MapBox - Development Seed The latest developments in our open source mapping stack. en Welcome Anna Scalamogna <figure class="align-center"> <img src="/assets/graphics/content/anna-welcome-card-card.jpg" /> </figure> <p class="dropcap"><a href="">Anna Scalamogna</a> just joined our team, and we are more than thrilled. Anna builds <a href="">scalable interfaces</a> and <a href="">space-age designs</a>. We also think she’s a <a href="">talented mapmaker</a>.</p> <p>As a developer and designer, Anna will be wrangling data and working across projects here at Development Seed to make intuitive, beautiful tools. She comes to us by way of National Geographic, where she designed interactive graphics for all facets of the magazine and website.</p> <p>Anna studied graphic design and journalism at American University. Chat graphics with her at <a href="">GitHub</a> and <a href="">Twitter</a>.</p> 2015-08-28T12:00:00+00:00 Development Seed Selamat Datang Lauren Jacobson <figure class="align-center"> <img src="" /> </figure> <p class="dropcap"><a href="">Lauren Jacobson</a> is now a member of the Development Seed team! We initially met Lauren at OSM’s State of the Map where her love of maps, mapping and open source were immediately obvious. Her dedication to others is also evidenced by her time spent as a Fulbright Fellow in Indonesia as well as her active involvement in the local community, helping to support the awesome <a href="">Women Who Code</a> program here in DC.</p> <p>Lauren will be utilizing the full extent of her flexible skillset working on things ranging from OSM data imports to frontend site implementation to data ingest scripting. We’re excited to have her join the team and make an immediate impact on our projects.</p> <p>You can find Lauren on <a href="">Twitter</a> or <a href="">GitHub</a> and out and about at various mapping and open data events in the area.</p> 2015-08-28T10:00:00+00:00 Development Seed Introduction to the geo command line <p class="dropcap">I used to fear the command line. As a cartographer, time and again I elected to use software with a visual interface - from QGIS to Mapbox Studio. But I kept running up against situations when I would look up how to do something and the advice told me to turn to a tool that was only available on the command line. And I kept running.</p> <p>But, as Zed Shaw eloquently states in his <a href="">command line crash course</a>:</p> <blockquote> <p>If you have an irrational fear of the command line, the only way to conquer an irrational fear is to just shut up and fight through it.</p> </blockquote> <p>This post is for people like me, who generally understand mapping and GIS, but don’t know much of anything about the command line. I’ll talk about why you should care about it, cover some of the basics, and go through a quick example of how to use two powerful command line tools for manipulating geodata - ogr2ogr and GDAL. I’ll leave you with other command line resources for further exploration.</p> <h3 id="what-is-the-command-line-and-why-should-you-care">What is the command line and why should you care?</h3> <p>The command line (also known as command line interface, or CLI) gives you access to your computer’s innards. It’s how you talk to your computer directly, without going through the intermediary of a graphical interface. You type things in, and the computer responds by typing things out on the screen. A <em>shell</em> is the interface where this typed interaction between human and computer takes place. On Mac and Linux, the default shell is called bash, which you access through the Terminal application. On Windows, it’s PowerShell.</p> <p>Attempting to turn your computer from a magical black box into something that you have more control over can be intimidating. But by learning the command line, you have access to geo tools that are only accessible there, or you can do your work much faster than through a heavy software program. At first, it might feel like you are blindly following instructions, but with more practice you’ll start to feel like you understand more of what you are doing.</p> <blockquote> <p>Why? Because if you want to learn to code, then you must learn this. Programming languages are advanced ways to control your computer with language. The command line is the baby little brother of programming languages. Learning the command line teaches you to control the computer using language. Once you get past that, you can then move on to writing code and feeling like you actually own the hunk of metal you just bought.</p> <footer>Zed Shaw, <cite title="The Command Line Crash Course"><a href="" title="Visit The Command Line Crash Course">The Command Line Crash Course</a></cite></footer> </blockquote> <h3 id="getting-started">Getting started</h3> <p>First, you’ll need to <a href="">install GDAL</a> so that down the road we can do some geoprocessing with it. GDAL and ogr2ogr are two amazing command line utilities - GDAL can be used for processing raster data, while ogr2ogr is for operations with vector data types (<a href="">read more</a> about vector vs. raster data types).</p> <p>Now let’s get started on the command line. You’ll need to open up your shell. On Mac, press <code>⌘ + spacebar</code> to open your computer search, type in <code>Terminal</code> and press <code>return</code>. On Windows, open up PowerShell. (I am working on a Mac, so that is where all of my screenshots and examples come from.)</p> <p>Once it’s open, you will see something that looks more or less like this, but with the name of your username instead of mine, followed by <code>$</code>. On Windows, you will see something similar but followed by <code>&gt;</code>.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-01-terminal.png" alt="The Mac Terminal" /> </figure> <p>We will use the shell to access different files and programs on your computer. So, first we need to find which directory (aka folder) we are in. Type <code>pwd</code> into your shell. This means <code>present working directory</code> and tells you the location that the shell is currently accessing.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-02-pwd.png" alt="pwd on the command line" /> </figure> <p>The shell prints out <code>/Users/robin</code>, telling me that that’s the directory I’m currently in.</p> <p>Now, I want to see everything that’s in the <code>/Users/robin</code> directory, so I type<br /> <code>ls</code> and get this on the screen:</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-03-ls.png" alt="ls on the command line" /> </figure> <p>Now we see the Desktop in there. To get to any directory or file within the present working directory, you just type in <code>cd</code> and then the name of the directory or file. <code>cd</code> stands for <code>change directory</code> and this allows you to move around to a different directory on your computer.</p> <p>So for me, I just type in <code>cd Desktop</code> (Side note: you can use the <code>tab</code> button as a shortcut - so if I type the first couple of letters of <code>Desktop</code> and then hit tab, it will autofill with the full directory name so I don’t have to type it out completely). Now when I type in <code>pwd</code>, the shell prints out <code>Users/robin/Desktop</code>, showing me that I’m inside my Desktop directory:</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-04-cd.png" alt="cd on the command line" /> </figure> <p>When navigating around your directories, <code>..</code> means <code>parent directory</code>. Now that I am in the Desktop, if I want to go back to <code>Users/robin/</code> I just type in <code>cd ..</code> and it takes me up to the “parent” of my current directory, meaning the directory that my current directory is within.</p> <h3 id="actually-doing-something-geospatial-using-gdal-and-ogr2ogr">Actually doing something geospatial using GDAL and ogr2ogr</h3> <p>We’re going to convert a polygon shapefile to a GeoJSON format and then clip a satellite photo to the bounds of that polygon, all using GDAL and ogr2ogr, which you downloaded earlier. You can download the <a href="">zipped shapefile</a> and the Landsat <a href="">satellite image</a>. We’re looking at Dane County, Wisconsin (because Wisconsin in the summertime is lovely).</p> <p>Go ahead and download those files, <code>wisconsin.tif</code> and <code></code>. They’ll probably be in your Downloads folder. In my case, I’ve placed them on my Desktop. Unzip the zipped shapefile, <code></code>. It will create its own folder within whatever directory the zipped file is in, so keep that in mind when you are doing the following commands.</p> <p>First, we’re going to convert the shapefile into a GeoJSON. You’ll need to navigate to the folder where your shapefile is saved using the <code>cd</code> command to change directory and <code>pwd</code> to check and make sure you’re in the right place. You can always use <code>ls</code> to list the files in that directory and make sure they’re in there. Then you’ll type in this command:</p> <p><code>ogr2ogr -f "GeoJSON" dane-county.json dane-county.shp</code></p> <p>Let’s break that down. <code>ogr2ogr</code> is the command you’re using, which you have access to because you installed GDAL above. As is the case with many command-line utilities, after the command you have some optional flags. These are extra conditions or parameters that you can set to tell the program how to do what you want it to do. In this case, we are using the <code>-f</code> flag, which tells the command that we want to specify a filetype that we want the output to be saved as. After the <code>-f</code> flag, we specify the filetype, in this case, GeoJSON. Then we put in a name for the output (I put <code>dane-county.json</code> but you can put <code>output.json</code> or <code>awesomeness.json</code> or whatever you want). Finally, we put the input file, <code>dane-county.shp</code>.</p> <p>You should see in your directory that you have a file called <code>dane-county.json</code> and you have converted your shapefile to a format better suited for the web!</p> <p>Now we’re going to practice using GDAL to clip a Landsat satellite image to the bounds of a shapefile. First let’s open up that beautiful TIF file to see what it looks like (you already know how to simply double-click on the file to open it, but that’s cheating! Use the command line!). The command we’re going to use is <code>open</code>, which opens the file using the default program. Using the commands <code>cd</code> to change directories and <code>pwd</code> to check which directory you’re currently in, navigate to the directory where your files are located. Then type in <code>open</code> and the name of your file (in this case, wisconsin.tif). For Windows users, navigate to your directory and then <a href="">just type in the file name</a>.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-05-open.gif" alt="open file" /> </figure> <p><code>open Users/robin/Desktop/wisconsin.tif</code></p> <p>(Pro-tip: You can also drag-and-drop the file from the directory onto the command line to get the full path name.)</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-06-drag-n-drop.gif" alt="drag and drop file" /> </figure> <p>Now we know what the image looks like. The GeoJSON that you just converted is a polygon of Dane County, Wisconsin, which takes up part of the area of the satellite image (you can always open it up in the GIS program of your choice to take a look). To crop the image to the GeoJSON, we’re going to use this command (I recommend typing the command in yourself, don’t copy/paste it):</p> <p><code>gdalwarp -cutline dane-county.json -crop_to_cutline wisconsin.tif dane-county.tif</code></p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-07-crop-to-cutline.gif" alt="gdalwarp crop to cutline" /> </figure> <p>Let’s break this down to make sure we understand what’s going on.</p> <p><code>gdalwarp</code> is a powerful <a href="">GDAL command</a> that allows you to reproject, crop, and warp georeferenced raster files. The documentation on <code>-cutline</code> says:</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-08-cutline.png" alt="gdal cutline" /> </figure> <p>That doesn’t really make much sense. That’s why I rely on other people for help because the formal documentation is often unclear. I got the above command from Derek Watkins’ <a href="">GDAL cheat sheet</a>.