The Geospatial Revolution ANZ 2010

The Geospatial Revolution Peter Batty December 2010 1

The Revolution 3

The changing nature of geospatial data 5

Crowdsourcing 6

The Cloudﬂic.kr/p/7Bzn5E 7

The Revolution 8

GIS has been a specialized backroom technology for many years 9

Players Utility GIS Smallworld Intergraph ESRI Network modeling Scalability Customizability Utility focus Non-network apps “Mainstream” core New technologies10

ﬂic.kr/p/7A6vVg 12

ﬂic.kr/p/7NEJzF Location is now Pervasive and Simple in consumer applicationsﬂic.kr/p/7rnNAD 13

Disruptive technology Functionality / performance og y h nol d tec h e a blis Est Mainstream Market requirements lo gy chno tiv e te ru p Dis Time 14

Consumer driven innovation 15

“But these new systems are justsimple web mapping, they’re not GIS” 16

Demo Ofﬂine 17

Mobile too!Runs on smart phones … and tablets including iPhone… including iPad 18

ﬂic.kr/p/7UjM5T 20

September 12, 2010 22

Microsoft Photosynth 29

Google Streetview 30

Microsoft “Street Slide”http://bit.ly/streetside 31

layar 32

Blaise Aguera y Arcas at TED 2010 Something on layarhttp://www.ted.com/talks/blaise_aguera.html 33

mapsManhattan 34

C3 Technologies Las Vegas 35

prototypegame.org Manhattan 36

The Sensor Web Need a spatial context to make sense of all this 37

Location sensingUWB GPS Wi-Fi RFID Cell towers 38

New TomTom trafﬁc speed dataset derived from 600 billion speed readings from users real time data within 3 minutes ﬂickr.com/photos/rutlo/3164449930/ 39

location based services are real at last! 42

Will have the ability to know where everything is - and what is happening - all the time 43

Crowdsourcing 44

Web Web 1.0 2.0publishing participation 45

Wikipedia 46

Hurricane Katrina New Orleans 47

Community generated data scipionus.com 48

OpenStreetMap 49

December 3, 2007Google OpenStreetMap July 7, 2009 50

200,000+ 300,000+ users 24m km of highways crazy 34m km of ways momentum!!OSM stats from May 2009NAVTEQ had 18m km of highways in Dec 2007 ﬂickr.com/photos/pimpmasterjazz/2601898276/ 51

What about quality? Dr Muki Haklay of UCL“OSM quality is beyond good enough, it is a product that can be used for a wide range of activities” Based on a detailed analysis http://tinyurl.com/mukiosm 52

Something on MapQuest / Microsoft OSM 53

Database 2007 data 69 countries 11m miles (18m km) of roads 18m points of interestPeople “Creating, maintaining and delivering a comprehensive, high quality map database is a Field force 700 multi-step, labor-intensive process. We Central production 270 currently employ over 270 employees in our centralized production facility and a global Technology 500 workforce of over 700 geographic analysts in 32 countries” Total 3349Financial Revenue $853m (~€604m) Data creation & distribution costs $396m (~€280m) 54

Crowdsourcing is a paradigm shift for data creation ﬂickr.com/photos/jamescridland/613445810/ 55

Running in the Cloudﬂic.kr/p/7Bzn5E 56

This is an old Googledata center (2005) Each container holds servers 1,160 This facility holds an estimated 150,000 servers 58

Google has an estimated 40 data centers… …holding an estimated 1 million servers 59

3 billionsearches per day 35,000 searches per second 60

2 billion videos per day 61

100 million monthly users 62

one zillion times more computer science PhDs than your company or mine has 63

HUGE economies of scale GoogleAmazon Your IT department 64

SaveBIG!! ﬂic.kr/p/64kFKH 65

EC2 PricingSeptember 5, 2010 66

My daily quotas (max $5 per day)46.5 CPU hours 43.2m web requests!6.5 CPU hours free 67

