I spent two years mapping the areas surrounding one of the most beautiful drives in the world. Each morning, I looked forward to discovering and mapping an area adjacent to a one I did the day before. This diary entry serves as an introduction to a series of texts covering the challenges I encountered and how I overcame them. Included below are a few snapshots taken in the field, which capture the majestic landscapes along the Icefields Parkway. I have also added a “before and after” map excerpt, as well as a short video, to illustrate how my work gradually took shape over time.

The Place

The Icefields Parkway (Highway 93N), in Alberta, owes its name to the spectacular icefields and glaciers that line the road. The parkway, which is a 232 km stretch of highway cutting through Banff and Jasper National Parks, attracts between 1.2 million and 2 million visitors per year. The parkway offers exceptional alpine landscape, with glaciers, rivers, turquoise lakes, and breathtaking mountain views directly from the roadside (Figure 1). Best of all, many of these landmarks can be reached in a single day’s hike (Figure 2).

… See full entry

Spent the morning with the Bicycle wiki page and a 300-post thread about access=dismount, and the same thing kept coming up: our cycling data is really good at describing the road, and a bit less good at describing the person on the bike.

We tag the surface, the separation, the access. But the thing that decides whether I actually ride a route is rain, fatigue, and whether I trust the stretch where the cycle lane ends and the traffic doesn’t slow down. None of that is in the data, and it’s the part that matters on a Tuesday morning when it’s pouring.

I keep coming back to the idea that the gap between “legal to ride” and “would actually ride” is where the interesting mapping is. Maybe that’s a project for a quieter evening.

DinoData is a web-based tag editor for OpenStreetMap nodes, ways and relations, designed for phones first. Open an object, see what it is and what each tag means in your language, edit the tags with inputs suited to them, and save as a changeset. Geometry is not edited.

It runs entirely in the browser and is hosted on Wikimedia Toolforge. It started as a Wikidata editor; the OpenStreetMap mode lives next to it in the same app.

Finding an object

Search for a place, or paste what you have: an openstreetmap.org URL (including edit and API URLs), an iD or JOSM remote control link, a short ID like n123 or way 123, or coordinates. Other text is a place search with Photon and Nominatim. “Near me” lists the tagged objects around you via the Overpass API, handy while surveying.

Understanding the tags

• What the object is (“Cafe”), from the translated presets of the iD tagging schema
• Every tag explained from the wiki’s data items: description in your language, status, image, links to the wiki page and the Wikidata concept
• Deprecated and discouraged tags marked
• A small map, the parent ways and relations, and the history: tags added, changed and removed per version, with user and changeset comment

Editing

• Preset fields with matching inputs: chips for yes/no and other choices, a value picker with translated options and the most used values from taginfo, chips for multiple values like cuisine=pizza;pasta, toggles for diet:*, names per language, a search for wikidata=*, and number, phone and URL keyboards
• Issues with one-tap fixes: deprecated tags, values not matching the format documented on the wiki, tags not meant for the element type, incompatible tags
• Undo per field or tag; pending edits survive reloads and the login

… See full entry

OSM formats compared with TOML

OpenStreetMap data comes in several text formats. OSM XML is exact but verbose, OSM JSON has no comments, and Level0L is easy to type but has no grammar beyond its parser. TOML has comments, a spec and parsers in most languages, including Python’s standard library. So what would OSM data look like as TOML? I came across this question while adding an OSM tag editor to DinoData. Below, each format is compared with TOML using the examples from the OSM wiki, and the TOML versions are checked with Python’s tomllib.

Level0L is the text format of the Level0 editor. DinoData reads and writes it in the OSM text view (src/osm/level0.ts, src/osm/tags.ts). It looks a bit like TOML: key = value lines, # comments, and headers that start a section. But the two formats are not compatible. This page compares them, together with the other formats DinoData meets, and shows how OSM data could look as TOML.

• Level0L (lines of text): editing in Level0. OSM API: –. DinoData: text view, read and write.
• JOSM file format (OSM XML with action, <bounds>, upload): saving edits in JOSM. OSM API: plain OSM XML is the default response, and the body of PUT for single elements. DinoData: –.
• OSM JSON (JSON): reading. OSM API: elements, history, full, map and more, with a .json suffix or Accept: application/json. DinoData: reading elements, history and changesets.
• osmChange (XML with <create>, <modify>, <delete>): changes. OSM API: GET /api/0.6/changeset/#id/download, and the only format of the upload POST /api/0.6/changeset/#id/upload. DinoData: saving tags.

The Level0 behaviour described here follows its parser, www/level0l.php.