</p> <p>I know from looking at the command that <code>-cutline</code> is followed by the vector file, so my guess is that <code>-cutline</code> is indicating the bounding area that you want to crop the raster to. Then comes <code>-crop_to_cutline</code>, followed by the original TIF file and then the name that we want to give to the output TIF file. The output will save in your present working directory unless you specify a path to a different directory.</p> <p>We know how to open a file now, so let’s open up the output and see what it looks like.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/geo-command-line-09-open-dane-county.gif" alt="open output file" /> </figure> <h3 id="go-forth-and-command">Go forth and command</h3> <p>You won’t learn the command line overnight, and this post isn’t going to get you all the way there, but it’s a start. If you want to keep exploring, check out the resources below. I downloaded the Landsat image used here using <a href="">landsat-util</a>, a command-line utility that lets you search for, download, and process Landsat imagery.</p> <p>Part of what’s intimidating about the command line is that many of the online resources out there are unclear or aimed at people who already have a deeper understanding of it. So don’t be afraid to ask for help - check out <a href="">Stack Overflow</a> or join the <a href="">Spatial Community slack group</a>. Feel free to reach out to <a href="">me</a> or <a href="">Development Seed</a> on Twitter, too.</p> <p>Most importantly, when you come across a situation in your normal workflow when you think to yourself, “If I knew the command line, this would be easier/faster,” force yourself to do it. Push through your fear.</p> <h3 id="resources">Resources</h3> <h5 id="learn-more">Learn more:</h5> <ul> <li><a href="">The Command Line Crash Course</a>: Zed Shaw’s course that served as inspiration for this tutorial.</li> <li><a href="">Learn the Command Line</a>: Codecademy just came out with a new course for learning the command line.</li> <li><a href="">GDAL Quickstart</a>: Getting to know GDAL.</li> <li><a href="">GDAL Cheat Sheet</a>: Derek Watkins’ cheat sheet of common ogr2ogr and GDAL commands.</li> <li><a href="">Terminal Cheat Sheet</a>: Cheat sheet for basic Mac Terminal commands</li> </ul> <h5 id="other-geo-command-line-tools">Other geo command line tools:</h5> <ul> <li><a href="">Mapshaper</a>: Used for editing vector file formats, including tasks like editing attribute data, clipping, erasing, dissolving, filtering. Especially useful for simplifying shapes.</li> <li><a href="">Landsat-util</a>: Allows you to search for, download, and process Landsat imagery. Created and maintained by Development Seed.</li> <li><a href="">rio</a>: CLI for rasterio, which reads and writes geospatial raster files. Uses GDAL, but “is designed to make working with geospatial raster data more productive and more fun.”</li> <li><a href="">Topojson</a>: Create and manipulate TopoJSON files, which are like GeoJSON but preserve topology.</li> <li><a href="">geojsonio-cli</a>: Shoot GeoJSON files from the command line to for visualization and editing.</li> <li><a href="">Fiona</a>: A CLI for Python programmers who want to use OGR (see also the <a href="">cheat sheet</a>).</li> </ul> 2015-08-27T14:00:00+00:00 Development Seed Open aerial imagery for building a better road map <p class="dropcap">OpenAerialMap indexes over a 1,000 scenes of satellite and drone imagery. We recently showed how OAM’s <a href="">post-disaster imagery</a> in the Philippines could be used for disaster response. The same imagery is being used to make investment decision on road projects.</p> <p>The Philippines has over 200,000 kilometers of road. The national government has only mapped about 15% of the road network. There is no good map of local roads (particularly municipal and barangay roads). Philippines government agencies and international donors spend millions of dollars each year on rural road improvements. A comprehensive map of the road network would lead to better funding decisions and greater ability to monitor progress.</p> <p><a href=",124.93652343749999,8">Imagery now available on OAM</a> from <a href="">Skyeye Inc.</a> and the <a href="">OpenRoads project</a> is being used toward creating a comprehensive map of the country’s road network.</p> <figure class="bleed-full"> <iframe width="100%" height="600px" frameborder="0" src=",zoompan.html?access_token=pk.eyJ1Ijoib3BlbnJvYWRzIiwiYSI6InJ0aUQ2N3MifQ.R3hdFqriZr6kEUr-j_FYpg#18/10.68063/124.81338"></iframe> <figcaption>Active bridge construction near Baybay, Philippines</figcaption> </figure> <p>UAV imagery can complement inspection by engineers to quickly evaluate the road network. Crucial attributes like surface type, approximate surface quality and the width of the road can be gathered from this imagery. Where flying in engineers for regular field inspections is costly or time consuming, UAV and satellite imagery can be a cost effective way to monitor the progress of a road improvement project or to track the state of the road network between inspections.</p> <figure class="align-center"> <a href=",124.81417179107665,15/10.790112856939722,124.9365234375"><img src="/assets/graphics/content/prose/baybay-new-road.png" /></a> <figcaption>A newly constructed road in Baybay, Philippines. Imagery provided to OAM by SkyEye Inc.</figcaption> </figure> <p>Thanks to <a href="">Celina Agaton</a>, Skyeye, and the World Bank, for coordinating to open up this imagery. Check out the imagery by going to <a href="">the OAM image browser</a> and click ‘Browse latest imagery’.</p> 2015-08-17T14:00:00+00:00 Development Seed Post-disaster mapping with aerial imagery <p class="dropcap">OpenAerialMap is growing. UAV operators around the world are making their imagery open and available. The latest imagery available in OAM is post-disaster imagery after Typhoons Haiyan and Ruby. Typhoon Haiyan, also known as Yolanda, struck the Philippines in <a href="">November 2013 killing more than 6,000 and left 4 million displaced</a> from their homes. The Philippines was still recovering from Yolanda when Typhoon Ruby struck a year later. Disaster response organizations in the Philippines used <a href="">imagery after typhoons to assess building damage</a>. The new imagery was collected as a part of disaster response activities by Philippines UAV companies <a href="">Skyeye Inc.</a> and <a href="">COREPHIL Data Services</a>.</p> <p>Aerial imagery after a disaster is a critical response tool. Imagery after these Typhoons and after the recent earthquake in Nepal was <a href="">immediately helpful for creating maps to aid response efforts</a>. Aerial imagery is used to measure the extent of the devastation, the locations of displaced people, and what existing infrastructure exists to support the response.</p> <p>Head over to <a href="">OpenAerialMap and hit “Browse latest imagery”</a> to check out the new imagery. Thanks to <a href="">Maning Sambale</a> for helping to import the imagery into OpenAerialMap. Also check out OpenAerialMap’s slick new imagery filtering and sorting features.</p> <figure class="align-center"> <a href=",124.93635177612303,12/11.307678544927487,124.9365234375"><img src="/assets/graphics/content/oam-typhoon-yolanda-cover.jpg" /></a> <figcaption>Imagery over Tacloban after Typhoon Yolanda, December 2013</figcaption> </figure> <figure class="align-center"> <a href=",125.46369552612305,12/12.082264792573714,125.4638671875"><img src="/assets/graphics/content/oam-typhoon-ruby-cover.jpg" /></a> <figcaption>Imagery over Oras after Typhoon Ruby, December 2014</figcaption> </figure> 2015-08-12T10:00:00+00:00 Development Seed Welcome Nick Bumbarger <figure class="align-center"> <img src="/assets/graphics/content/nick-welcome-card.jpg" /> </figure> <p class="dropcap"><a href="">Nick Bumbarger</a> is joining the Development Seed team! Nick will expand our satellite imagery and remote sensing work. He will improve tools like <a href="">landsat-util</a> and create new analytical tools that make it easier to process satellite data and draw meaningful insights into global development challenges. We are excited to have Nick helping us to build the future of open geographic science for global development.</p> <p>As a scientist, Nick uses complex geospatial data to solve complex questions. Nick comes to Development Seed from the aerospace and defense industries where he deployed large scale imagery and mapping projects. Recently, Nick is experimenting with <a href="">Lego Mindstroms</a> bricks to build a personal robot. Nick has a Bachelor’s degree in Geography and a Master’s degree in Geographic Information Science from Clark University. Give him shout on <a href="">github</a> and <a href="">twitter</a>.</p> 2015-08-10T10:00:00+00:00 Development Seed Introducing Collecticons <p class="dropcap"><a href="">Collecticons</a> is an ever growing compendium of crisp icons for responsive web apps. Collecticons includes <a href="">a processing script</a> to generate an icon font for use with your CSS or SASS.</p> <h2 id="icons-matter">Icons matter</h2> <p>We use a lot of iconography. Clean icons convey meaning simply and quickly. Clean icons help us to build an <a href="">intuitive interface to search satellite imagery</a> and to tell a clear story that <a href="">some oil is dirtier than others</a>.</p> <p>There are several ways to include icons in a project. We find the icon font approach most suitable to our process, simply because icon fonts are great:</p> <ul> <li>They are ridiculously lightweight;</li> <li>Only one single font file is needed instead of multiple images, requiring just one HTTP request;</li> <li>The icons can be scaled to any size, and easily styled using CSS;</li> <li>By using <a href="">PUA code points</a> for the glyphs, you prevent screen readers from reading these characters and thus addressing accessibility issues.</li> </ul> <h2 id="a-better-way-to-manage-icons">A better way to manage icons</h2> <p>We built <a href="">Collecticons</a> to improve our own development process. Previously we used <a href="">IcoMoon</a> to create a webfont but any small change to an icon required that we manually generate the font all over again. Over a project with many careful iterations to the icon set, this time adds up. So we built our own tools for icon management hat we could plug right into our development process.</p> <p>The <a href="">Collecticons Library</a> allows us to easily manage and reuse icons. Some of the icons were built on top of libraries. Most we made in house. With the <a href="">Collecticons Processor</a>, we generate the font and sass files automatically when we do our site build.</p> <p>Collecticons saves us time, makes our development process more efficient, and lets us borrow from tested icon conventions. We hope it does the same for you. We look forward to seeing you involved in the project, which is licensed under <a href="">the open source MIT License</a>.</p> 2015-07-27T00:00:00+00:00 Development Seed Field testing open source for road planning <p class="dropcap">We are getting out of Manila and are field-testing some of our tools and assumptions in Quezon this week, a province about 3-4 hours travel. This is part of our engagement with the World Bank to help the Philippine government <a href="">use open source tools to manage road data across government</a>. We are meeting government officials across a number of road-building national agencies and local government road managers, to design tools that can support them to make better investment decisions.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/warm_welcome_lucban.jpg" alt="A warm welcome in Lucban" /> <figcaption>A warm welcome by the municipality of Lucban. Photo by Kai Kaiser.</figcaption> </figure> <p>We are using <a href="">Mapillary</a> and <a href="">Routeshoot</a>, two mobile apps that record images and GPS tracks as you walk or drive. Routeshoot also records elevation data, which is useful for planning road improvements. It also captures the speed the surveyor was moving, from which you can infer road quality. Both apps work on simple Android devices and allow local engineers to do first-pass assessments on the cheap. They also work offline by design.