My total server costs for development of myWorld so far 13c 68

ﬂic.kr/p/4iCux7but is it SECURE??? 69

There is a strong case that your data is more secure in the cloud 70

AgilityGoogle Maps had 400releases in its ﬁrst 4 years ﬂic.kr/p/7BM8zr 71

Cloud benefits Ultra scalable Secure Major cost savings Hardware, admin, upgrades Much faster enhancements & fixesflic.kr/p/7Bzn5E 72

Summary 73

Fast train?ﬂic.kr/p/7b5sKd 74

ﬂic.kr/p/4vLS2o 76

Cloudﬂic.kr/p/6aJDqo 78

?peter.batty@ubisense.netgeothought.blogspot.com twitter.com/pmbatty 80

BackupDemo screen shots 81

Matches Smallworld styles 82

Google basemap gives context 83

Google satellite map 84

Display attribute data 85

Street View gives extra info 86

Google style one box search Something on “Google style search” 89

Autocomplete search 90

Autocomplete search 91

Reports in search too 92

Reports from search 93

Google geocoding very ﬂexible 94

Google local search 95

“Enterprise mashups” Outages Trucks WMS CIS Smart Grid 96

Ability to link to maps 97

Click link, no software needed! Demo 98

Different Approachesto integration ﬂic.kr/p/pWMRk 99

Sparse data Sparse data Trucks Outages Jobs Often don’t need detailed “GIS” maps KML and GeoRSS are good formats 100

KML is child’s play!<?xml version="1.0" encoding="UTF-8"?><kml xmlns="http://www.opengis.net/kml/2.2"> <Placemark> <name>Simple placemark</name> <description>Attached to the ground. Intelligently places itself at the height of the underlying terrain.</description> <Point> <coordinates>-122.0822035425683,37.42228990140251,0</coordinates> </Point> </Placemark></kml> 101

So is<entry> <title>M 3.2, Mona Passage</title> geoRSS <link href="http://example.org/2005/09/09/atom01"/> <id>urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a</id> <updated>2005-08-17T07:02:32Z</updated> <summary>We just had a big one.</summary> <georss:point>45.256 -71.92</georss:point></entry> 102

You can use KML in lots of places KML uploaded to GeoCommons and used for thematic mapping 103

KMLﬂic.kr/p/8eZBZk The Good Simple Good for sparse and dynamic data Works with multiple platforms / products 104

KML The Bad Hard to match GIS styles Hard to scale to large data volumesﬂic.kr/p/6hwSwn 105

Dense Data 106

Raster map tilesPre-render maps into image ﬁlesApproach used by Google, Microsoft, etcVery high performance and scalabilityEasy to integrate with Google Maps etc 107

“If a dataset available on the web is in a format that cant be indexed by Google, does it make a sound?” Kevin Wiebe Safe Software 108

…/a2e/data/datasources/Pole/90974REST APIs 114

…/a2e/data/datasources/Pole/90974?f=gjsonREST APIs: simple and powerful 115

…/a2e/data/datasources/Pole/search?f=gjson &lat=42.600&lon=-76.1780&d=4 116

REST is good!! <Picture of Jason> Jason Birch City of Nanaimo 117

“Neo GIS” examples 118

CartographyAndy Allan, Cloudmade 119

Data creation and maintenance Upcoming Mapzen editor Cloudmade 121

Data creation and maintenance “Walking Papers” for OpenStreetMap Stamen DesignHere’s a print of Chinatown, San Francisco. 122

Geospatial analysis Stamen Design 123

Geospatial analysis FortiusOne / GeoCommons 124

Usability ﬂic.kr/p/kjwKm 125

Usability ﬂic.kr/p/kjwKm 126

I highly recommendthis book 127

My top 3 rules for good usability1 Do usability testing 2 Do usability testing 3 Do usability testing 128

Doing a usability testPut application in Shut up Watch front of user flic.kr/p/7UqPo1 flic.kr/p/53WUkg flic.kr/p/72M6L 129

Don’t make me thinkKrug’s ﬁrst law of usability ﬂic.kr/p/28o3cz 132

The Geospatial Revolution ANZ 2010

by Peter Batty on Dec 09, 2010

Accessibility

Categories

Tags

Upload Details

Usage Rights

Statistics

The Geospatial Revolution ANZ 2010 — Presentation Transcript