Differences

Section header

… See full entry

Written 2025-10-30… Published 2026-11-03

About three years ago, during my day-to-day mapping activities, I came across a golf course that had some obvious errors. I didn’t think much of it at the time. Just a fairway that was drawn right across the top of a green. But the more golf courses I looked at, the more frequently I encountered this error.

I soon learned about a golf course simulator video game that was built on OpenStreetMap data and was attracting a large influx of simulator enthusiasts who weren’t seasoned OSM mappers. Along with that came various tools to aid in converting OSM data into these simulators, as well as tutorial videos that made some very bad assumptions about how golf courses should be mapped. These factors set the stage for some very bad mapping practices to spread around the world.

My understanding is that attempts were made to correct these bad habits at the source, but the tutorial author wasn’t interested in revising or removing his video. Combine that with YouTube’s bad decision to hide down votes on videos, and you’re left with de facto training material that is widely disliked but that new golf course mappers have no way of knowing is a poor source of truth.

So this is where I come on the scene and try to make the map better.

I’m a big fan of the broken windows theory. I’ve seen it play out in my own town. A minimally maintained but abandoned pizza parlor near me stood for years waiting for a new occupant to lease it. You’d see prospective tenants touring it every once in a while, but no takers. Eventually, someone tagged the building with graffiti. It wasn’t cleaned up in a timely manner, and things spiraled downward from there: more graffiti, broken windows, stolen copper, garbage left behind by homeless people, and eventually, the building was razed.

… See full entry

I’ve been building Wandering Around, a free browser game in which you can walk around any town or city on Earth. The whole world comes from OpenStreetMap, so I’d like to share how it uses OSM data and ask this community for feedback.

Play it here: https://wanderingaround.games (works on phone and desktop, no account needed) Screenshots and a short video: [link naar je itch.io-pagina]

How the map is built

• Streets, buildings, water, parks and bridges come from OpenFreeMap vector tiles (zoom 14, OpenMapTiles schema). The game fetches the tiles around the player, decodes them in the browser, stitches buildings that are split across tile edges and turns everything back into OSM-style features with tags. The game logic works on those tags, so it is the same code whether the data comes from tiles or (as a fallback) from Overpass.
• 2D and 3D: the same data drives a 2D map and a 3D eye-level view. Building heights come from OSM where they are tagged.
• Transit: train routes and stations, city buses, ferries and signposts come from Overpass queries, so you can travel between real stops and stations.
• Country map: towns, airports and ferry ports within 100 km come from pre-built tiles made from Overture, OurAirports and Natural Earth.

What the game adds itself

Vector tiles leave out some details, such as single trees, benches, crossings, traffic lights and roof shapes. The game fills those in itself so streets don’t look empty. These are decoration, not data: they are never written back anywhere.

Wikidata postcards

Every feature with a wikidata tag (and, in the Netherlands, ref:rce for national monuments) becomes a landmark. When you visit one you get a postcard with:

• a photo from Wikimedia Commons (P18), with the photographer and licence shown,
• year built (P571), height (P2048) and floors (P1101),
• a summary from Wikipedia in your language.

… See full entry

While walking around, surveying, I’ve often theorised it would be possible to work out where I lived based on the centre of all the edits I’ve made.

So I made a tool to estimate it. And it’s pretty close.

It gets closer or further away depending on how much history you look at.

But you might find it informative.

https://yamatt.github.io/osm-user-home/

Edit:

It’s an interesting problem.

OSM needs and wants contributions. But by it’s very nature it can give away users data, without them even realising.

I don’t think there is even a technical fix for that.

But there does need to be a serious discussion with contributors about their data privacy.

Some might not care.

Some might be alarmed.

And alarmed is good.

We need to talk about how we balance users privacy and the power of OSM contributions.

X marks the spot to the crossroads of America. Route 30 travels east to west and Route 41 north to south spanning the whole country. Schererville, Indiana

A fantastic guide. But my god, could it make the changeseys a bit more bulky instead of creating a new one for every single edit?

I feel like I’m spamming now…

I had a great time at SotM LATAM in Mexico City this past week. I got to hear so many talks on diverse subjects, I met friendly and interesting people, and there was a general acceptance that a random person from Canada would be there. Kudos to the organizing team and volunteers who put on a smooth, well run event.

I have never been to a State of the Map before despite thinking about going since way back in 2016 when I happened to already have a visa for Brazil. That certainly would’ve been harder with my level of Spanish understanding then. I will definitely be back.

Gracias a la comunidad y hasta la próxima. Espero encontrar oportunidades para colaborar.

Surprised myself and participated for rrrreal! (well, ok, missed the first day)

Very nice event, liked it a lot.