</p> <p>Stay tuned for more updates as we continue building out OSM as a platform in the Philippines.</p> 2015-07-19T00:00:00+00:00 Development Seed Access the Planet Labs archive <p class="dropcap">We have access to the full Planet Labs archive! We can mine the Planet Labs imagery archive to create analytical and derivative products in order to better understand and monitor our changing planet. Scores of Planet Lab satellites circle the planet returning a regular stream of data.</p> <p>And the imagery just got better! Yesterday Planet Labs <a href="">acquired our friends at Blackbridge</a>, meaning they now have the entire Rapid Eye constellation of 5-meter resolution satellites. Planet Labs will add six years of Rapid Eye imagery, which includes data specialized for agriculture, forestry, and climate change applications.</p> <p><a href="&#109;&#097;&#105;&#108;&#116;&#111;:&#105;&#110;&#102;&#111;&#064;&#100;&#101;&#118;&#101;&#108;&#111;&#112;&#109;&#101;&#110;&#116;&#115;&#101;&#101;&#100;&#046;&#111;&#114;&#103;">Email us</a> to get access to <a href="/set/understanding-our-changing-planet/">imagery and analysis for global development projects</a>.</p> <figure class="align-center"> <img src="/assets/graphics/content/set/doves-in-space.jpg" width="1022" height="374" alt="Doves in Space" /> <figcaption>Doves emerging from International Space Station. <small>Image by <a href="" title="View original image">Steve Jurvetson</a></small>.</figcaption> </figure> 2015-07-16T00:00:00+00:00 Development Seed Open Drone Imagery <p class="dropcap">Since the <a href="">beta launch of OpenAerialMap</a> we’ve been approached by satellite companies and drone operators looking to share images with resolutions ranging <a href="">from 30 meters</a> down to 3 cm. Some of the drone imagery we’ve seen is particularly impressive. The drone imagery below is of Dar Es Salaam, Tanzania and comes from the Tanzanian Commission of Science and Technology and the Tanzania Open Data Initiative collaborating with Drone Adventures exploring how to use drones in Tanzania. At the moment <a href="">Ramani Huria</a> community mapping projects are using the imagery and a Digital Surface Model to explore how communities in Dar es Salaam can become more resilient to periodic urban flooding.</p> <figure class="bleed-full"> <iframe width="100%" height="500px" frameborder="0" src=",share.html?access_token=pk.eyJ1IjoiZHJvbmVhZHYiLCJhIjoiYmU0ZXQtcyJ9.8Fh95YZQ_WdYEDlgtmH95A"></iframe> </figure> <p>If you are a drone pilot with imagery to share, open an issue on the <a href="">OpenAerialMap Github page</a>.</p> 2015-07-06T00:00:00+00:00 Development Seed UX/UI Designer Wanted <p class="dropcap">Design is an important part of <a href="/projects">every project that we do</a>. We take complex open data and open source technology and make it useful and usable to people solving some of the worlds hardest problems. To do that we consider every design trick and push the limits of what is possible.</p> <p>We are looking for opinionated designers who want to build amazing products side-by-side with other committed developers, designers, and strategists in Washington DC and Lisbon. In addition to design skills, you should have some tech chops or an interest in developing them.</p> <p>Interested? <a href="/about/jobs/#uxui-designer">Check out the job description</a>.</p> 2015-06-26T00:00:00+00:00 Development Seed How to Organize a Mapathon <p class="dropcap">Today I gave <a href="">a lightning talk</a> at State of the Map about how to organize a mapathon. A mapathon, also known as a mapping party, is when a bunch of people get together to edit <a href="">OpenStreetMap</a>, the editable map of the world. Here’s a guide for organizing a mapathon, based on my experience organizing mapathons in Bogota, Colombia, and Madison, Wisconsin, along with great advice from people who have organized OpenStreetMap mapathons all over the world (thank you, you lovely people!).</p> <p>First things first: there is no one right way to run a mapathon. I will outline things to consider, but you don’t need to do all of these. Do what works for you, and <a href="">lean on the wonderful OSM community</a> for support when you have questions.</p> <h2 id="preparation">Preparation</h2> <h4 id="find-a-co-organizer-or-two">Find a co-organizer (or two)</h4> <p>A lot can go into organizing a mapathon, including the preparation beforehand and the actual event. It’s helpful to have someone to split up the duties with and who can help attendees on the day of the event.</p> <h4 id="determine-your-priority">Determine your priority</h4> <p>Are you organizing a mapathon to add a lot of data to the map, or are you trying to introduce new people to this awesome project? These things aren’t mutually exclusive, but it’s a good idea to keep your main goal in mind as you are planning and running the mapathon.</p> <h4 id="find-a-location">Find a location</h4> <p>Common places to hold mapathons include local schools, universities, businesses, libraries, restaurants, cafes, or parks. You’ll want to <a href="">check on the internet connection</a> at the location to try to avoid connectivity issues. Some mapathons have had issues when lots of people try to upload or download data from OpenStreetMap, so you can also check if different computers can connect through different IP addresses.</p> <p>You might be able to find a local business or organization to sponsor the event to provide food, drinks, and supplies.</p> <h4 id="decide-on-the-format">Decide on the format</h4> <p>Do you want to do surveying of a local area? Or will you be doing “armchair mapping,” contributing data to the map using aerial imagery or other existing data? If you are going to do outdoors surveying, I recommend meeting at a central indoors location to introduce everyone to OpenStreetMap, splitting up into groups to go off and map nearby, then coming back together at the end to add the collected data to the map. Think ahead of time of possible routes that people can take or certain things that people can map, like addresses, mailboxes, or restaurants. During the event, people are free to map whatever they want, but it’s good to have some ideas ready.</p> <h4 id="pick-your-tools">Pick your tools</h4> <p>There are tons of great tools out there to help you contribute to OpenStreetMap. You might want to stick with just one or two to teach at the mapathon so people don’t get overwhelmed by too many options. Or you can provide information about all of them and let people decide what works for them.</p> <ul> <li>For local mapping: <ul> <li><a href="">Field Papers</a></li> <li>Phone apps (OSMTracker, OsmPad, GoMap!!, Pushpin, to name a few)</li> <li><a href="">Mapillary</a></li> <li>Cell phone / camera</li> <li>A notepad</li> </ul> </li> <li>Remote mapping: <ul> <li><a href="">HOT Tasking Manager</a></li> </ul> </li> <li>Either: <ul> <li><a href="">Battle Grid</a></li> <li><a href="">MapRoulette</a></li> </ul> </li> </ul> <h4 id="pick-an-editor">Pick an editor</h4> <p>JOSM and iD are the main OpenStreetMap editors, with <a href="">JOSM</a> being the primary desktop application and <a href="">iD</a> as the main web-based editor. If you are expecting a bunch of new mappers, you should focus on iD, which has a simpler interface and doesn’t require a download. If you are an expert JOSM user, make sure to familiarize yourself with iD ahead of time so that you can help answer questions that come up. If you are planning to teach JOSM, make sure it is downloaded on available computers or advise people to download it on their own laptops ahead of time.</p> <hr /> <h2 id="outreach">Outreach</h2> <h4 id="collaborate">Collaborate</h4> <p>Reach out to other local organizations to see if they want to work together and help with outreach. This can include a local <a href="">OSM Meetup group</a>, <a href="">Maptime chapter</a>, school or university, or the general open-source community. Leverage other organizations’ email lists and social media presence. You can also organize a mapathon in coordination with a <a href="">nationwide U.S. mapathon</a> to get extra coverage.</p> <h4 id="get-out-the-word">Get out the word</h4> <p>Twitter is great, but there are a bunch of other ways to get out the word, too. You can put up flyers around the neighborhood, send out messages to local neighborhood or city lists, or post on <a href="">Nextdoor</a>.</p> <h4 id="asking-for-rsvps">Asking for RSVPs</h4> <p>It’s a good idea to ask people to RSVP for the event on Meetup, Eventbrite, or Facebook. People can still show up at the door, but it will help with planning if you have an idea whether you’ll have 5 or 55 people.</p> <hr /> <h2 id="the-day-of">The day of</h2> <h4 id="intro-to-osm-and-the-tools">Intro to OSM and the tools</h4> <p>Give a brief introduction to OSM. Some people prefer to do this as a 10-30 minute presentation at the beginning of the event, while others prefer giving 3-5 minute presentations throughout the event on different facets of OSM. You can talk about what OpenStreetMap is, different ways to contribute to the project, give examples of ways that it is used, and why open data is important.</p> <p>Be sure to introduce the tool(s) that you’re going to use during the mapathon. You can mention additional tools that people can use, but focus on just one or two.</p> <h4 id="welcome-newbies">Welcome newbies</h4> <p>Hopefully you’ll have people at your event who have never even heard of OpenStreetMap. That’s great! If a new mapper has a positive experience at a mapathon, that will make them more likely to support it in some way. There are lots of ways to do this:</p> <ul> <li>Emphasize that you and other experienced mappers are available to help with any questions.</li> <li>Match up experienced mappers with new mappers so that they have a dedicated person who they can go to with questions.</li> <li>Print “Get Started” guides with basic information about OSM and how to edit.</li> <li>Make sure people know that they don’t need to stay for the whole event.</li> <li>Point to additional resources that people can turn to after the event is over.</li> </ul> <h4 id="get-mapping">Get mapping</h4> <p>Go out and survey or start making edits. If you are doing outdoors mapping, make sure someone stays behind at the meeting location in case people show up late or one of the participants has mobility issues.</p> <h4 id="most-importantly-have-fun">Most importantly: have fun</h4> <p>Be enthusiastic and have a great time. You are the best ambassador for OpenStreetMap and getting people excited about contributing in whatever way they can.</p> <p>If you were able to get a sponsor, you can do something social after you’re done with the serious stuff.</p> <h3 id="resources">Resources</h3> <p>If you need more help, here are a few additional resources:</p> <ul> <li><a href="">LearnOSM</a></li> <li><a href="">TeachOSM</a></li> <li><a href="">Mapgive</a></li> <li><a href="">Humanitarian OpenStreetMapTeam</a></li> <li><a href="">Maptime</a></li> </ul> <p><em>Card photo by <a href="">Harry Wood</a></em></p> 2015-06-07T16:30:00+00:00 Development Seed OpenAerialMap Interface Design <p class="dropcap">We recently built several highly <a href="">usable</a> <a href="">imagery</a> <a href="">browsers</a>. Each allow users to get right to the data they want by browsing a map. This interface works well with Landsat imagery, which has a consistent coverage area. For large, complex imagery datasets like <a href="">OpenAerialMap</a> we created a new type of grid interaction.</p> <h2 id="designing-for-complex-map-data">Designing for complex map data</h2> <p>Last week we launched the <a href="">beta version of OpenAerialMap</a> a tool for finding open satellite and drone imagery. We knew that even in beta, <a href="">usability is going to be critical</a> to the adoption and success of OpenAerialMap.