What’s new, October edition
Welcome back to the Map of Life blog! I’m Gaurav Vaidya, a first-year graduate student at the Guralnick lab in beautiful Boulder, Colorado. I joined the Map of Life team just under two months ago, and have been having a great time working on the project. In these two months, Map of Life has had a lot of fantastic new features approaching completion, and we thought the time was ripe to show some of them off to you!
Our most impressive new feature is our map rendering. As you may recall from earlier blog posts, mapping is handled by our backend, set up on a Linode VPS server. Aaron recently restructured our backend to use Windshaft, a high-speed map tiler generously released under an open-source license by Vizzuality, based on cutting-edge open-source tools such as Mapnik, PostGIS, Redis and node.js. Between Windshaft and Aaron’s work on HTML5 canvas, we’ve achieved some unbelievable results. At the moment, we’re rendering tiles without caching in the hundreds of milliseconds time-frame. This includes support for species occurrence data, protected areas, expert species range maps, ecoregions or any other geo-referenced vector or point data you care to throw at it. Aaron is currently working on merging these innovations into our main data preparation work flow.
Meanwhile, John has been busy adding raster environmental layers to Map of Life, probably with PostGIS’ upcoming raster support. Supporting raster layers is a first step towards linking all the species geographic distribution data already in Map of Life to local environmental variables such as climate, land cover and vegetation. This will also help us facilitate analyses on our platform, meeting one of the key goals for Map of Life. We’ll definitely be talking more about the planned analyses in future blog posts, so stay tuned!
Map of Life depends on data producers, compilers, and aggregators to add their data to the project, so it’s vitally important to ensure that they can do so quickly and easily. With a lot of help from the entire team, I am taking a first stab at this process. Before the TDWG 2011 Annual Conference (only a week away!), we hope to have all the scripts in place to provide an efficient pipeline from shapefiles and accompanying Map of Life-specific metadata to pretty, searchable layers on your favourite browser. So far, we’ve managed to compress most of the functionality we need into one easy-to-use program. This is a great area of Map of Life on which to have started, since it lets me connect with all the components of the software — from data preparation to map visualization — while also giving me a chance to work closely with the entire team.
There are many other exciting things we’d like to talk to you about, from analyses we have planned, to our TDWG demonstration next week, to the first release of a demo you can play around with yourself. For now, though, this brief glance will regrettably have to suffice. Please ask us any questions in the comments — it’s a huge help for us in gauging interest in features for the software, as well as for topics we need to cover in future blog posts. As always, Map of Life’s source code and open issues are available online, so please do contact us or contribute there if you have any specific concerns.