Sessions at Where 2012 on Tuesday 3rd April

Opening remarks by Brady Forrest, Where Conference program chair.

Our personal devices provide endless streams of data set in context of who we are, where we are, who we know, and what we do. But what can we realistically expect the future to look like, and how soon will it be before it gets here? The key is to understand what new opportunities are unleashed through the combination of these different contextual data – and overlay what is probably given business model, social, technology, and political constraints. We’ll also look at what this brave new world means in terms of actions you must take today.

Open source tools let you design fast and beautiful interactive maps using your own data and share them on the web and mobile. This keynote will be a walk through showing how to use TileMill, and how it integrates with the web. Eric will take you from a spreadsheet to a custom designed map and then share it from a cloud map hosting service using embeddable widgets and the MapBox API.

Responsive Design is all about how structure can adjust to various environments, user activities and form factors. First we’ll look at the core concepts of Responsive Design and how they’re applied in a typical web setting, noting the methods used. Then we’ll see how maps naturally follow some of those basic principles. For example, maps alter both design and information based on scale – not unlike adjusting detailed website information for small screen display.

An overview of the technical status of online mapping shows that the move away from a “baked” raster tileset to a vector-based world of geographic data opens the door to even more flexible display and response. What is particularly exciting is how Responsive Design might be applied to a wide variety of environmental and behavioral factors. We’ll take a look at some of these factors beyond just adjustment for screen size and orientation. For instance, consider factors like: speed (main roads and type becoming more prominent as speed increases); user input (map detail responding to touch gestures); actual physical environment (location-based weather and seasonal information altering the look of the map); moods and modes; and various location-based activities (when you’re in a park, showing contour information). Each factor/response demonstrates how the design and display of geographic information might shift and respond to provide the best map to meet the moment.

Lots of visuals.

You see satellite imagery on Google Maps all the time, but how often do you stop to admire the amazing structures, patterns, and colors of our planet? Stratocam is a new web app, described as “Hot-or-Not for maps,” that lets you discover and vote on the best satellite imagery around the world, and take your own snapshots of the planet for others to see.

If you notice a black box hidden behind a plant while you’re here, don’t be alarmed. We’re capturing Media Access Control (MAC) addresses to track location and flow patterns of Where Conference attendees. This keynote will explain exactly what we’re doing and how, data privacy precautions, as well as visualization and open data use cases.

For decades, planners and engineers have been painstakingly hand-counting people and vehicles in handfuls of locations for short windows of time in attempts to understand how people use space and how it can be adapted to be more comfortable and effective for different user groups.

The past few years have seen a revolution in cheap GPS receivers and mobile data devices, so now, to augment these small, scattered surveys, we are awash in noisy but vast data from geotagged messages and photos revealing where hundreds of thousands of people were at different times on different dates and something about what they were doing there. At the same time, public location feeds for vehicles provide a constantly updating picture of traffic conditions at thousands of locations.

In both of these real-time sources, for most places, the same patterns repeat over and over, so when the pattern for a place changes, it means something significant has happened there. A physical or social change has somehow made the place more or less popular with people, or has reallocated road space to a different mode, or has changed the speed or pattern of traffic.

By knowing that a change has happened, we can then hope to learn why it happened, and, if it is a change for the better, apply the lesson to other places, and if for the worse, avoid making the same mistake again. The dream is to learn how to best balance the needs of different transportation modes, including, critically, pedestrians, and to understand from the past what effect can be expected from different decisions in the future.

We all love maps, because maps are great. That’s why we’re here, right? But it turns out that sometimes maps aren’t the right answer when it comes to visually presenting data with a spatial component. Noah Iliinsky will discuss why, and how to figure out when to not map your data. And of course some examples will be shown where a better choice is a non-standard map, and some where the best representation doesn’t involve a map at all.

