Opening Session, Martijn van Exel
1
Tom MacWright, iD, a New Editor for OpenStreetMap
Track 1
iD is the new in-browser map editor for OpenStreetMap that focuses on usability and simplicity. It’s designed to radically improve the first-time editing experience while providing a fast and intuitive interface for anybody mapping on OpenStreetMap. This will session will demo iD and talk about the dynamics of building this editor, the roadmap ahead, and the potential for similar open source projects in the OSM space.
2
Patrick Wilson, Zombie Based Learning OSM Style
Track 2
Geography is becoming less and less emphasized in the Florida public school curriculum partly due to the standardized assessment system. In an attempt to inspire our youth and promote geography as a life skill, Patrick has spent some of his free time volunteering at local schools. In this talk he’ll talk about Zombie-based geography lessons that utilize OpenStreetMap data and other open tools he’s used in an effort to grow students’ interest in contributing to OpenStreetMap and learning how geography will benefit them in the event of a zombie apocalypse.
1
Coleman McCormick, Pushpin, Stats, and Mobile Editing
Track 1
In the fall of 2012, the mobile OpenStreetMap editor for iOS devices - Pushpin - was launched. It’s now been used by thousands of people in over 100 countries for casual editing, mapping parties, and humanitarian aid projects. Pushpin has lowered lower the bar for mappers, making it simpler for newcomers to make meaningful contributions. This talk will review some of the experiences of Pushpin’s development, statistics on its usage worldwide, and how Spatial Networks wants to grow the app in the coming year.
2
Nuala Cowan and Richard Hinton, I'll Grade Your Edits: OpenStreetMap in Higher Education
Track 2
For the last few years, George Washington University’s upper division GIS class has integrated OpenStreetMap assignments into the course curriculum. This past semester, the class worked closely with the International Services Division at the American Red Cross to map disaster prone communities in Columbia and Indonesia. These efforts were in conjunction with the National Red Cross chapters in these countries that are using the data toward disaster mitigation efforts. The students loved the assignment - contributing to real data.
Nuala and Richard are interested in formalizing the workflow developed to assign, operate, and grade this type of assignment for undergraduate classes. The result would be a small module that could be used in any undergraduate class (not just GIS or Geography) that promotes the collective creation of open source data. This session will discuss experimentations with various editors, task managers, and desktop GIS software to achieve this module, and will open the floor up ideas from you on how to make this better, more concise and more accessible.
1
Frederik Ramm, The Geofabrik Download Server
Track 1
Ever since the planet file became too large for the average end user to process, Geofabrik have made country extracts available for people to download. It was just a few countries initially, but meanwhile there are daily updates for most countries in the world, and even sub-regions for some of those where OSM is big. This technical talk explains the chain of open source tools that powers the download server, explains the problems and limitations of the approach used on that server, and compares it to other ways of producing daily or even sub-daily regional OpenStreetMap extracts.
2
David Emory, Adding Traffic to OpenStreetMap
Track 2
This session is all about using real-time GPS data to add traffic and roadway speed data to OpenStreetMap. Kevin will discuss traffic-engine, the open source platform for data processing and analysis. As a bonus, he’ll also walk through how to leverage large amounts of GPS data to improve OSM center lines.
1
Grant Slater, OSM Core Architecture and DevOps
Track 1
This session will provide an introduction to the OpenStreetMap operations team, the core OpenStreetMap setup, and its operational management. It will discuss the website and API architecture, including infrastructure, monitoring and stats, its frontend and backend, testing and deployment, and future plans. Grant will also briefly touch on the tile and caching architecture. If you’re interested in what powers OpenStreetMap and make it tick, come to this session.
1
Artem Pavlenko, The Secrets of Mapnik
Track 1
Artem Pavlenko, creator of Mapnik, will lift the hood on the open source rendering engine. This will be an almost non-technical snapshot of tricks that can lead to awesome performance for the user, debunking a few myths along the way. As planet OSM gets bigger, speed and scalability become paramount - but meanwhile users want more features. This session will look at ways to reconcile these two (or not), and where the balance should be. It will also examine possible directions the Mapnik project could take, with the aim of kickstarting a focused discussion in the OSM community.
