TurfJS - The simple, modular, geospatial library. I have done a presentation on this that included glitch demos. Available (here)[https://siliconrob.github.io/github-web/turfjs/intro_turfjs.pdf]. I would for State of the Map cut out the geojson material and focus on the library examples. Also open to the 5 minute lightning talk, would cut the material down.

Machine learning can be used to predict map features with high levels of accuracy. However, the outputs of the process are not usually suitable for direct import into OpenStreetMap because of their format, noise in the data, or duplication of existing OpenStreetMap features. We'll present a variety of technical workflows and tools for solving these problems. We'll also show how these approaches can be abstracted for non-machine learning use cases like better bulk uploads and data conflation.

Over the past year, OSMUS has been conducting the first-ever community census. With a couple hundred responses to-date, the results are ready to be shared. This lightning talk will describe the demographics and the editing trends discovered in our community so far through this survey. Rather than list off statistics, the talk aims to be graphics heavy to promote easy sharing /posting of results to social media and be digestible in a short amount of time.

In the Ecuadorian Amazon the Waorani indigenous people are eighteen months into a five year mapping project. Their aim is to create maps that communicate the unique way they use and relate to territory, to help defend their land from external threats. They are doing this by collaboratively mapping hunting and fishing grounds, medicinal plants and historical and spiritual sites. However, despite their desire for exactly the kind of simple and collaborative mapping that OSM offers, as it stands OSM is neither a possible nor completely suitable tool for them. Waorani communities lack cellphone and internet coverage and the nature of some of the GIS data makes it too culturally sensitive for automatic upload into an open-data situation. Digital Democracy has developed an offline, open-source mapping tool, Mapeo, based on iD Editor, with edits stored in a local peer-to-peer database that implements the OSM API. The Waorani mapping team learnt Mapeo in one day, and now manage all their own data, choosing when and how to make it public. This talk will describe the Waorani’s mapping process, and the benefits which an offline peer-to-peer database can offer to use-cases within the OSM community.

I used census data, OSM streets, pgrouting, and a k-mean clustering algorithm to build 11 voting districts in Virginia. https://github.com/openbrian/districtbuilder/wiki/sampled-data,-plain-distance-function

Crossing a railroad at grade is common in the US, and also quite dangerous. More than 2000 traffic incidents occurred at level crossings in 2016 alone, resulting in 265 fatalities(1). Those numbers could go down if drivers were aware of them. That requires accurate data about the location of at-grade crossings. Data that even the Federal Railroad Administration does not have. OpenStreetMap community member Mike Nice decided to do something about it. As a follow-on to the MapRoulette V1 rail crossings challenge (which was only partially completed), he created a MapRoulette V2 challenge for the community to look at the remaining 63,000 locations where a railroad crosses a road on OSM and 5300 locations where a railroad crosses a pedestrian path. Ten months later, this challenge is complete, and OSM now has what is likely to be the most accurate and up-to-date inventory of at-grade railroad crossings in existence! 1 (Data from FRA)[https://oli.org/about-us/news/collisions-casulties]