</p> <p>OpenAerialMap is a site that has to show a lot of data. It features imagery from different providers, with different aspect ratios, and captured on different dates. Presenting such data in a meaningful and usable way was challenging, especially considering that individual areas can contain multiple imagery.</p> <h2 id="enter-the-grid">Enter the grid</h2> <figure class="align-center"> <img src="/assets/graphics/content/prose/battleship-board-game.jpg" width="3456" height="2304" alt="The grid in action" /> <figcaption>Battleship board game. <small>Image by <a href="" title="View original image">John Morgan</a></small>.</figcaption> </figure> <p>Drawing inspiration from <a href="">hexgrid experiments using Turf</a> and good old <a href="">battleship game</a>, we created the grid for showing the density of imagery at any area. The grid breaks the world up into units that are easy to interact with.</p> <p>We used an “always square” grid disregarding the map projection. The result is a beautiful, clean, easy-to-use grid where all interactions with the map are consistent and as visual appealing as possible.</p> <p>We color grade each grid cell according to the number of images intersecting them. This way we avoid bloating the map with useless information and provide a sense of density. The grid works as a visual guide where at a quick glance the user can easily grasp where and how much imagery is available.</p> <figure class="bleed-full"> <img src="/assets/graphics/content/prose/oam-browser-interaction.gif" width="2545" height="764" alt="The grid in action" /> <figcaption>The grid in action.</figcaption> </figure> <p>Selecting a cell reveals a panel with a list of imagery. From there, the imagery can be previewed on top of the map, downloaded or, when available, used as the baselayer of another map.</p> <p>Designing usable map interactions requires real thoughtfulness. But it makes all the difference between a tool that <a href="">people love</a> and one that is forgotten.</p> 2015-06-05T10:00:00+00:00 Development Seed Off to State of the Map <p class="dropcap">We will be at <a href="">State of the Map US</a> this weekend at the United Nations. It promises to be a great weekend, packed with interesting people, talks and other events. If you want to catch up come show off your Lego skills at our table, or join us at any of our sessions:</p> <ul> <li><strong><a href="">OpenAerialMap - Birds of a Feather</a></strong> <br /> <strong>Saturday, 4pm (tentative)</strong> <br /> OpenAerialMap is a platform to access openly-licensed satellite and drone imagery. Discuss the architecture and development roadmap for OpenAerialMap and the Open Imagery Network. Find <a href="">Nate</a> for more info.</li> <li><strong><a href="">OSM as a platform - Birds of a Feather</a></strong> <br /> <strong>Sunday, 1pm (tentative)</strong> <br /> What if we could use the software that powers OSM in other collaborative mapping projects? Its harder than you might think. We’ll trade tips on running an OSM infrastructure and identify a roadmap to make it easier. Hit me up <a href="">on Twitter</a> if you want to know more.</li> <li><strong><a href="">OSM as a platform</a></strong> <br /> <strong>Sunday, 3pm - room CR2</strong> <br /> I will talk about <a href="">OpenRoads</a>, a platform for The Philippine government to manage their road network built entirely on OSM software.</li> <li><strong><a href="">How to organize a mapathon</a></strong> <br /> <strong>Sunday, 3:30pm - room CR3</strong> <br /> Catch <a href="">Robin’s</a> lightning talk on organizing a mapathon.</li> <li><strong><a href="">OSM Metadata - Birds of a Feather</a></strong> <br /> <strong>To be defined</strong> <br /> Interested in using OSM’s changeset meta data to learn about the OSM community? Get in touch with <a href="">Marc</a> and keep an eye on the BOF whiteboard.</li> </ul> 2015-06-04T00:00:00+00:00 Development Seed Opening up Air Quality Data <p class="dropcap">Air pollution is a leading cause of death across the globe, and contributes to stroke, heart disease, lung cancer, and other respiratory illness. While the vast majority of deaths are in low and middle income countries, air quality continues to worsen in cities across the world.</p> <p>So we made an open data pipeline and API for citizens to better understand the quality of the air they breathe.</p> <figure class="align-center"> <a href=""><img src="" /></a> <figcaption>Pollution seen in Ulaanbaatar, Mongolia.</figcaption> </figure> <p>Using accurate, low-cost sensors, people can measure the concentration of harmful particulate matter in their homes or places of work. Feeding into our data pipeline, these sensors create a central repository of data about air quality that anyone can access through the API. With several sensors scattered across an area, citizens create meaningful data that they can use to advocate for better policy, zoning laws, and regulation.</p> <p>We’ve been working with a group of hardware engineers, infrastructure builders, and journalists to develop an air quality monitoring system powered by an open API and low-cost sensors – some great folks at <a href="">InfoAmazonia</a>, <a href="">Feedback Labs</a>, <a href="">FrontlineSMS</a>, <a href="">Groundtruth Initiative</a>, <a href="">EJN</a> and <a href="">Internews</a>.</p> <p>Earth Journalism Network (EJN) and InfoAmazonia designed, manufactured, and deployed <a href="">Dustduinos</a>, Arduino-based sensors that detect particulate matter at 2.5µm and 10µm. The Dustduino uses an open spec that was <a href="">optimized for low power consumption and SMS communication and wifi</a>. To ensure connectivity in areas without wifi, <a href="">FrontlineSMS</a>, adapted their SMS services to support the Dustduino.</p> <p>Our role was to turn the raw input data into actionable information. We built a data pipeline to make this data publicly available through a <a href="">flexible API</a> as well as for download. The API allows anyone to build apps on top this information, or integrate it with other tools. Downloads will allow researchers and advocacy organizations to work with the data in tabular format.</p> <p>Open and available air quality data can empower citizens in vulnerable areas to have more say in the policies that affect their local air quality as well as providing researchers with valuable insight into potentially understudied areas. This kind of change happens from the ground up. We will continue to support this change by building open-source data pipelines, better sensors, and a robust community for open air quality data.</p> 2015-06-03T10:00:00+00:00 Development Seed Welcome Robin Tolochko! <p><img src="" alt="Robin with bike" /></p> <p><a href="">Robin Tolochko</a> loves maps. Her favorite map is a 1868 map of South America. We know that because it’s on her resume.</p> <p>Robin shares her map love. She directed a mapping lab in Bogota and teaches others to make maps at Maptime Madison. Come see her <a href="">lightning talk on Mapathons</a> at State of the Map US.</p> <p>At Development Seed, Robin will make beautiful maps that are intuitive and purposeful. Robin brings curiosity and a detail-oriented eye to all the <a href="">work she does</a>. She’s committed to advancing women’s rights and renewable energy. She also owns a <a href="">small business</a> that sells handmade leather goods from Colombia.</p> <p>Hit up Robin on <a href="">twitter</a> or <a href="">GitHub</a> to chat map projections, hiking, or ultimate frisbee.</p> 2015-05-29T10:00:00+00:00 Development Seed Introducing OpenAerialMap <p class="dropcap">Today we’re releasing a beta version of <a href="">OpenAerialMap</a>. OpenAerialMap makes it easy to share and find open satellite and drone imagery. This is critical to the work of the disaster response community. We are launching this tool in close partnership with the <a href="">Humanitarian OpenStreetMap Team (HOT)</a>.</p> <figure class="align-center"> <img src="" /> </figure> <p>OpenAerialMap is a set of tools for searching, sharing, and using open satellite and drone imagery. This initial release includes the core infrastructure to catalog petabytes of open imagery. It also includes an extremely usable API and an elegant web interface to submit, search and download available imagery.</p> <figure class="align-center"> <img src="" /> <figcaption>Search for available imagery.</figcaption> </figure> <figure class="align-center"> <img src="" /> <figcaption>Select scenes by grid.</figcaption> </figure> <figure class="align-center"> <img src="" /> <figcaption>Preview imagery and get metadata information.</figcaption> </figure> <h3 id="rebooting-a-great-concept">Rebooting a great concept</h3> <p>The OpenAerialMap concept has bounced between several attempts over half a decade. Previous attempts failed to take off. HOT <a href="">reinitiated the concept</a> this year with funding from the <a href="">Humanitarian Innovation Fund</a>. We worked with the HOT and others in the open imagery community to reimagine an approach to OpenAerialMap that we expect to be much more successful.</p> <ul> <li>We focused on simple, usable toolset that meets the clear needs of the humanitarian response community. The underlying architecture is flexible enough to be immediately useful to research, resource management, urban planning, and other communities. However, we decided to first build a frictionless interface for the clear needs of the disaster response community.</li> <li>We are extremely focused on community. From day one, we involved other organizations and developers like <a href="">Azevea</a>, <a href="">Planet Labs</a>, <a href="">Cadasta</a>, <a href="">OpenDroneMap</a> and HOT’s own developers.</li> <li>With these and other groups, we reimagined OpenAerialMap along a network model. Rather than try to house all the open imagery out there, OpenAerialMap is a node and index for a larger network or open imagery - the <a href="">Open Imagery Network</a>.</li> </ul> <h3 id="a-network-approach-to-open-imagery">A network approach to open imagery</h3> <p><a href="">Open Imagery Network (OIN)</a> is a simple framework and license for placing imagery into an open source license. Participants in the OIN adopt a common metadata scheme to describe the imagery they are making available, and standardized ways to broadcast and access that data. This allows us to build tools that search across all open imagery data without requiring one entity to host all of it. We’re working with HOT, Planet Labs, Cadasta, Azavea, OpenDroneMap, and others to develop OIN and to build OpenAerialMap as the first node in that network.</p> <h3 id="your-feedback">Your feedback</h3> <p>A vibrant community will be critical to the success of OpenAerialMap. Check out the beta version and send us feedback on <a href="">Twitter</a>. Or open an issue or make a contribution <a href="">on Github</a>. Currently OpenAerialMap has imagery from the Nepal earthquake response, high resolution satellite imagery of Finland, <a href=";subject=prog&amp;topic=nai">agriculture imagery</a> over Nebraska, and drone imagery from Vanuatu. We’ll work to add functionality and features toward an initial release candidate later this summer.</p> 2015-05-27T12:00:00+00:00 Development Seed Open Geospatial Happy Hour <p class="dropcap">The <a href="">Geospatial World Forum</a> is well under way in Lisbon and we’re impressed with the diversity of people and talks at the event.</p> <p>On Thursday night, we are organizing an Open Geospatial Happy Hour at Café Fábulas with our friends from <a href="">Planet Labs</a>. This Happy Hour is a great opportunity to meet people and talk about open data, mapping and satellites in a more informal setting.</p> <p>You don’t have to be attending the conference to join the happy hour, just let us know if you’re coming. You can <a href="">RSVP here</a>.</p> <p>If you are participating in the GFW, make sure to check out our <a href="">workshop and talk</a> in the exhibition hall on Thursday morning.