Balloons are a central tool in the Public Laboratory mapping kit. Mathew Lippincott will demonstrate a camera-bearing helium balloon that is small enough to fly without prior FAA clearance to altitudes up to 4000 ft. It is an approachable, inexpensive way for civic organizations to document their events and environments. Small-scale mapping offers communities high temporal as well as spacial resolution, capturing phenomena ranging from Deepwater Horizon oil to vegetation growth to occupy camps.

Today’s smartest brand marketers get the importance of speaking to people on their own terms. The ideal cocktail of location-based messaging includes a dash of permission and a strong dose of context to turn a one-way push message into a two-way dialogue. We’ll dig into how adding in customer preferences, demographics, environmental elements, along with location will take mobile messaging into new dimensions.

Dwolla’s architecture, technologies, accessibility and even its price point, serve as the foundation for what the startup believes to the payment network of the 21st century. Ben Milne, founder and builder at Dwolla, will elaborate on how new advancements, like location-based technologies, are helping pave the way a chance at ubiquity.

Think back 10 to 15 years ago, there was probably a kid in your neighborhood that you could pay a couple bucks to wash your car or mow your lawn. We’ve lost that sense of community over the years because the age of the internet has siloed us. With the social networking in full force, that is changing. Social, location, and mobile technologies are creating a meaningful foundation to connect a neighborhood both on and offline. Technology has finally caught up to human behavior, and we are able to use the internet to get off the internet and build a strong community together. Leah will discuss how these technology trends are paving the way for new marketplace businesses and why they are so important to the success of these models.

Disease, especially infectious disease, has always been dependent on location. The first quarantines occurred in 14th century Venice, requiring visiting sailors to spend 40 days (‘quaranta’) aboard before they could disembark, lest they spread ills from some distant land. And when syphilis began to march across Europe in the late 15th century, it was first known in Italy as the French Disease (thanks to French soliders), while in France it was called the Italian Disease. Meanwhile, in Holland it was the Spanish Disease; in Russia, it was the Polish Disease, and so on. (In fact, the disease most likely came over from the Americas in 1494, one of less auspicious of Columbus’ discoveries.)

Even before the Germ Theory was developed in the 1870s, it was impossible to separate disease from place. When an outbreak of disease occurred, it created, all at once, a new map that created new rules for trade, politics, and society. But not all infections were outbreaks – the slow, unstoppable march of tuberculosis changed the perception of space and logistics on a different pace, creating a less sudden but no less significant dislodging of social conventions and forging new patterns of migration and development.

This talk will explore the legacy of infectious disease on our perceptions of geography and space. It will distinguish between the “fast maps” that came with outbreaks, and the “slow maps” that emerged as entire nations tried to outrun a ferocious killer like TB. And it will connect these fast and slow maps to our contemporary quest to eliminate infectious disease altogether – tracking it down town by town until it fades from the map altogether.

Geo-data isn’t just for supporting decisions and creating actionable interfaces. Geo-data can create nuance, giving new understandings that lead to further questioning—rather than just actionable decisions. In particular, curiosity, and creative thinking can be driven by combining different data sets and techniques to develop a narrative around a set of data sets that tells the story of a place—the emotions, history, and change embedded in the experience of the place.

In this session, we’ll see how far we can go in exploring one street in San Francisco, Haight Street, and see how well we can understand it’s geography, ebbs and flows, and behavior by combining as many data sources as possible. We’ll integrate basic public data from the city, street and mapping data from Open Street Maps, real estate and rental listings data, data from social services like Foursquare, Yelp and Instagram, and analyze photographs of streets from mapping services to create a holistic view of one street and see what we can understand from this. We’ll show how you can summarize this data numerically, textually, and visually, using a number of simple techniques.

We’ll cover how traditional data analysis tools like R and NumPy can be combined with tools more often associated with robotics like OpenCV (computer-vision) to create a more complete data set. We’ll also cover how traditional data visualization techniques can be combined with mapping and augmented reality to present a more complete picture of any place, including Haight Street.

Data is becoming more and more prevalent. But neither all data is created equal, good, nor useful. The age of data and algorithm has come, where data mining and machine learning become necessities for all sorts of business, education, research and commerce.

