B. Cartography and
toolkit. BT Technology Journal. (2004). 73. Maurer, D. Four Modes of Seeking Geographic Information Systems as
52. Kidner, D., Jones C. A Deductive Information and How to Design for Semiotic Systems: A Comparative
Object-Oriented GIS for Handling 64. Liu, H. Computing Point-of-View: Them. boxesandarrows.com (2006). Analysis. ACM Symposium on GIS.
Multiple Representations. In Proc. Modeling and Simulating Judgments of (2000).
International Symposium on Spatial Taste. (2006). 74. Mauss, M. Les Techniques du
Data Handling. (1994). corps. Journal de Psychologie. 84. Prasolova-Forland, E. Supporting
65. Lieberman, H. Powers of Ten Reprinted in Mauss, Sociologie et Social Awareness in Education in
53. Knez, I. Attachment and identity as Thousand: Navigating In Large anthropologie. (1934). Collaborative Virtual Environments.
related to a place and its perceived Information Spaces. Conference on (2002).
climate. Journal of environmental User Interface Software Technology. 75. Miles, S.B., Ho, C.L. Applications
psychology. (2005). Grounding for a (1994). and Issues of GIS as Tool for Civil 85. Relph, E. Place and
Computational Model of Place Engineering Modeling. Journal of Placelessness. (1976).
Matthew Hockenberry 66. Lurie, G., Sydelko, P., Taxon, T. An Computing in Civil Engineering.
Object-Oriented GIS Toolkit for Web- (1999). 86. Rivest, S., Bedard, Y., Marchand,
54. Kramer, J., Noronha, S., Vergo, J. Based and Dynamic Decision Analysis P. Towards better support for spatial
A user-centered design approach to Applications. Journal of Geographic 76. Montello, D. Scale and Multiple decision-making: Defining the
personalization. Communications of Information and Decision Analysis. Psychologies of Space. Spatial characteristics of Spatial On-Line.
the ACM. (2000). (2002). Information Theory: A Theoretical Geomatica. (2001).
Basis for GIS. Lecture Notes in
55. Koffka, K. Principles of Gestalt 67. Lynch, K. The Image of the City. Computer Science. (1993). 87. Robertson, G., Czerwinski, M.,
Psychology. (1935). (1960). Larson, K., Robbins, D.C., Thiel, D.,
77. Morville, P. Rosenfeld, L. van Dantzich, M. Data Mountain: using
56. Kuhn, W. Defining semantics for 68. MacEachren, A.M. How Maps Information Architecture for the World spatial memory for document
spatial data transfers. Proceedings of Work: Representation, Visualization, Wide Web: Designing Large-Scale management. In Proc. Symposium on
the International Symposium on and Design. (2004). Web Sites. (2002). User Interface Software and
Spatial Data Handling. (1994). Technology. (1998).


References:
88. Rodden, T., Benford, S. The 94. Siegel, A.W., White, S.H. The 101. Tollmar, K., Sandor, O., Schömer, 107. Werner, S., Krieg-Bruckner, B.,
evolution of buildings and implications development of spatial representations A. Supporting social awareness work Mallot, H.A., Schweizer, K., Freksa, C.
for the design of ubiquitous domestic of large-scale environments. Advances design and experience. (1995). Spatial Cognition: The Role of
environments. In Proc. CHI in child development and behavior. Landmark, Route and Survey.
Conference on Human Factors in (1975). 102. Traynor, C., Williams, M.G. Why Informatik. (1997).
Computing Systems. (2003). are geographic information systems
95. Silverman, K. The Subject of hard to use? In Conference 108. Whyte, W. City: Rediscovering
89. Rumbaugh, J., Blaha, M., Semiotics. (1983). Companion of Human Factors in the Center. (1988).
Premerlani, W., Eddy, F. Object- Computing Systems Conference.
oriented modeling and design. (1991). 96. Singh, P. The Open Mind Common (1995). 109. Worboys, M., Duckham, M. GIS:
Sense Project. KurzweilAI.net. (2002). A Computing Perspective. (2004).
90. Sarker, M., Brown, M.H. Graphical 103. Tuan, Y. Space and Place. The
fisheye views. Communications of the 97. Skyhook. Skyhook Wireless. http:// Perspective of Experience. (1977). 110. Zipf, A. User-Adaptive Maps for
ACM (1994). www.skyhookwireless.com (2006). Location-Based Services (LBS) for
104. Vertesi, J. Mind The Gap: The Tourism. In Proc. ENTER. (2002).
91. Schuurman, N. Trouble in the 98. Smith, B. The formal ontology of ‘Tube Map’ as London’s User
heartland: GIS and its critics in the space: An essay in mereotopology. In Interface. hciresearch.org. (2005). 111. Zubin, D. Natural Language
1990s. Progress in Human Geography. Hahn, L. (Ed.) The Philosophy of Understanding and Reference Frames.
(2000). Roderick Chisholm. (1994). 105. Virrantaus, K. Markkula, J., In Mark, D., Frank, A., Egenhofer, M.,
Garmash, A., Terziyan, V. Developing Freundschuh, M., McGranaghan, M.,
93. Shneiderman, B. Designing the 99. Stevens, A., Coupe, P. Distortions GIS-supported location-based White, R.M. (Eds.) Languages of
User Interface: Strategies for Effective in judged spatial relations. Cognitive services. Web Information Systems Spatial Relations: Initiative Two
Human-Computer Interaction. (1997). Psychology. (1978). Engineering. (2001). Specialist Meeting Report. National
Center for Geographic Information
100. Therborn, G. The Ideology of 106. Waquant, Loic. An Invitation to Analysis. (1989).
Power and the Power of Ideology. Reflexive Sociology (with Pierre
(1980). Bourdieu). (1992).


Design of Spatial Applications

Recomendados

Recomendados

Más contenido relacionado

Destacado

Destacado (11)

Similar a Design of Spatial Applications

Similar a Design of Spatial Applications (20)

Último

Último (20)

Design of Spatial Applications