2
Benjamin Blanchet, OpenStreetMap and Flight Simulation
Track 2
This session will explore the use of OpenStreetMap data in flight simulation software and specifically in osm2xp, software that injects OSM data into flight simulators. Benjamin will show the evolution of terrain representation in flight simulators and discuss how OpenStreetMap can improve the existing landscape in this area. He’ll show how close to reality you can get thanks to OSM, discuss the pros and cons of using OSM in flight simulation, and talk about the different approaches out there.
1
Andy Allan, Putting the Carto into OpenStreetMap Cartography
Track 1
Earlier this year, Andy embarked on a rewrite of the OpenStreetMap standard Mapnik stylesheets, porting them to the CartoCSS styling language. This opens the door for developing the styles using TileMill, as well as making them easy to customize for other projects.
In this session Andy will discuss some of the background behind the rewrite, tips that he’s picked up, tricks that were developed, and advice on the common pitfalls when styling OSM data in TileMill. He’ll also explain how to wrangle the stylesheets to suit your own projects and how to help further improve our cartography.
1
Saman Bemel Benrud, A New Vision for openstreetmap.org
Track 1
What could a ground-up redesign of the OpenStreetMap website look like? In this session, Saman will discuss the strengths and shortcomings of the current site design, present a vision for a modern, user-friendly, and responsive openstreetmap.org, and outline possible approaches for implementing such a redesign.
Among other things, the redesign will introduce concepts for strengthening community aspects of the site, improving the on-boarding experience for new users, and defining a more logical information architecture that makes the relationship between the community, data, and map layers of the site clearer. Saman will present both wireframes and fully rendered mock-ups illustrating his plans.
2
Christine White, I Dream of Data: Building Worldwide, Multi-Source Maps
Track 2
Start with local government content, add a dash of commercial navigation, a pinch of OSM, and wa-lah! You’ve got a recipe for creating the best in breed of authoritative maps. Simple? Not a chance. If your goal is to create beautiful, complete, and authoritative maps that describe essential data layers, using disparate sources is often required.
This session will discuss issues for achieving this goal - data quality assessment, metaphors for contribution, handling imports and data conflicts, and licensing paradigms. Christine will discuss the Esri Community Maps program, its results from a recent assessment of integrating OSM data into Community Maps, and ways OSM and Community Maps can learn from and benefit one another in the near future.
1
Tom MacWright and Eric Fischer, OSM Data Report: Visualizing Our Progress
Track 1
How complete and how well maintained is OpenStreetMap’s data around the world and in different cities? How has it grown over time? Who has been doing the work? What needs more attention? Tom and Eric have been working to answer these questions by visualizing the data behind OpenStreetMap contributions in helpful ways. They’ll present their findings and show the true state of OpenStreetMap’s data and identify areas where growth and improvement can be made moving forward to make OpenStreetMap a better map.
2
Andrew Hill, Using OpenStreetMap Data to Build Your Dynamic Maps
Track 2
Many projects today are collecting data continuously and want to build maps that show the latest trends. Whether the data is documenting street congestion, air quality, human movement, or a host of other subjects, this data can be combined with OpenStreetMap to build dynamic and real-time maps for the web. This session will demonstrate how CartoDB can be used to store, manage, and sort OSM data, and then be used to combine dynamic data with OSM layers to create real-time maps. While publishing dynamic maps on the web has traditionally been a technological challenge, CartoDB now makes it easier than ever before. CartoDB serves maps from a live database with on-demand filtering and styling. It also utilizes caching strategies that mean your maps are always fast but never get out of sync from the data on your account.
2
Steve Kashishian, Cohesive OSM Experiences Across the Screen and Paper
Track 2
In this session Steve will discuss his last year working as lead developer on an ambitious successor project to the popular Field Papers platform called CaerusGeo, currently being tested in Syria, Afghanistan, and Libya.
CaerusGeo is different than the typical web mapping platform in that an substantial amount of conceptualization and testing was undertaken to ensure its viability in “frontier” environments where analog mediums such as paper and verbal communication solely rule data collection. This talk will focus on the extensive amount of work needed to create an system that gelled with the environment’s workflow, as opposed to replacing it outright, and also the trials and tribulations experienced in integrating with Mapfish Print, an extremely powerful but lightly documented component for producing excellent wallmap and atlas PDFS.
Steve will also discuss the fork brought into the Mapfish codebase to meet the project’s needs, why he chose Leaflet for the frontend, and how contributing multiple pulls to that project has made him a better Javascript programmer.