Commerce has traditionally been “offline” and still super majority of transactions take place in retail stores than online. But the behaviors of consumers are changing as devices get more connected and mobile, and where information is readily available anytime, anywhere, any way.

To win in this new world of commerce, where the lines between online and offline are blurred, new tools of trade become necessary for survival.

In this talk, we discuss big data, algorithms, and use cases necessary to win in the new age of commerce.

The tablet market has exploded with over 60 millions units sold to date. The result is a new type of user experience that goes beyond smartphones and desktops. Some call it the “curl up” experience, in which users both absorb and engage content. It’s an experience driven by touchscreens, screen size and connectivity, and it unlocks new and exciting opportunities for map-makers. We will discuss examples how some of the core features of mapping - navigation, search, and cartography - are being upgraded for the tablet experience, and what new mapping concepts can be brought into play on these devices.

Evan is a software engineer on Google Maps. Evan joined Google in 2004 and worked on the Ads team for 2 years before joining the Google Maps team in 2006. Since then Evan has worked on Street View, the integration of the Google Earth plugin into maps.google.com, web app performance, and a mobile version of maps.google.com.

Evan has a bachelors and masters in computer science with a focus in computer graphics from Stanford University.

The future of mobile payments contains real excitement: not around simple convenience, but rather the return of purchase history back to the hands of the user. While the commercial implications are of course enormous, the geotechnical potential – employing an artifact of the payments process – is huge. This brief talk aims to provide an overview to some of that hugeness, and discuss what how the game is taking shape on the table.

A single mobile payment records the time, date, and place of purchase (like any purchase today), but doing so on a mobile device provides the means to record that transaction and re-purpose it for further user benefit outside the commercial domain. The next generation of geo applications will all be ‘Big Data’ apps: using data for a purpose orthogonal to its purpose at collection, and enriching it to become more than a sum of its parts. We’ll see mobile apps that check-in, comb for deals, and geolocate people with every purchase, to start.

This talk will introduce the idea that the future of geo lies in mobile payments. It will introduce some of the interesting players on the board now (who may not see themselves as geo companies) will become increasingly so in the future. The role of Data as a crucial ‘technological enabler’ will be, perhaps unsurprisingly, the tenor of the entire talk.

The confluence of social media, mobile computing, the democratization of data and technology, and government transparency leaves us in the midst of an information tsunami of sorts. The tidal wave of data has many catalysts: passive broadcasting of geographic data, the phone as probe (gyroscope, accelerometer, compass, proximity, ambient light, distance/elevation, location, audio, video, time), and openness in government initiatives (opening access to location-based data, including crime reports, macroeconomic data, and other datasets). From mass transit systems to open (or proprietary) formats, lawsuits to metadata, the intersection of GIS and public policy looks messy.

Unanswered and intersecting questions abound: what geodata should be available, to whom, and under what conditions? How does the “firehose” of data affect reasonable expectations for government transparency? Who truly “owns” this data, and what is the appropriate means of distribution? With real-world experience standardizing and sourcing unique sets of geodata, Urban Mapping CEO Ian White will provide an updated perspective on the good, the bad, and the ugly from his company’s work in the trenches. The presentation will also highlight emerging trends, recent advances, and hopes in public access and bureaucratic morasses.

“But I thought indoor GPS doesn’t exist.” If you work in the location-based services industry, you’ve probably heard that sentiment at least daily. In a 15-minute session, Nick Farina will discuss the technical, design, and infrastructure challenges of indoor navigation, as well as how some mobile developers and IT teams have overcome them. Borrowing from 10 years experience in mobile development for navigation apps — namely his experience developing the first indoor GPS-like app with Cisco for the American Museum of Natural History — Farina will review a number of successul strategies for helping visitors navigate indoor locales.

StreetEasy rolls their own maps. They depend on OSM data and a custom tileset served by Mapbox.

