![]() ![]() ![]() To that end, I use the boundary extents of my spatial data of interest to create an appropriate bounding box for my download. There’s a few ways of doing this, but I find with my workflow I will typically start off with some existing spatial data for which I want to have a vector tile-based background. You can download tiles by specifying bounding box of coordinates. Thanks to an AWS Research Award Nextzen will continue to host the vector tile service through 2018 (and hopefully longer). Unfortunately Mapzen is now defunct but a number of other providers continue to host Mapzen’s open-source tools. This included an excellent vector tile service using OpenStreetMap data. These tools included a digital map rendering engine, search and routing services, open-source data tools for map data, terrain, and transit, gazetteers, and tile servers. Mapzen was an innovative business that tried to be viable by making unreal open-source software with an emphasis on web mapping and geography products. The project is available on github as well as CRAN. Though Mapzen itself has gone out of business, rmapzen can be set up to work with any provider who hosts Mapzen’s open-source software, including geocode.earth, Nextzen, and NYC GeoSearch from NYC Planning Labs. Rmapzen is a client for any implementation of the Mapzen API. This post relies on Tarak Shah’s excellent rmapzen package which provides an R front-end to several Mapzen APIs. The metadata embedded within vector tile objects is what provides information on what those objects are such as type and name–and that allows a lot of flexibility in choosing what is useful and what is not. In this simple demo, I want to focus on the advantages that come with using vector tiles. Google Maps has been vector tile based for over 5 years now. Because they use far less data to transfer, vector tile based maps are much faster and more responsive, and most map services now use them instead. Unlike raster tiles, vector tiles are just layers of numbers and strings containing geometry and metadata. Even if using raster tiles from different providers offering different styles (e.g. Stamen or Thunderforest or OSM), in the end the user was resigned to essentially having an unalterable image in their plot with no control over what exactly is displayed in those features. R packages like ggmap can retrieve raster-based tiles to use as backgrounds for ggplot maps. As these tiles were images, that the only information available was pixel and colour. For many years after their introduction, these slippy maps used raster-based bitmaps as their tiles. The concept underpinning those maps was the use of tiles: pre-rendered map cells for every specified zoom level that would be loaded by your browser as you scrolled through a map. When MapQuest and later Google Maps came on the scene we were blown away by the detail, speed, and convenience of “slippy maps” that you could scroll, pan, and drag across. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |