The SparkGeo team has tested Lovable and asked it to create a simple interactive map showing Canadian cities and their population.
results weren’t always consistent. Running the same prompt twice sometimes yielded drastically different outcomes: one time clean and functional, the next time oddly styled or partially broken. In one case, even after explicitly stating not to use any frameworks, Lovable still generated a React-based implementation. A quick follow-up prompt corrected this, and to its credit, the revised output was clean, readable vanilla JavaScript.
That responsiveness is promising, but also highlights an important caveat: some knowledge of development (especially in geospatial UI) is still necessary. Without it, spotting issues or debugging quirks could become a frustrating barrier. Lovable gets you 80% of the way there, but that final 20% – the part where accuracy, accessibility, and usability matter most, still depends on human expertise.
Even for simple applications, you need a knowledgeable engineer to create a finished product. This is fine if you treat generative AI assistants as a tool, not the solution.
But an important question lingers. While engineers rely more and more on generative AI to build applications, they spend less time understanding the concepts and libraries they’re working with, making mistakes and gaining experience allowing them to understand and improve the code they write. By extension, engineers never learn enough about the technology they’re putting to work, maybe even unlearn some of the things they know now. If that turns out to be the case, who will maintain applications largely built with AI in the future?
I’d love to see a similar interface for generating GDAL commands; Cameron Kruse:
Tippecanoe has great documentation in its ReadMe with all the info you need to get started, but I’ve often found that half the battle is finding the commands, deciding which ones to use and stringing them together into a coherent command you can actually run in your terminal. I’ve made this a little easier by taking many of the Tippecanoe features and converting them into an interface you can use to generate your commands. The premise of this tool is you select all the options you want from the dashboard and a command is generated below you can copy and paste in your terminal.
The good folks at Heigit have released ohsome-planet, a handy tool to turn OpenStreetMap history data from PBF into GeoParquet files, ready to use in common GIS applications.
Working with raw OSM data presents several challenges due to its complex structure. Typically, users require data that is readily compatible with Geographic Information System (GIS) applications. Our new tool streamlines this process, providing a structured and GIS-ready dataset for improved usability.
The tool also enriches OSM element data by integrating information from OSM changesets and administrative boundaries. This additional contextual data allows for more efficient and straightforward spatial analysis, further improving the utility of OSM datasets.
The tool is written in Java and you have to build it yourself; a small price to pay for more easily accessible free and open data.
Apple Maps now shows the borders of Indigenous land in Australia and New Zealand.
Apple Newsroom:
Beginning today, Apple Maps now displays Indigenous lands in Australia and Aotearoa New Zealand. By gathering information from Indigenous advisors, cartographers, Traditional Owners, language holders, and community members, Apple Maps will show reserves and Indigenous Protected Areas, Indigenous place names, Traditional Country, and dual-language labels. Indigenous lands place cards feature information about the local area and Traditional Owners, and can be curated to allow communities to add their own photos, destinations on their land, and text in their own language. Representation of Indigenous lands in Apple Maps provides users with a more comprehensive experience while also recognising the stories and significance behind them.
Apple Maps will now include more than 250 dual placenames for cities and towns across the country, with more to be added.
I’m struggling to confirm this. On the latest versions of both macOS (15.3.2) and iOS (18.3.2), no traditional place names are shown instead or in addition to the English names. It’s technically feasible to show two names simultaneously for a location, as we’ve seen with Gulf-of-Mexico travesty—it just hasn’t been done.
When I search for “Naarm,” the traditional name of the Melbourne area, only “Melbourne” shows up as a match. Likewise, a search for “Gadigal” or “Cadigal” returns matches in the Sydney area, like “Gadigal Station,” but not Sydney itself.
I love a good deep dive into often-overlooked map-related challenges. Here Hanbyul Jo wonders why Mapbox chose to vertically display labels in Hangul, whether the choice improves legibility, and since when Koreans horizontally read and write Hangul instead of vertically.
Meta, one of the founding members of Overture Maps Foundation, has successfully transitioned its suite of global basemaps used across apps such as Facebook and Instagram to Overture’s base data layers
The goal was to build an up-to-date, validated, global basemap using OpenStreetMap that could power all of Meta’s use cases. Daylight included validation checks designed to find and correct mapping errors, building footprint detections, lidar derived building heights, name translations, and a global land cover layer. This global dataset was made publicly available and has served the maps at Meta for the past five years.
As a founding member of Overture, Meta has been deeply involved in developing the processes that produce Overture’s published data. In fact, the very same validation processes and pipelines that were used in Daylight are also now used to produce Overture’s regular data releases.
Notice the past tense. There is no official announcement confirming Daylight’s end of life. But there hasn’t been an update since November 2024 after more than four years of at least twice-monthly releases.
If you are a student or early-career developer and want to make your mark in open-source geospatial, the following geospatial organisations are participating in the 2024 Google Summer of Code:
The Protomaps project, a set of protocols and software for serving fast map tiles over the Web, is transitioning to a new funding model.
Previously, Protomaps’ main source of funding was selling one-time downloads of basemap tiles. Now you can purchase access to commercial, hosted tile APIs and hands-on support through the project’s GitHub sponsorship page:
For $14, you get one million tile requests via the commercial API. The tileset is derived from the Daylight Map Distribution, which includes data from OpenStreetMap but has gone through additional quality checks.
For $140, you get fifty million tile requests via the API, plus an optional downloadable tileset in case you’re looking to host the tiles more cost-efficiently should you exceed the allowance.
Then there are two access tiers for $2,000 and $4,000, which give you access to Protomaps developers to provide support and consulting services.
In addition, all components needed to produce PMTiles and run the Protomaps infrastructure are open source; you can run the APIs yourself for the infrastructure cost only.
Developing open-source projects and financing them by selling professional services for development, maintenance and knowledge transfer is an idea some German geospatial businesses, like Terrestris or Wheregroup, have successfully implemented for years. But I haven’t seen it yet for a company selling map tiles.
Development Seed’s data team in Ayacucho, Peru has spun out to form GeoCompas:
I am pleased to announce that the skilled data and annotation team from DevSeed is separating to become an independent employee-owned company called GeoCompas. The Geo AI practice at Development Seed has grown, and the GeoCompas team has been essential in providing crucial mapping, annotation, QA, and automation support. Their expertise in geodata is now available to everyone.
The new company focusses on OpenStreetMap editing, data labelling and annotation for AI projects, but also development of data processing pipelines and web services.