NACIS 2026

SpatialpIpe
ADDA Youcef, TECH
SpatialPipe is an ETL platform built around a native C++ engine, with a modern UI powered by Vue 3, TypeScript, Vite, and Electron.
The ambition is to build a serious open alternative to commercial ETL tools: fast, transparent, scalable, and designed to last.
Today, the platform already integrates multiple data readers, including OSM and WFS. The roadmap goes much further: GNSS PPP (Precise Point Positioning), photogrammetry workflows, 3D reconstruction from imagery, alignment with ground control points, and more direct access to free APIs and public data sources.
A lot of the ideas behind SpatialPipe are still missing from most ETL tools. That is also why C++ was chosen for the backend: to build on a strong foundation for performance, scalability, and long-term evolution from day one.

Be A Legend, or Maybe Just Make a Better One: Five Tips and Tricks for More Effective Map Legends
Vicky Johnson-Dahl
Break out of default mode and discover five ways to better integrate your legend into your map. In this lightning-fast talk we'll explore different methods and approaches for using the legend to enhance your cartographic storytelling skills.

GLIDE: An Open Source Ecosystem for Interactive Web Mapping Applications
Julio Perez, Frost Institute for Data Science and Computing at University of Miami; Chris Mader - Frost Institute for Data Science and Computing at University of Miami
GLIDE (Geospatial Layered Information Display and Exploration) is an open-source ecosystem for building and deploying interactive web-based mapping applications. Built on SvelteKit, Leaflet, MapLibre, and Apache Solr, GLIDE supports dynamic bivariate rendering, layered visualization, and on-the-fly attribute searching within vector tiles.
This session will provide an overview of the GLIDE architecture and include a live demonstration showing how the platform supports the rapid development of modern, multi-tier mapping applications capable of rendering vector tiles, raster datasets, and other interactive mapping layers. The presentation will conclude with a complete web application demonstrating multiple rendered datasets and interactive mapping capabilities.

How to create novel OSM-based maps more easily with Sourdough vector tiles
Jake Low, OpenStreetMap US
Vector tiles power the zoomable interactive maps seen in many websites and apps. A vector tile schema defines what data is made available in these tiles, and how it's organized.
Sourdough (https://sourdough.osm.fyi/) is a new vector tile schema designed to make the wealth of data in OpenStreetMap more easily accessible to map makers. It differs from prior schemas by including far more data in each tile, in order to support the widest possible range of cartographic expressions. Sourdough makes it easier to create novel maps that depict data which exists in OSM but isn't shown on most general-purpose basemaps.
This talk will show you how you can use open-source, browser-based tools to start creating your own maps with Sourdough tiles.

Using Blender to make Big Maps of Small Places
Maddy Grubb, self employed
Making a large-format map of a geographically small place means getting close, really close.
This talk walks through a Blender workflow for large-scale interpretive mapping using a historic district in the Tetons as a case study. I'll cover what I've learned about modeling a landscape beyond the scale at which data exists: using bump mapping to add texture, integrating aerial imagery and LiDAR, scattering vegetation across ecosystems, and modeling historic buildings. I'll also share how I photographed the actual walls and chinking in the field and brought those textures directly into the map. The goal is to create a map that captures how a place feels and makes visitors want to get out and connect with the surrounding landscape.

New Methods for Generating Vector Hachures
Daniel P. Huffman, somethingaboutmaps; Andrew Tyrrell, South Arrow Maps
We'll share some new techniques for depicting terrain in a sketched style that's reminiscent of hachuring methods from centuries past. Taking only a DEM as an input, a bit of GIS trickery can be used to automatically draw thousands of lines that illustrate major elevation features from a planimetric (top-down) view. With some additional work in Blender, we can even draw a sketch our terrain from an oblique view, all with minimal human effort. These methods unlock new styles for cartographers who don't want to do the painstaking work of manual terrain sketches.