</p> 2015-05-27T00:00:00+00:00 Development Seed Oil-Climate Index Launches <p class="dropcap">Nearly 100 million barrels of oil flow through the global oil supply chain every day. But not all oils are created equal. When you consider the full oil processing lifecycle, some types of oil are responsible for nearly twice as much greenhouse gas as others. This is important information for oil procurement and energy policy. Smarter oil selection can lead to significant reduction in greenhouse gases without even touching overall oil consumption. We worked with the Carnegie Endowment to launch the <a href="">Oil Climate Index website</a> to help consumers and policymakers make smarter decisions on oil.</p> <h3 id="extracting-oil-data">Extracting Oil Data</h3> <p>Oil emission data has traditionally been extremely hard to find. The source data for most oil fields is often secret and some of the models needed to calculate total greenhouse gas emissions have been proprietary. The Carnegie Endowment made a major contribution to understanding the climate impacts of fossil fuels with the <a href="">Oil-Climate Index</a>. The Oil Climate Index is the first study using entirely open-source models for evaluating greenhouse gas emission. These models were developed in a collaboration by Carnegie Endowment’s Energy and Climate Program, Stanford University, and the University of Calgary. In addition, the Oil Climate Index also collected model input data for 30 popular and emerging oils.</p> <figure class="align-center"> <img src="" alt="Scatter plot" width="759" height="587" /> <figcaption>Explore relationships between oil properties</figcaption> </figure> <p>The data is extremely complex and nuanced. Depending on what you want you want to do with an oil, making diesel vs jet fuel, different oils may be better. An improvement in an extraction or refinement process may have a significant climate benefit for one oil but not for another. To make this data useful to scientists, investors, policy makers, and interested citizens we built a <a href="">flexible data exploration tool</a>. The tool makes reasonable assumptions to allow immediate comparison, but also allows users to explore how specific factors change the overall climate impact of each oil. Most importantly, all of the data and the modeling methodologies are open and available for download.</p> <h3 id="designing-for-complexity">Designing for complexity</h3> <p>We limited graphing options to the most meaningful properties to provide flexibility without unnecessary complexity. To visualize the Oil-Climate Index, we weren’t just showing static data, we were visualizing results of complex models. Processing all the data in the browser is impractical. Instead, we picked several model input parameters that are most meaningful and calculated their data up front. We built processing tools behind the site to recalculate this data as Carnegie extracts data on new oils.</p> <figure class="align-center"> <img src="" alt="Model parameters" width="370" height="172" /> <figcaption>Model parameters allow the user to customize the model and data output</figcaption> </figure> <p>For more information on the Oil-Climate Index hit us <a href="">on Twitter</a>, ask <a href="">@DxGordon</a>, or check out these resources:</p> <ul> <li><a href=""></a></li> <li><a href=""></a></li> <li><a href=""></a></li> </ul> 2015-05-26T18:00:00+00:00 Development Seed We're sending a delegation to State of the Map US <p class="dropcap"><a href="">State of the Map US</a> is around the corner. We’re sending a robust delegation to the United Nations for the biggest OpenStreetMap conference ever.</p> <p>This is a critical meeting for OSM and for open data. OpenStreetMap is becoming too big to fail. It is now critical infrastructure for everything from urban planning to disaster response. We’ll join 1000 open mappers from dozens of countries at a truly international venue to discuss new opportunities and responsibilities for OSM.</p> <p>We’ll be talking about:</p> <ul> <li><strong>OSM-as-a-platform</strong> - What if governments could roll OSM internally as an open source stack to manage road data between agencies? Will wider use of software behind OSM lead to better OSM data? Better OSM software? Check out <a href="">Olaf’s talk</a> on our experience <a href="">rolling out OSM-as-a-platform in the Philippines</a>. Or talk to <a href="">Olaf</a>, <a href="">Derek</a>, or <a href="">Anand</a>.</li> <li><strong>Organizing mapathons</strong> - <a href="">Robin</a> will deliver a <a href="">lightning talk on organizing mapathons</a>. Hit her up to talk about making mapathon magic.</li> <li><strong><a href="">OpenAerialMap</a></strong> - OpenAerialMap will be an open catalog of satellite and aerial imagery. Find <a href="">Nate</a>, <a href="">Alireza</a>, and <a href="">Marc</a> to talk about our contributions to OAM.</li> <li><strong>OSM Meta Data</strong> - Want to understand the OSM community? The answers are in the metadata. Find <a href="">Marc</a> and <a href="">Drew</a> to talk about our efforts to <a href="">open OSM changeset metadata</a> and <a href="">measure OSM growth</a></li> <li><strong>Humanitarian OpenStreetMap Team</strong> - <a href="">Nate</a> is always scheming about how to improve the HOT imagery pipeline. We all have a ton of HOT love and thoughts on OSM in disaster response.</li> </ul> <p>We hope to see you there. There are <a href="">still a few tickets left</a> if you want to join us in New York and geek out on open mapping.</p> 2015-05-26T00:00:00+00:00 Development Seed Geospatial World Forum kicks off <p class="dropcap">The <a href="">Geospatial World Forum</a> kicks off today in Lisbon. We will be talking with NGOs, government agencies, and researchers about how they can use open geospatial data and open source software. Our Lisbon team will host an open geospatial happy hour on Thursday.</p> <p>Open geospatial data is some of the most valuable data that government produce, generating billions of dollars in economic value and tremendous social good. Open geospatial data underlies everything from critical climate change research to many of the apps in your favorite app store.</p> <p><a href="">Libra</a> and <a href="">Landsat-util</a> are powerful open source tools for processing and analyzing Landsat data. On Thursday at 11:00 I’ll talk at the <a href="">Open Data track</a> about using these tools in combination with other open software like <a href="">QGIS</a> to work with open satellite data. At 9:00 Thursday <a href="">Alireza</a> and <a href="">Nate</a> will offer a hands on training on open tools for satellite data at the <a href="">the Open Source Imagery Tools workshop</a>.</p> <p>We’ll continue the conversation about open data, mapping and satellites at an <a href="">Open Geospatial Happy Hour</a> that we’re organizing with our friends from <a href="">Planet Labs</a> on Thursday evening. If you are interested in open geospatial data, please join us. You don’t need to be at the conference to attend.</p> <p>We look forward to meeting old and new geo friends while you are in Lisbon. If you want to meet me and our Lisbon team <a href="">hit me up on Twitter</a>.</p> 2015-05-25T00:00:00+00:00 Development Seed GitHub cards <p class="dropcap">We love <a href="">GitHub</a>. We work in the open. We have well over 100 public projects on GitHub and contribute to many others. We’ve helped dozens of NGOs, International Organizations, and Government Agencies to open source their code on GitHub.</p> <p>When we redesigned our website, we wanted to promote open source and to make it easy to find our code on GitHub. So we created GitHub cards.</p> <h3 id="whats-a-github-card">What’s a GitHub Card?</h3> <p>A GitHub card is a styled card that clearly calls out that this is an open source project, links to the project on GitHub, and shows statistics like forks and stargazers that show the depth of the community around that tool.</p> <figure class="align-center"> <img src="/assets/graphics/content/prose/github-card-example@2x.png" alt="GitHub card" width="370" height="172" /> <figcaption>A GitHub card example</figcaption> </figure> <p>GitHub cards are a visual way to promote open source and to make it easier for others to participate in our projects by requesting features or contributing code.</p> <h3 id="let-a-thousand-github-cards-bloom">Let a Thousand GitHub Cards Bloom</h3> <p>GitHub cards aren’t natively supported by GitHub. We built them into our site using a simple visual wrapper and a bit of code to pull live data from our public GitHub repos.</p> <p>We’d love to see more sites use GitHub cards. A simple visual flag for open projects across websites can reinforce how much of the web is built on open technologies. There are solid open source options for nearly any technology need. GitHub cards can make them more obvious.</p> <p>If you want to add GitHub cards to your own site, feel free to use our code. Its <a href="">open</a>.</p> <p data-height="268" data-theme-id="0" data-slug-hash="wBVOXe" data-default-tab="result" data-user="danielfdsilva" class="codepen">Check out the source code on <a href="">codepen</a>.</p> <script async="" src="//"></script> <p><br /></p> <p><em>Update: Within a few hours or posting, <a href="">@gmaclennan</a> from <a href="">Digital Democracy</a> built a tool to generate github cards. Awesome. Check it out:</em></p> <iframe width="320px" height="155px" frameborder="0" src=";title=Github Card&amp;link="></iframe> 2015-05-14T00:00:00+00:00 Development Seed The Connectivity Atlas <p class="dropcap">We recently launched the first iteration of the <a href="">Connectivity Altas</a>, a project to map infrastructures on a global scale. From roads and rivers to internet and electricity lines, these intricate and vast networks exist everywhere we are. Considering that infrastructure is a broad classification, mapping out these different systems provides a unique insight into how these networks are dispersed onto the globe. Since the Connectivity Atlas was our first significant dive into <a href="">Mapbox GL</a>, we wanted to share some notes about our experience.</p> <h4 id="the-challenge-of-complex-data">The challenge of complex data</h4> <p>I started to build the map using <a href="">Mapbox Studio</a> importing each layer into Studio as a separate data source and styling each layer using <a href="">Carto CSS</a>. This workflow worked fine for a while, but over time might have led to hundreds of sources and styles pushing the limit of our Mapbox account and our organizational skills. This approach would also lead to performance issues as our data sets increase.</p> <p><img src="" /></p> <h4 id="rendering-vector-data-in-the-browser">Rendering vector data in the browser</h4> <p>We considered alternative workflows and ultimately decided to use Mapbox GL for various reasons.</p> <ul> <li>We like vector tiles (super crisp)</li> <li>We like rendering styles in the browser (it makes for faster styling / instant results)</li> <li>Better organization - we used a jekyll collection to add styles into a master JSON</li> </ul> <p><img src="" /></p> <h4 id="what-we-learned">What we learned</h4> <p>Vector tile platforms like GL allow for much faster rendering of complex data by using the graphics processor in the same way that a video game does. The tools for working with Mapbox GL are still limited. For example, we had to write our own tooltip functionality to expose the meta information in each layer. But we know the Mapbox folks are cooking up some new tools for GL, and we look forward to using GL for other projects.</p> <h4 id="open-infrastructure-data">Open Infrastructure Data</h4> <p>All of the data on the Connectivity Atlas is open and available for reuse. This is a collaborative project, and you can participate by sharing and suggesting data. This way we can produce a map of our profound inter-connectedness as well as an oddly beautiful web of global infrastructure.</p> 2015-05-01T02:00:00+00:00 Development Seed Our new website - still Jekyll <p class="dropcap">We recently redesigned our website from the ground up. While the look and organization changed, the underlying technologies have not. We are still using Jekyll to power the site, just as <a href="">we’ve done since 2011</a>. Jekyll has come a long way and we’re taking full advantage of some of its new features.