Report 🫡

https://m-d.me/notes/osm-dailychallenge-2026/

31DayOSMChallenge #Maptober #Maptober2026

Registered in 2014, but I only started using OpenStreetMap actively last year.

After watching some Aboriginal movies of Australia, I switched from the Himalayas to Down Under, and then with chasing tracks in the outback of Australia—uncovering hidden reservoirs, basins, homesteads, cattle stations, farms, and surprisingly, some generous peaks. From there, I switched over to China, where the clever use of land through greenhouse horticulture stands out as one of the best examples of resourcefulness.

Flipping to Ethiopia brings a blend back to Nature: homes and huts constructed from natural elements offer a grounding perspective, even as the region slowly turns toward the so-called “blinded modernization.” Crossing over to South America, dense forest covers reveal hidden streams and alpine lakes tucked high in the mountain ranges, though the sheer scale of deforestation is disheartening to see.

Jumping to Japan and the Korean peninsula, the structured, disciplined approach to farming is among the finest in the world, matched by remarkable urban and human development.

Further north, the vast expanses of Alaska and Siberia ground you completely—reminding us of our fragile place within Nature and how quickly our so-called "development" will simply be reclaimed by time.

OSM has helped keep me calm, though it is still easy to get agitated. Now, I’ve even started mapping via my mobile browser as well, which has been a lifesaver while traveling on public transport. In a world full of towering billboards and aggressive roadside slogans, mapping on the move gives me the perfect excuse to ignore the visual pollution.

Of course, whenever I'm passing through true valleys and breathtaking landscapes, the phone goes straight into my backpack so I can take it all in.

Title: OpenHVRA: a free hazard, vulnerability and risk assessment tool built on OpenStreetMap

---

Hello mappers!

For years I have watched the same thing happen. A flood or any hazard hits a place, and the first questions are always the same: how many people live near the river, which schools and health posts are in the way, and what will it cost? Very often the best data to answer those questions is already sitting in OpenStreetMap, mapped by volunteers. But turning those buildings and waterways into a risk picture usually needs desktop GIS, paid software, or a specialist.

So I built a tool based on my professional experience, OpenHVRA (Open Hazard, Vulnerability and Risk Assessment) with help of my friend Claude, a free web map that does that analysis straight in the browser, using only open data.

Try it: gomalhunzai.github.io/openhvra Code (MIT License): github.com/gomalhunzai/openhvra

No login, no install.

How it works

… See full entry

Motion extraction tutorial: https://youtu.be/NSS6yAMZF78?t=226

This mapping was a while ago. This is also a diary entry repost, I figured out how to display GIF-files. I was using imgur before this, where all GIFs you upload are converted to mp4 automatically so I can’t display it in the diary. Imgur is also full of ads. I assume most people don’t use adblockers, even with adblockers the UI distracts from your file. I’ll go back through my imgur links on OSM and reddit and link them instead to images on my little github page.

OSM LATAM 2026 México 🇲🇽 — Day 3: Beyond the Map!

And just like that, we reached the final day of OSM LATAM 2026 México! 🚀

What an incredible journey these three days have been. Day 3 brought another packed schedule of inspiring talks, practical experiences, new ideas, and, above all, more opportunities to connect with the amazing people behind OpenStreetMap.

🔗 Event website: https://www.osmlatam.org/sotm/2026-en/

📅 Full schedule: https://pretalx.cienciaenmexico.com/sotmlatam2026/schedule/

🌎 Learning from the global community

The day featured keynote talks from the Humanitarian OpenStreetMap Team (HOT) and OpenStreetMap US (OSM-US), alongside a rich program of presentations and discussions.

Throughout the day, we enjoyed a wide variety of talks covering practically everything OSM: mapping, routing, tagging, data quality, community engagement, and much more.

We also learned about MapRoulette challenges for local governments, an interesting way to encourage collaborative mapping and help improve geographic data in areas of particular interest.

❤️ Mapping with and for communities

Perhaps the most meaningful part of the day was hearing about projects that put communities at the center of mapping.

Teams shared their experiences working with vulnerable communities across Latin America, showing how OpenStreetMap can help local governments make informed decisions and prioritize actions where they are needed most.

Another inspiring project introduced elementary school children to mapping through playful activities, using playdough and stickers on top of OSM maps.

Particularly remarkable, these initiatives reached remote communities, some hours away by car from the nearest place with telecommunications access!

… See full entry

OSM LATAM 2026 México 🇲🇽 — Day 2: Transmute the Map!

Day 2 of OSM LATAM 2026 México brought us another incredible day of learning, collaboration, and community!