Custom map tileset, covering NYC (all 5 boros) zoom levels 10 to 19
- developed in-house using TileMill
- data from OSM, NYC.gov, and our own
Maps served from MapBox
- Using their server-side map composition tools
- Serving about 2M tiles per day (for 500K-700K pageviews)
Maps embedded using Leaflet
JS generated via our own ruby abstraction classes used on a rails app.
Marker clustering done server-side with custom ruby/rails code, updated via ajax
Custom polygon editing, adapted/ported to leaflet (and in the process of being cleaned up for contribution back into leaflet)
Geocoder built in-house (using, yes, you guessed right, ruby), based on NYC.gov data plus our own updates

GeoIQ provides real-time exploration and analysis of millions of geospatial features. The technology stack combines open data standards with the development speed of Ruby on Rails, the asynchronous processing of node.js and our own highly scalable data storage - deployable horizontally across EC2 or on laptops in Afghanistan.

The proliferation of publicly accessible real time data streams has given us a new substrate to create compelling visualizations. We can now map the public’s thoughts, opinions, and beliefs across time and space. This has lead to a renaissance in cartography, but we still use classical techniques to analyze the data fueling our beautiful maps. The answers we generate with real time data are currently snap shots in time that loose their currency as soon as they are generated. How do we coax our analysis into keeping up with our data?

Crowdsourcing made us embrace data fluidity and disrupted the market. What now happens when your analysis become fluid – creating new answers as the global pulse shifts across billions of interconnected devices? How do we handle an ever changing sample size that is unevenly distributed across geography and time? Academia, large corporations and start ups are all starting to make forays into a world were analysis can no longer be a batch job. We’ll cover those endeavors and dig into some case studies on how the spread of the Arab Spring was analyzed and how Pepsi’s managed real time reaction to their Superbowl advertisements as examples of real time analysis in action.

The proliferation of publicly accessible real time data streams has given us a new substrate to create compelling visualizations. We can now map the public’s thoughts, opinions, and beliefs across time and space. This has lead to a renaissance in cartography, but we still use classical techniques to analyze the data fueling our beautiful maps. The answers we generate with real time data are currently snap shots in time that loose their currency as soon as they are generated. How do we coax our analysis into keeping up with our data?

Crowdsourcing made us embrace data fluidity and disrupted the market. What now happens when your analysis become fluid – creating new answers as the global pulse shifts across billions of interconnected devices? How do we handle an ever changing sample size that is unevenly distributed across geography and time? Academia, large corporations and start ups are all starting to make forays into a world were analysis can no longer be a batch job. We’ll cover those endeavors and dig into some case studies on how the spread of the Arab Spring was analyzed and how Pepsi’s managed real time reaction to their Superbowl advertisements as examples of real time analysis in action.

This session encompasses speakers and content from the following sessions:

Tiling GeoData with TileStache
Zain Memon, Trulia
TileStache is an open source Python app that serves map tiles. It can create image, vector, or data tiles from sources like a Mapnik config, a PostGIS database, a shapefile, or another Python application. In this talk, I’ll cover some of the ways I’ve used TileStache to trivialize the task of accessing geodata.

Popularizing Satellite Data for Crisis Mapping
Georgy Potapov, R&D Center SCANEX
Vladimir Ermakov, University of Carnegie Mellon
Accurate real-time data is crucial for effective response in crisis situations. Access to satellite monitoring data is too difficult to obtain in time required from specialized services designed for the scientific community. This talk describes a web-based solution that aims to provide near-real data about wildfires and other natural disasters for use in crisis mapping and emergency response.

CartoDB: How Working With Geospatial Data Can Be A Joy and Not A Pain
Javier de la Torre, Vizzuality
CartoDB makes beautiful, fast online maps powered by PostGIS and Mapnik, with easy import tools and a user friendly interface. In this workshop you’ll produce insightful, thought provoking maps of police activity, and along the way will learn how to use data from open data repositories, perform powerful spatial analyses, and share a stunning map online.