</p> <h3 id="collections">Collections</h3> <p>Collections are the <a href="">Jekyll</a> way to separate content types. They allow us to create new content types and to treat them differently. We are using collections for our team and project pages. Collection items have the attributes of “posts” in Jekyll with some subtle differences, like being ordered alphabetically in ascending order or not needing to have the date on the name. With collections we can achieve a greater content separation thus making the whole site more organized.</p> <h3 id="list-views">List views</h3> <p>Jekyll still has some challenges when it comes to paginate collections or categories. We solved this problem by rendering our list views (projects and blog posts) client side.</p> <p>Jekyll outputs a JSON file that we use to render the content with ejs templates. This allows us to have greater control over content output and prevents your browser from blowing up when trying to render our 650+ blog posts at once.</p> <h3 id="wrangling-the-json-api">Wrangling the JSON API</h3> <p>Jekyll’s JSON output can lead to a familiar comma problem. When Jekyll is generating a JSON file is common to see code like this:</p> <pre><code>{% raw %}[ {% for post in site.projects %} { "title": "{{ post.title }}" } {% unless forloop.last %},{% endunless %} {% endfor %} {% endraw %}] </code></pre> <p>This will work fine when every element of the array is being printed, but if you start introducing conditions the <code>forloop.last</code> can’t be used to control the comma anymore.</p> <p>For example we’re only showing posts that are not hidden. Image we’ve only three posts of which the third one is hidden. The resulting JSON would be:</p> <pre><code>[ { "title": "Project one" } , { "title": "Project two" } , ] </code></pre> <p>See the trailing comma? That little thing would break the JSON. It’s there because our last visible project is <code>Project two</code> but that is not the last element in the array so <code>forloop.last</code> doesn’t kick in and the comma is printed anyways.</p> <p>The workaround for this situation is easy. You just have to print the comma before the element except for the first time:</p> <pre><code>{% raw %}[ {% assign comma = false %} {% for post in site.projects %} {% unless post.hidden == true %} {% if comma %},{% endif %} {% assign comma = true %} { "title": "{{ post.title | escape_once }}" } {% endunless %} {% endfor %} {% endraw %}] </code></pre> 2015-04-20T00:00:00+00:00 Development Seed A Modern Imagery Processing Pipeline <p class="dropcap">Satellite data is a tremendously powerful resource for governments and development organizations. We built a <a href="">suite</a> <a href="">of</a> <a href="">tools</a> to make open Landat data more accessible and useable. These allow our development partners to process imagery and perform analysis quicker, and that can make all the difference in rapidly evolving situations.</p> <p>Often our partners need commercial imagery with greater resolution and refresh times than what Landsat 8 offers. We have great partnerships with commercial imagery providers, to offer all sorts of imagery. Too often receiving and processing commercial imagery is a huge pain point that slows us down and makes it harder to make use of the data. As developers we know it could be better.</p> <p><a href="">Astro Digital</a> gave us an opportunity to rebuild this workflow from the ground up. We’ve worked closely with their team to build a satellite imagery pipeline for developers and end users. We just launched a browsing and publishing platform with Astro Digital to allow anyone to discover, process, and share satellite imagery in an incredibly quick and intuitive manner. A process that previously could take days has now been cut down to minutes.</p> <h3 id="api-first">API first</h3> <p>We built an end-to-end data processing pipeline that feeds a powerful data API that unlocks possibilities for others. We broke down the fundamental goals of the Platform and built API calls around each. Those goals were to search, process, and publish.</p> <p>We built and exposed a perfomant Elasticsearch-powered endpoint, based on our previous <a href="">landsat-api</a> work, that will allow for complex queries to find exactly the data that is needed.</p> <p>{% highlight bash %}<br /> $ curl[0+TO+20]<br /> {% endhighlight %}</p> <p>But how to process the imagery? We extended our existing open source <a href="">landsat-util</a> tool to handle varying band combinations and the API offers several including true color, vegation health false color and urban false color.</p> <p>{% highlight bash %}<br /> $ curl<br /> {% endhighlight %}</p> <p>And finally, there is a simple request that can be made to process the imagery and receive a tiled map URL. This URL can be used with tools like <a href="">Mapbox</a> or <a href="">Leaflet</a> to build upon the processed imagery in any way. Full documentation, including interactive samples, can be found at <a href=""></a>.</p> <p>{% highlight bash %}<br /> $ curl -X POST –data “sceneID=LC80430332014262LGN00&amp;process=urbanFalse&amp;satellite=l8”<br /> {% endhighlight %}</p> <p>{% highlight json %}<br /> {“status”:”Image is being processed.”}<br /> {% endhighlight %}</p> <h3 id="frictionless-publishing">Frictionless publishing</h3> <p>We are using this data pipeline to power an extremely easy and visual imagery browser and publishing tool. We started with <a href="">Libra</a> as a base and modified it to meet the Astro Digital specific workflow. Libra was already designed to be quick and intuitive. We added a simple publish workflow that will process and publish images and email a link to the tiled map after processing has completed. For images processed within this visual workflow, the email contains a link to an embeddable map that can be used anywhere across the web.</p> <p><img src="" alt="publish" /></p> <p>Working with Astro Digital, we built a modern publishing pipeline that we hope will push the entire industry to build more usable tools. This is good for the industry and good for users, particularly the small governments and development organizations that are the next wave of power satellite data users.</p> 2015-04-17T15:00:00+00:00 Development Seed An OpenStreetMap for Government <p class="dropcap">The software that powers OpenStreetMap (OSM) is a fantastic open source tool for governments to manage road data. We are working with the World Bank and the Philippine government to deploy OSM software to create a collaborative tool for national and municipal authorities to manage road data.</p> <p>OSM-as-a-platform gives governments a powerful open source option for enterprise management of road data. The OpenStreetMap community benefits with more investors in the OSM ecosystem and more OSM-ready data.</p> <h3 id="osm-and-government">OSM and Government</h3> <p>OpenStreetMap is a powerful tool for collaborative mapping of critical infrastructure. Over two million OSM users have contributed 280 million road, rail, and waterway segments.</p> <p>Governments are ideal OSM users. Governments build and maintain roads and manage public transit. Governments have powerful incentives to provide road data in a manner most useful to citizens. However legal, policy, and cultural considerations prevent governments from engaging in OSM.</p> <ul> <li>Government may have a requirement for authoritative control of data for 911 purposes.</li> <li>OSM’s share-alike requirements may run foul of government publishing obligations.</li> <li>Agency IT Departments may be hesitant to authorize external services for maintaining critical infrastructure data.</li> <li>Governments may have additional data or workflow requirements that aren’t supported by OSM.</li> </ul> <h3 id="openroads---osm-for-the-philippines">OpenRoads - OSM for the Philippines</h3> <p>The National Government in the Philippines needs the ability to manage and collaboratively edit road data in much the same way as OSM, but with their own unique set of data and users. Different portions of the road network are managed by authorities at the national, provincial, municipal, or barangay. These authorities also need tools to manage data on road improvement projects, from evaluating proposed resurfacing projects to tracking the proper completion of a funded road expansion.</p> <p>The OSM software provides fantastic tools for a whole-of-government approach to road management. Working closely with the World Bank, we are building the OpenRoads Network Editor, an OSM-based system for road management across Philippines’ various road authorities. We built this using OSM-as-a-platform and utilizing tools like <a href="">iD</a> and <a href="">to-fix</a>.</p> <p><img src="" alt="" /></p> <p>This tool will allow the Government to create a full map and inventory of the countries 200,000 KM of roads. Ultimately it will also provide the analytical tools to allow the Philippines to make better decisions on infrastructure investments with real-time data.</p> <p>The World Bank recognizes the value of government managing data with collaborative open source tools. This week the World Bank is introducing this platform at a summit of the Philippine’s 1490 mayors. Today we are at the World Bank for <a href=""><em>Big Data for a More Resilient Future</em></a> to present OpenRoads and how OSM-as-a-platform can benefit other governments.</p> <h3 id="open-tools-for-open-government">Open Tools for Open Government</h3> <p>OpenRoads is part of Development Seed’s ongoing efforts to create <a href="//">open tools to support open government</a>.</p> <p>The OSM software ecosystem provides governments with a powerful open source alternative for managing road data. The OpenStreetMap community also benefits. These deployments create additional investors in the OSM software and more OSM-ready data.</p> 2015-04-15T14:00:00+00:00 Development Seed Tracking Metadata in Real-time <p class="dropcap">This week we’re releasing more tools to track OpenStreetMap metadata. Together with the American Red Cross, we’re launching <a href="">OSM Metadata API</a>, a tool to help enable analysis of OSM’s rich metadata at the user and comment level. By using hashtags in changeset comments such as <a href="">#MissingMaps</a>, API access can enable groups like the Red Cross to gain feedback from OSM deployments.</p> <p>We’re building off our <a href="">osm-meta-util work we released several weeks ago</a> to store and index OSM metadata. We built the API on <a href="">Elasticsearch</a> and <a href="">Node.js</a>. Using Elasticsearch, we index and store metadata information such as changeset_id, user, created_at, bounding box, comment. You can filter these logs by hashtag and keyword as they come in real-time, as well as build a database of historical metadata logs.</p> <p>We’re rolling out a sample API today with the code. You can access the endpoint here: <a href=""></a>.</p> <p>You can use the full power of Apache Lucene Search to browse the data. For example, to browse for the #missingmaps hashtag between two dates:</p> <pre><code>[2015-04-08T00:00:00Z+TO+2015-04-09T00:00:00Z]+AND+comment:"missingmaps" </code></pre> <p>In the past week, #missingmaps events have helped make 69,865 edits to OpenStreetMap in places like Haiti, Iraq, Zimbabwe, and Tanzania.</p> <p><img src="/assets/graphics/content/osm-meta-api-release-cover.png" alt="" /></p> <p>Check out the <a href="">API guide</a> for full endpoint documentation.</p> <p>All code is available via a <a href="">GitHub repository on OSM Lab</a>. To get started quickly, you can <a href="">deploy the API as a Heroku app</a>. Try it out and contribute back to the OSM community.</p> 2015-04-14T14:00:00+00:00 Development Seed Howdy, Anand Thakker! <p><img src="" alt="Anand looking good" /></p> <p><a href="">Anand Thakker</a> likes the messy stuff. Whether he’s using historic satellite imagery to measure electrification or finding better ways to teach high school math, Anand embraces hard problems. He brings creativity, energy, and thoughtfulness to every project.</p> <p>Anand is going to help us to build powerful tools that make a real difference. Anand has always been engineering. He built a search engine with his high school friends. At a startup right out of college, he worked on tools for analyzing and debugging XML-based web services. Then, for seven years, he devoted his energy to teaching high school math and computer science in Baltimore. There too, Anand engineered curriculum and teaching methods for improving students’ problem solving skills.