🔗 Event website: https://www.osmlatam.org/sotm/2026-en/

📅 Full schedule: https://pretalx.cienciaenmexico.com/sotmlatam2026/schedule/

As every great adventure, the day had its ups and downs. We ended up with two breaks instead of one, but what could have been a downside turned into another opportunity to meet fellow mappers, exchange ideas, and strengthen the connections that make our community so special.

🌎 Collaboration over competition

One of the highlights was hearing keynote talks from sponsors HERE and Overture Maps, learning more about who they are, how they work, and how they collaborate with OpenStreetMap to improve geospatial data.

Sometimes, companies working in the mapping world can be perceived as rivals to OSM. But today’s conversations remind us that the geospatial ecosystem is much bigger than any single project or organization. There are meaningful ways to collaborate, share knowledge, and support open data. Different approaches can complement one another, and the entire community can benefit.

We also had a special talk from Daniela Waltersdorfer, an OpenStreetMap Foundation (OSMF) Board Director, about the Foundation’s role in the OSM ecosystem.

A particularly important takeaway was that the OSMF’s mission is to support OpenStreetMap, not control it. The Foundation helps keep our community healthy and growing while providing the organizational and technological backbone needed to sustain the project for the long term. The map belongs to its community, and that distinction matters!

🗺️ Mapping that makes a difference

Throughout the day, we enjoyed amazing talks exploring how mapping helps local communities, how public transport mapping can improve mobility, and why open geospatial data matters.

… See full entry

1) In Vespucci, create a custom layer named “Strava-cookies” with the following URL: https://content-a.strava.com/identified/globalheat/all/hot/{z}/{x}/{y}.png?px=256.

2) Next, you need to retrieve cookies from the Strava website. To do this, install the JOSM Strava Heatmap browser extension on a PC. Use this extension to obtain the cookies.

3) Create an XML file with the following content:

<?xml version='1.0' encoding='UTF-8' ?>
<Configuration>
    <Section name="tilelayers" table="headers" 
version="0.1" truncate="false">
        <ColumnTypes id="TEXT" name="TEXT" 
value="TEXT" />
        <PrimaryKeyColumns />
        <Row id="Stravacookies" name="Cookie" 
value="***"/>
    </Section>
</Configuration>

Replace *** with the cookies you retrieved.

4) In Vespucci, go to “Tools” and select “Import configuration.” Locate the file you created.

5) Enable the “Strava-cookies” layer.

Alternative method (without using a PC): Open the link mentioned in step 2 using the Firefox mobile browser. Follow the link provided there for Firefox. Enable “Desktop site” mode in the browser, install the extension, and then proceed as described above.

Cookie values change every few days; therefore, if the layer stops working, you will need to update the XML file and re-import it.

OSM LATAM 2026 México 🇲🇽 — Day 1: Connecting, Learning, and Mapping Together!

I had the wonderful opportunity to attend OSM LATAM 2026 México, and what an amazing first day it was!

It was a pleasure to meet so many incredible mappers, connect with fellow community members, and share knowledge, experiences, and ideas around OpenStreetMap. I also enjoyed participating in community talks and learning about the new technologies and tools being developed to further support OSM and its community.

One of my personal highlights of the day was discovering Maprizon, a state-of-the-art street-level imagery platform from Kaart, and fAIr, a collection of AI-assisted mapping tools from HOT. It was exciting to learn about these developments and explore how emerging technologies can help us improve our mapping workflows and contribute even more effectively to OpenStreetMap.

Events like this remind us that OpenStreetMap is much more than a map. It is a global project powered by people, collaboration, shared knowledge, and a common passion for open geographic data.

A heartfelt thank you to all the sponsors, allies, organizing committees, speakers, volunteers, local governments, venue staff, and all the amazing people attending and making this event possible.

It has been an inspiring first day, and we’re already looking forward to continuing tomorrow, learning more, exchanging ideas, and honing our mapping skills together!

OSM LATAM 2026 México! 🗺️❤️

 #OpenStreetMap #OSMLATAM #OSMLATAM2026 #Maprizon #fAIr #MappingTogether

I made Walkably, a free walkability map for 31 European countries.

Streets, paths and places come from OpenStreetMap contributors (ODbL). Every street gets a score from 0 to 100.

How the score uses OpenStreetMap

• Reach is 60% of the score: the real walk on streets and paths to eight daily needs, such as groceries, parks and public transport. A shorter walk scores more, and the second and third option nearby count too.
• Street quality is 40%: traffic, busy roads nearby, summer shade and slope.

… See full entry

I lost my wife because of cancer on 22 September. Half of my soul and motivation also lost with her..