1
Steve Gifford, Vectors, OpenStreetMap, and Mobile
Track 1
Vector maps are hot right now, with Google and Apple deployments in the news. On the web side, HTML5 and WebGL make vector maps workable and fast, respectively, and on the mobile side, we have the two above examples to show us it works. So what do we do for OpenStreetMap?
mousebird consulting makes an open source toolkit called WhirlyGlobe-Maply - WhirlyGlobe for the 3D interactive globe and Maply for the 2D map. Both are based on OpenGL ES, and both are very fast for iOS (iPad, iPhone) and (soon) Android. The openstreetmap.us server is returning vector based tiles for several levels. This gives us a nice, semi-official platform to try out high performance vector map display on iOS. This session will present a simple experiment in that area. It will take a lot of work to match the quality (Apple jokes aside) of the commercial vector maps. Our hope is that a nice framework for fetching and displaying vector tiles can nudge us further along that path.
2
Eric Fischer, Updating OpenStreetMap's TIGER Imports to TIGER 2013
Track 2
A lot of the OpenStreetMap data in the United States was originally imported from the U.S. Census Bureau’s TIGER map data. Unfortunately at the time of the import, TIGER was in the middle of a multi-year project to correct misalignments and omissions, and most of these corrections have never been applied to OpenStreetMap. This talk is about Eric’s attempt to merge TIGER updates into OpenStreetMap and the things that make doing this difficult.
1
Dane Springmeyer, Vector Tiles for Distributed, High Performance Rendering of OSM
Track 1
The potential for custom and real-time maps from OSM is growing fast, but the size and complexity of the data increasingly limits new and creative uses. Many seek to craft beautiful maps from OSM only to find their energy and time mostly taken by importing the data into a database and massaging it into manageable pieces.
A powerful solution to this problem is to allow the rendering of OSM data directly from vector tiles. Vector tiles combine the strengths of image tiles - fast, cacheable, and easy to distribute - with the flexibility of source vectors - high resolution, interactive, and renderable in infinite ways. Supporting multiple named layers and containing a space efficient serialization of geometries and attributes, vector tiles once created can be cached and distributed to any number of map rendering clients.
In this session Dane will describe a new high performance and open source library for creating vector tiles and rendering vector tiles to various formats using Mapnik. Mapnik in this context becomes both server– producing vector tiles on demand - and client - rendering beautiful maps from vector tiles. After covering the details of how to write your own software to use vector tiles, Dane will demo an experimental version of TileMill that leverages this technology to enable immediate styling of a global set of OSM vector tiles.
2
Serge Wroclawski, The U.S., Imports, and Bots, Oh My!
Track 2
Imports are a sensitive subject in OpenStreetMap, usually frowned upon, often blamed, and not spoken about in polite company. In this talk, Serge will describe the changing landscape of U.S. imports and bot efforts, from manual imports to automated bots, and tool standardization.
1
Aaron Ogle, Walkshed.js: Client-side Raster Processing with OSM Data Tiles
Track 1
OSM is commonly used for map tile generation and routing, but much less so with data rasters. Rasterizing OSM data can open up a whole new set of mapping and analytical possibilities using map algebra. Walkshed.js is a project combining OSM data tiles and cost-distance raster calculations to measure pedestrian access for a location. This talk will be a technical dive into creating OSM data tiles and the JavaScript code used to do raster processing in the browser.
2
Paul Norman, Address Import Review
Track 2
In 2010 an address import was done in Surrey, BC, but no one has yet reviewed it to determine if it has been a success. This talk will look at how the import was done, examine how the data has been integrated with other OpenStreetMap data, and ultimately used by mappers.
1
Jacob Beaudoin, Pure Vector Pathways for OSM Display
Track 1
Since its inception, the primary display methodology for OSM data has been raster tiles. There are several problems with this approach, including size and inflexibility at runtime. This session will explore a pure vector pipeline for handling and displaying OSM data on mobile and desktop platforms in a performant way. Advantages include smaller file sizes, the possibility of on-device data, instant restyling and dynamic label placement.
2
Jeff Meyer and Clifford Snow, Using Imports to Build Community: The Seattle Import Case Study
Track 2
This session will discuss some of the traditional concerns with imports and review how the Seattle import into OpenStreetMap was designed to address those concerns and improve OSM map quality and community participation in the Pacific Northwest.