</p> <p>Anand studied math and computer science at Carnegie Mellon University and has a masters in education from Harvard University. Give him a shout on <a href="">github</a> or <a href="">twitter</a>.</p> 2015-04-02T16:00:00+00:00 Development Seed Welcome Emily Bremner! <p><img src="" alt="Emily and the tree" /></p> <p>International development is hard. <a href="">Emily Bremner</a> first saw this at a school in Kenya for Somali refugees. She was 16. Over more than a decade since, Emily has pushed to improve peoples’ lives in places from El Salvador to Afghanistan. She’s been teargassed in Djibouti and battled bureaucracy back in DC.</p> <p>Emily has worked on all sides of international development and has seen its successes and its failures. Emily is going to work on our business strategy and operations. She is going to help us to better solve hard social problems and to make sure our work is impactful.</p> <p>We had the opportunity to work with Emily when she was at <a href="">Democracy International</a> to open up election data in <a href="">Lebanon</a> and <a href="">Tunisia</a>. She understood then the opportunities and the limits of data and technology to make meaningful change. We look forward in working with her to expand those limits.</p> <p>Say Bienvenue or Hola to Emily <a href="">on twitter</a>.</p> 2015-04-02T14:00:00+00:00 Development Seed Optimizing Landsat-util <p class="dropcap">Two weeks ago we launched a <a href="">new version of Landsat-util, v0.5</a>, our utility for searching and processing Landsat satellite imagery. This version is now faster than it was before. To do this we rewrote most of the internals to use flexible python frameworks.</p> <p>Below is a deep dive into how we’ve improved Landsat-util to make it faster and easier to use.</p> <h2 id="dependency-hell">Dependency Hell</h2> <p>Landsat-util downloads Landsat files, pulls out the individual bands representing wavelengths of light, corrects contrast, warps them and combines them to make a colored image you can add on a web map.</p> <p>Initially, landsat-util was written as a command line wrapper to <a href="">existing pipelines</a>:</p> <ol> <li>Scale bands with <code>gdal-translate</code></li> <li>Warp with <code>gdal-warp</code> to the correct projection</li> <li>Combine bands with <code>ImageMagick</code></li> <li>Contrast correct with <code>OpenCV</code></li> <li>Pansharpen with <code>orfeoToolbox</code></li> <li>Gamma correct with <code>ImageMagick</code></li> <li>Add geographic information back with <code>gdal_edit</code></li> </ol> <p>We needed a lot of dependencies to process an image, and they’re not particularly optimized for scripting. ImageMagick, GDAL, orfeoToolbox and openCV are monolithic frameworks that don’t allow for cherry-picking functions. Installing all these frameworks can be a <a href=";q=install">painful experience</a>.</p> <p>This toolchain combination is also tedious because it creates inherent bottlenecks. To communicate between all tools, we need to write temporary files to disk and read them back in at every step.</p> <h2 id="enter-rasterio">Enter rasterio</h2> <p><a href="">Rasterio</a> is a great python library written by <a href="">Sean Gillies</a> at Mapbox to work with raster data. It wraps around gdal and abstracts the band data as <a href="">Numpy</a> arrays.</p> <p>By standardizing the input, rasterio allows us to minimize our dependencies and use fast, in-memory, scientific libraries like <a href="">scikit-image</a>.</p> <p>Here’s the guts of our new process:</p> <h3 id="read-in-bands-with-rasterio">Read in bands with rasterio</h3> <p>{% highlight python %}<br /> with rasterio.drivers():<br /> with‘LC82040522013123LGN01_B4.TIF’) as band4:<br /> with‘LC82040522013123LGN01_B3.TIF’) as band3:<br /> with‘LC82040522013123LGN01_B2.TIF’) as band2:<br /> with‘LC82040522013123LGN01_B8.TIF’) as band8:<br /> band4_s = band4.read_band(1)<br /> band3_s = band3.read_band(1)<br /> band2_s = band2.read_band(1)<br /> band8_s = band8.read_band(1)<br /> {% endhighlight %}</p> <h3 id="scale-bands-with-scikit">Scale bands with <code>scikit</code></h3> <p>{% highlight python %}<br /> from skimage import transform as sktransform</p> <p>band4_s = sktransform.rescale(band4_s, 2)<br /> band3_s = sktransform.rescale(band3_s, 2)<br /> band2_s = sktransform.rescale(band2_s, 2)<br /> {% endhighlight %}</p> <h3 id="warp-with-rasterio">Warp with <code>rasterio</code></h3> <p>{% highlight python %}<br /> for color, band in zip([r,g,b,b8], [band4_s, band3_s, band2_s, band8_s]):<br /> reproject(band, color, <br /> src_transform = src.transform,<br /> src_crs =,<br /> dst_transform = dst_transform, <br /> dst_crs = dst_crs,<br /> resampling = RESAMPLING.nearest)<br /> {% endhighlight %}</p> <h3 id="pansharpen-using-numpy-operations">Pansharpen using <code>numpy</code> operations</h3> <p>{% highlight python %}<br /> m = r + b + g<br /> pan = 1/m * b8<br /> r = r * pan<br /> b = b * pan<br /> g = g * pan<br /> {% endhighlight %}</p> <h3 id="contrast-correct-and-gamma-correct-using-scikit-image">Contrast-correct and gamma correct using <code>scikit-image</code></h3> <p>{% highlight python %}<br /> # using CLAHE<br /> from skimage.exposure import equalize_adapthist<br /> for band in [r,g,b]:<br /> band = equalize_adapthist(band, clip_limit=0.02)<br /> {% endhighlight %}</p> <h3 id="write-to-disk-using-rasterio">Write to disk using <code>rasterio</code></h3> <p>{% highlight python %}<br /> with<br /> tiffname,’w’, driver=’GTiff’,<br /> width=dst_shape[1],height=dst_shape[0],<br /> count=3,dtype=numpy.uint8,<br /> nodata=0,<br /> transform=dst_transform,<br /> photometric=’RGB’,<br /> crs=dst_crs) as dst:<br /> for k, arr in [(1, r), (2, g), (3, b)]:<br /> dst.write_band(k, arr)<br /> {% endhighlight %}</p> <p>The toolchain consists of only python libraries, and no other dependencies. Rasterio inherently supports GeoTiff so we don’t lose geo-information along the way.</p> <p>Landsat-util is open source, and we encourage developers to improve on our process. Fork <a href="">our repo</a>!</p> <h2 id="how-fast">How fast?</h2> <p>By using <code>rasterio</code>, <code>numpy</code> and <code>scikit</code>, we eliminate the disk bottleneck, and we regain transparent control over the pipeline.</p> <p>We ran tests on a couple of AWS machines. Each test was conducted 5 times and the resulting times were averaged.</p> <ul> <li>R3.large: 2-core 2.5 GHz Intel Xeon (Ivy Bridge) with 15GB RAM and 32GB SSD.</li> <li>C4.2xlarge: 8-core 2.9GHz Intel Xeon (Haswell) with 15GB RAM and 32GB SSD.</li> </ul> <h3 id="results">Results</h3> <div class="table-responsive"> <table> <thead> <tr> <th>instance type</th> <th>process type</th> <th>old-landsat-util</th> <th>new-landsat-util</th> <th>speedup</th> </tr> </thead> <tbody> <tr> <td>R3.large</td> <td>non-pansharpened</td> <td>252.7s</td> <td>122.05s</td> <td><strong>2x</strong></td> </tr> <tr> <td>R3.large</td> <td>pansharpened</td> <td>846.9s</td> <td>349.17s</td> <td><strong>2.4x</strong></td> </tr> <tr> <td>C4.2xlarge</td> <td>non-pansharpened</td> <td>216.6s</td> <td>106.67s</td> <td><strong>2x</strong></td> </tr> <tr> <td>C4.2xlarge</td> <td>pansharpened</td> <td>438s</td> <td>290s</td> <td><strong>1.5x</strong></td> </tr> </tbody> </table> </div> <p>On all but the largest machine, the new landsat-util is at least twice as fast as previously.</p> <p>And it gets better: a significant amount of time (about a minute on average) is used up to decompress the NASA bundle after downloading it. Landsat-util v0.5 takes advantage of <a href="">AWS’s new landsat archive</a> of unzipped bands, saving even more time.</p> <p>We hope you enjoy the new landsat-util! Fork it, modify it, break it or just use it and tell us about all the ways you’re incorporating satellite data in your apps.</p> 2015-03-28T02:00:00+00:00 Development Seed We're in the Philippines <p class="dropcap">This week, <a href="/team/ian-schuler/">Ian</a> and I are in the Philippines working alongside the World Bank to set up collaborative, open source tools for government to track critical infrastructure, particularly road data.</p> <p>Good road data is crucial for smart development and disaster response. To help the national and local governments manage their road infrastructure, we’ll be building on open tools such as <a href="">iD</a>, the popular editor for <a href="">OpenStreetMap</a>. Our goal is to use open-source tools to create better data infrastructure, at an enterprise level, tailored to the information needs of government.</p> <p>We are <a href="">confessed fans</a> of OpenStreetMap and the ecosystem that surround it. These tools are battle-tested and optimized for the quantity of data that flows daily through OSM. We think these tools can make government work better, and so far we’ve seen terrific responses from officials here.</p> <p>We’ll be in the Philippines until next Tuesday. Ping us on <a href="">Twitter</a> if you’d like to meet and talk OSM and managing infrastructure data.</p> 2015-03-27T10:00:00+00:00 Development Seed Bem-vinda Caroline Portugal <p>Welcome Caroline! Designer and architect <a href="/team/caroline-portugal/">Caroline Portugal</a> is joining Development Seed.</p> <p><img src="/assets/graphics/content/caroline-portugal-card.jpg" alt="" /></p> <p>Caroline is going to build thoughtful, beautiful software. Caroline hails from Brazil, where she studied and practiced Architecture and Urban Planning. She worked on architecture projects in Brazil and the US ranging from <a href="">Rio’s Olympic Park Master Plan</a> to the National Museum of African American History &amp; Culture. Caroline’s latest work has been in visual design. She recently completed a graduate degree in Graphic Design from MICA.</p> <p>Caroline’s work is moving. It is visually stunning. It is grounded in solving practical challenges with humanity, creativity and fun. Check out her work on <a href="">bike lanes</a> and <a href="">the future of contraception</a> to get a sense of why we are delighted to have her on the team.</p> <p>Caroline speaks fluent Portuguese. Say bem-vinda to <a href="">Caroline on twitter</a>.</p> 2015-03-24T10:00:00+00:00 Development Seed Powering Landsat Power Tools <p class="dropcap">Amazon Web Services just opened Landsat on AWS, a publicly available archive of Landsat 8 imagery hosted on their reliable and accessible infrastructure. This investment by the AWS open data team has a big impact on our work to make satellite imagery more accessible.</p> <p><a href="">Libra</a> is an open source Landsat imagery browser that we built with <a href="">Astro Digital</a>. Libra now has options on some scenes to download individual bands related to specific types of imagery analysis like NDVI, or Urban False Color. The most recent Landsat–8 images are now available for download up to two days sooner. Last week we rolled out a <a href="/blog/2015/03/10/releasing-landsat-v05/">new version of landsat-util</a>, our open source utility for processing Landsat imagery. The new version is much faster and allows you to build <a href="">false color composites</a> on the fly. These improvements to Libra and landsat-util are possible because we started using Landsat on AWS, which is a publicly available archive of Landsat 8 imagery hosted on Amazon S3 that is <a href="">publicly available today</a>.</p> <p><a href=""><img src="" alt="libra-band-download" /></a></p> <p><em>Libra now allows you to download the satellite bands for advanced analysis</em></p> <p>Our newest releases of Libra and landsat-util utilize Landsat on AWS for 2015 imagery. Landsat on AWS provides 2015 imagery as unzipped individual bands. AWS makes this imagery available extremely quickly, often within hours of capture. We can pull only the data that we need and to work with it immediately.</p> <p>Landsat 8 imagery is an incredibly powerful resource. It is some of the most valuable open data produced by the US Government. Our partners rely on Landsat data for everything from evaluating droughts to tracking conflict. However, until now, individual bands of Landsat imagery has never been available via predictable download endpoints that we can integrate into our tools.