Incident Action Plan Project (IAP Project)
Sam Lanier, FireWhat Inc.
In 2009, Wildfires accounted for nearly 3.7 Billion in Federal Disaster Funds. The Incident Action Plan Project moves to automate how Emergency Personnel respond to such catastropic incidents. The project aims to eliminate the antiquated practices and utilize existing data sources to streamline Crisis Mapping for Emergency Responders by use of readily available technology.

ql.io
Jonathan LeBlanc, X.commerce
ql.io is a declarative, evented, data-retrieval and aggregation gateway for HTTP APIs. Through ql.io, application developers can increase engineering clock speed and improve the end user experience. ql.io can reduce the number of lines of code required to call multiple HTTP APIs while simultaneously bringing down network latency and bandwidth usage in certain use cases.

ql.io consists of a domain-specific language inspired by SQL and JSON, and a node.js-based runtime to process scripts written in that language. ql.io is an open source project available on Github and demos, examples, and docs are available on the main site.

Web-based interfaces cover a spectrum between discovery through active intent (search) to more passive intent (browsing). But as we share more real-time passively gathered data, in particular location, through our mobile phones, service providers have a more granular awareness of their users’ real-time situation, opening up a deeper and more fluid understanding of context.

As service providers transform this situational awareness into an understanding of proper timing for certain opportunities, they can offer value that won’t wait for a user to check his feed—new opportunities that are even more passively triggered than traditional discovery and browsing sessions. To deliver this value, we’ll see more emphasis on revealing opportunities serendipitously through push notifications. Balancing contextual signals such as location, availability, proximity, interest relationships, social relationships, and social context is the key to engineering serendipitous delivery to maximize effectiveness and delight.

Through examples culled our two years of experience making real-time mobile introductions between nearby people who share common interests, I’ll share our best practices for wielding this powerful dataset to deliver situationally relevant serendipitous opportunities and orchestrate small-world moments. When availability, location accuracy, and social context must be taken into account, the most challenging factors can be knowing when, in what form, and how often to reach out. Attendees will take away lessons learned from managing multiple location sources, determining user availability, interpreting proximity in different urban densities, and variations in response rates by situation.

Uber is an on-demand transportation service. We use node.js to handle thousands of requests per second and utilize all of this data to create new geospatial prediction algorithms.

360 Cities runs 360cities.net and panomonkey.com. We make use of both stable long-lived VPS and shorter-lived EC2 instances by dividing up the system into pieces that fit each usage pattern. We have also found challenges while working with our 10s of Terabytes of panoramic imagery.

CartoDB is a new Open Source Geospatial Database in the cloud. It leverages the full power of PostGIS and Mapnik to effortlessly stand up the services needed to create maps and develop location aware applications. It has been built with a love for design and interaction so that playing with geospatial data is a joy and not a pain.

The software has been developed by Vizzuality, a data visualization company focus on geospatial analysis and visualization. CartoDB is offered as a free service, on dedicated cloud instances or as an elastic service enabling users to process large amounts of data and only pay for what they use.

Importing geospatial data is as simple as dragging files into it, editing the data is just the way it is supposed to be, the maps are fast and responsive, it just works! Plus set the tables to real time and you will be able to have real time visualizations on your data changes.

Participants on this workshop will learn how to do beautiful and speedy maps using CartoDB. We’ll start with a file of 600,000 “stop and frisks” carried out by the NYPD, and make sense of the data using CartoDB’s analysis and mapping tools, including thematic mapping by block.

Google Earth as a geo-browser; new features and lesser-known gems: We will talk about the latest capabilities of the Google Earth desktop and mobile clients, as well as review several of the lesser-known yet powerful features such as touring and historical imagery.

Geo-browsing on Google Earth mobile: mobile is the future of geo. We’ll talk about how to find, share, and view geo-content on both iOS and Android devices.

Advanced features in KML: KML is a standard for creating and sharing geo-content. We’ll cover some of the newer Google additions to this powerful language.

3D and the web: the Google Earth API gives you easy access to the power of Google Earth, in your own web application. We’ll highlight some of our favorites and talk about how to create a compelling 3D web application.