</p> <p>Libra and landsat-util now allow our partners to get imagery sooner and process it faster. Speed and ease are critical to our partners who use this data to respond to natural disasters, prevent hunger, and monitor elections.</p> <p><img src="" alt="gulf of guinea" /></p> <p><em>Landsat images of the Gulf of Guinea, processed with <a href="">landsat-util</a></em></p> <p>Thanks to the AWS team and collaborators–<a href="">Frank Warmerdam</a> at <a href="">Planet Labs</a>, <a href="">Charlie Loyd</a> at <a href="">Mapbox</a>, and Peter Becker and others from <a href="">ESRI</a>–for building a Landsat archive with developers in mind.</p> <p><a href="">Libra</a> and <a href="">landsat-util</a> are open source and on Github. Go ahead and fork or contribute.</p> 2015-03-19T14:00:00+00:00 Development Seed Launching v0.5 of landsat-util <p class="dropcap">We just released a new version of landsat-util, <a href="">version 0.5</a>. This version is lighter and has fewer dependencies. Landsat-util v0.5 downloads and processes images much faster and gives users more control.</p> <p><a href="">Installing landsat-util</a> now is considerably easier for Mac and Linux users. We are still working to make it as easy to run on Windows.</p> <p>This new version reflects a significant change in our thinking on landsat-util. We’ve moved to using faster and simpler frameworks to optimize processing. We removed heavy dependencies, such as ImageMagick and Orfeo Toolbox, that were causing installation problems. We leveraged faster processing frameworks like <a href="">Rasterio</a>, <a href="">numpy</a>, and <a href="">scipy</a>. These changes significantly optimize disk and memory usage resulting in faster and less error-prone processing. Landsat-util can now process images 3x faster.</p> <p><img src="{{ site.baseurl }}/assets/graphics/content/landsat-util-v05.gif" alt="Running the utility" /></p> <p>We’ll post more of the technical details on how we rebuilt landsat-util and what’s under the hood. In the meantime hit us up at FOSS4G in San Francisco all week to learn more.</p> 2015-03-10T15:30:00+00:00 Development Seed Collect and verify mobile reports <p>A common scenario for mobile reporting looks like this:</p> <ul> <li>A group wants to collect reports from their own trusted network and also to crowdsource reports from the public.</li> <li>The group has some process to try to verify crowdsourced reports and needs to track verification.</li> <li>The group wants to publish this data using simple visuals that answer their core question and invites comparison.</li> </ul> <p>We recently worked with Pursue Lebanon to build a system for reporting breakdowns in service delivery in Lebanon’s Palestinian refugee camps. Pursue has been working with community organizations in all twelve refugee camps for the past five years. With more attention on the refugee situation in Lebanon, there are opportunities to push for improved service delivery with better, timely data. With so many tools for mobile data collection, you’d think that this should be easy to do with open source tools. It’s not. Here is our experience.</p> <h3 id="collecting-data">Collecting Data</h3> <p><a href="">OpenDataKit</a> is a great open source tool for data collection on Android. When you have control over the device your volunteers are using, ODK is great. <em>(Side note: If you like ODK, check out <a href="">OpenMapKit</a> an exciting project of the American Red Cross, SpatialDev, and Ona).</em></p> <p>Like many organizations, Pursue relies on reports from volunteers using their own phones and can’t expect them to report through an Android app. This is why the ODK ecosystem and tools like <a href="">Enketo</a> and <a href="">Formhub</a> are interesting. Enketo allows us to serve a simple web form on any device. Formhub gives us the ability to convert and manage Excel-based forms. While this ecosystem is great for authoring forms, we found that this system had some shortcomings:</p> <ul> <li>managing multiple datasets can be a challenge,</li> <li>API lacks some features,</li> <li>system is resource hungry, and</li> <li>difficult to deploy and maintain.</li> </ul> <p>We use these tools for what they are best at, generating web forms from Excel documents. We pull out the data via the API to use in other open source tools.</p> <p><img src="" alt="" /></p> <h3 id="verification">Verification</h3> <p>No tool that we looked at had good workflow for verification. Most had no verification or had only a simple yes/no toggle. Many organizations need to append further verification data and notes to each report. Pursue’s partners are collecting and verifying thousands of reports. To make this manageable, the system must have a workflow for quickly reviewing whether reports are verified true, verified false or are unverified.</p> <p><img src="" alt="" /></p> <p>We used Django to build a verification platform that is useful and usable for the data verifier who is triaging hundreds of reports a day and for the field organizer who is trying to track the status of single report.</p> <h3 id="publishing">Publishing</h3> <p>To publish and visualize the collected data, we designed a map and report interface that highlights the trends in service breakdowns between camps and within neighborhoods in camps.</p> <p>It is important to clearly show users whether information is verified. For simplicity, we use a simple checkbox to filter out all reports that are “unverified” or “verified false”, even though we distinguish between these in the verification platform. We showed the verification status prominently on all report listings.</p> <p>Reporters can select a neighborhood and choose a level of geographic precision. To protect the privacy of reporters, the system does not require reports to submit precise location information. Trusted administrators use the precise location when it is provided to verify reports. Regardless of the location precision, the public platform only shows reports to the neighborhood level.</p> <p><img src="" alt="" /></p> <p>Join <a href="">our talk on Wednesday morning</a> to hear more. We’ll follow up later in the day with a Birds of a Feather session discussing how the open source data collection community continues to grow.</p> 2015-03-10T14:30:00+00:00 Development Seed Hello San Francisco; Ready for FOSS4G <p class="dropcap">We’re out in San Francisco this week for <a href="">FOSS4G North America</a>. Look for <a href="">Alireza</a> and I at the conference or at our session <a href="">Wednesday morning at 10:30am</a>. Alireza will discuss using the OpenDataKit ecosystem to build mobile reporting and verification tools for refugee camps in Lebanon. We’ll also be at the Birds of a Feather session discussing the future of OpenDataKit.</p> <p><img src="" alt="SF Landsat" /><br /> <em>Landsat 8 image over San Francisco on December 31, 2014</em></p> <p>This week we will release new versions of <a href="">landsat-util</a> and <a href="">Libra</a> with some powerful new features. We are also working on some tools for <a href="">managing OpenStreetMap data</a>. We look forward to collaborating with old and new friends in the open source community to improve the tools for mobile data collection, Open Geo, and satellite imagery processing. Tweet us at <a href="">@nas_smith</a> or <a href="">@scisco7</a> to meet up.</p> 2015-03-09T18:00:00+00:00 Development Seed Tapping into OpenStreetMap Metadata <p class="dropcap">We just launched <a href="">v0.1 of a new tool</a> to tap into OpenStreetMap changeset metadata. We built the tool in partnership with the American Red Cross as part of the infrastructure for tracking efforts such as <a href="">#MissingMaps</a>.</p> <p><a href="">OpenStreetMap changesets</a> give us access to a wealth of metadata information that is not specifically geographic but incredibly rich. Metadata is helpful in understanding the changing nature of OSM. This is different from using geographic APIs like <a href="">Overpass</a> because metadata contains commit text, number of edits, which editor was used, etc. With metadata, we can <a href="">track hashtags</a>, analyze commit text or aggregate user metrics.</p> <p>In 2014 alone, users committed over 6 million changesets to OSM. As OpenStreetMap’s metadata grows, dealing with the sheer amount can be daunting. We built <a href="">osm-meta-util</a> as an experiment in making OSM metadata easier and faster to use.</p> <p><img src="{{ site.baseurl }}/images/gifs/osm-meta.gif" alt="Running the utility" /></p> <p>osm-meta-util focuses on two core functions: downloading the minutely compressed metadata files and serializing into JSON. We convert compressed OSM XML files containing multiple commits to a stream of JSON objects that can be piped to any tool or API.</p> <p>You can use the library in a Node application or as a command-line utility to download all the data between two dates:</p> <p>{% highlight javascript %}<br /> MetaUtil({<br /> ‘start’: process.argv[2],<br /> ‘end’: process.argv[3],<br /> ‘delay’: process.argv[4]<br /> }).pipe(process.stdout)<br /> {% endhighlight %}</p> <p>In combination with <a href="">jq</a>, to get a commit history we can simply run:</p> <p>{% highlight sh %}<br /> node app 001181708 001181721 1000 | jq -c ‘{date: .created_at, text: .comment}’<br /> {% endhighlight %}</p> <p>If you don’t give the tool any parameters, it will get the latest changesets and update every minute.</p> <p>We’re using this utility to experiment building a metadata API with the American Red Cross. But we know there are many more uses of the rich OSM metadata and want to see what others can do with the tool. Together with the American Red Cross we’ve put this on <a href="">OSM Lab</a>, a Github organization for OSM related projects. Follow the development of osm-meta-util on <a href="">Github</a>.</p> 2015-02-19T18:00:00+00:00 Development Seed Open Data Day Garage Party <p class="dropcap">We love open data. And we love to talk about it over drinks with other open data lovers. Next Friday, February 20th at 7:00pm we are hosting an <a href="">Open Data Day celebration</a> in the Mapbox Garage. Head over to the Garage after the first day of <a href="">Open Data Day DC</a> to talk about open tools for open data.</p> <p><a href=""><img src="" alt="" /></a></p> <p>Even if you can’t make it to Open Data Day DC (or didn’t grab a spot on the now closed list), come over and share the #opendatalove.</p> <p>Let us know if you’re in by <a href="">RSVPing now</a>.</p> 2015-02-13T10:00:00+00:00 Development Seed Data Hungry Happy Hour <p class="dropcap">We’ll be at the <a href="">Thought For Food Global Summit</a> next week working with some brilliant people on the biggest challenges in agriculture and feeding the hungry. Look for <a href="">Olaf</a> at the Summit.</p> <p>If you are in Lisbon for the Summit, please join us for a <a href="">Data Hungry happy hour</a> on Thursday evening. We’ll be geeking out on better data, satellites, and sensors can contribute to better food policy. The Happy Hour kicks off at 18.30 in Fabulas in the center of Lisbon. The first couple of rounds are on us. You can <a href="">RSVP here</a>.</p> <p><a href=""><img src="" alt="Data Hungry Happy Hour" /></a></p> 2015-02-06T10:00:00+00:00 Development Seed Welcome Mariano Arrien-Gomez <p class="dropcap"><a href="/team/mariano-arrien-gomez/">Mariano Arrien-Gomez</a> is an artist and designer. He builds <a href="">beautiful graphics</a> and data visualizations that make our products more compelling, creative, and humane. Mariano directs his design skills toward the issues and topics he is passionate about, from preserving local parks, to genetically modified food, to soccer. He utilizes a range of <a href="">visual mediums</a> including illustration, painting murals, and photography.</p> <p><img src="" alt="mariano" /></p> <p>In addition to his work at Development Seed, Mariano is active in the DC art scene. Mariano received a Bachelors of Fine Arts in Graphic Design from Virginia Commonwealth University. He is fluent in Spanish.</p> 2015-02-02T10:00:00+00:00 Development Seed