OpenStreetMap
Diary Entries in English
Recent diary entries
The world of OpenStreetMap…
One person is sanding down a rusty gate,
Another is painting over the rust,
Yet another is attaching a new handle to the corroded gate,
Some are even building an entirely new gate to replace the old, rusted one…
And a few are wondering, “Why are they doing all this? The gate still works just fine…“
Everyone is doing what they want to do and what they can do,
but since no one is coordinating or adjusting things,
it’s a situation where people are working inconsistently and in a disorganized manner.
오픈스트리트맵 세계…
어떤 사람은 녹슨 대문을 사포질하고 있고
또 다른 사람은 녹이 슬어 있는 대문에 페인트칠을 하고 있고
또다른 사람은 녹이 슨 대문에 새로운 손잡이를 달고 있고
또 어떤 사람들은 녹이 슬어 낡은 대문을 교체하려고 새 대문을 만들고 있고…
그리고 몇몇 사람은 대문이 아직도 쓸만 한데 왜들 저러고 있나 의아해 하고 있고……
각자 제 하고픈 일을 하고 할 수 있는 일을 하지만,
누구도 조정해 주거나 조율해 주지 않아서
서로 일관성 없고 뒤죽박죽으로 일하고 있는 형국.
EN translation below
PT abaixo
En 2025, il existe encore plein d’endroits dans OSM où les villes sont encore peu cartographiées dans le détail – malheureusement pour l’exhaustivité de la base, mais heureusement pour la motivation des cartographes OSM. Il est possible de changer radicalement la cartographie d’une petite ville en y consacrant quelques heures lors d’un séjour de quelques jours.
Par le passé, lors d’une formation en ligne, j’avais montré comment j’avais cartographié peu à peu quelques centaines de POI sur un quartier très vivant de ma ville de résidence, et cela avait motivé un étudiant pour faire la même chose chez lui (voir ce billet).
À nouveau, je relate mon expérience pour inciter d’autres à l’adapter à leur manière. Mais j’irai cette fois au-delà des aspects techniques et logistiques. En effet, cartographier des POI, c’est bien sûr utile à toutes celles et ceux qui réutiliseront la donnée, c’est un bon moment à passer seul ou à plusieurs, mais cela permet également de découvrir le lieu autrement, de mieux le connaître et le comprendre.
Au-delà des POI, en portant un œil de géographe, on se rend compte comment la localité s’organise et comment elle évolue. Je rajoute donc quelques petites pastilles sur ce que les POI m’ont appris sur le territoire cartographié.
Le contexte
J’ai passé trois jours dans les environs de Mucugê, petite ville de la Chapada Diamantina dans l’intérieure de l’État de Bahia au Brésil, dont l’histoire est justement marquée par la période de ruée vers le diamant, avec avant tout un agenda touristique, mais aussi en tête la possibilité de cartographier en fin de journée, au retour de visites ou de randonnée, ou au contraire tôt le matin. Mais dans les deux cas, seulement si j’avais suffisamment d’énergie pour le faire et sans que cela n’interfère avec les activités prévues ou improvisées avec les personnes qui m’accompagnaient.
It sometimes feels like OSM is of the West, by the West, for the West.
An OpenStreetMap of the West, by the West, for the West.
Over the past 20 years, OpenStreetMap has become one of the world’s largest and most impactful crowdsourcing projects. Thousands of communities have formed around collaborative mapping, connecting people and fostering the sharing of geographic knowledge.
Today, OpenStreetMap is an essential source of geographic data, powering a wide range of map services used by thousands of people daily and serving as a vital tool in managing climate and humanitarian crises. Although the main site, OSM.org, is not designed as a direct navigation tool, the experience of new users plays a crucial role in attracting new mappers. Improving usability for the public, therefore, benefits the entire community, alongside the millions of organizations that use OSM data.
In recent years, the OpenStreetMap ecosystem has grown significantly, expanding to every corner of the world and yet, the user interface of OSM.org is much the same as it was when OSM was founded in 2004. Among other challenges, it is not optimized for mobile devices, nor does it meet modern accessibility standards. We believe this is the time to prioritize a more cohesive, intuitive, and enjoyable user experience while making it easier for developers and contributors to create solutions.
To achieve this goal, we are developing a design system, a set of reusable guidelines and components that unify the design and functionality of digital interfaces. This system has been designed to meet the community’s needs, promoting consistency, accessibility, and scalability. It will ensure that interfaces are inclusive, modern, and capable of addressing the needs of a diverse global community.
Discover the Atlas
After one more year of French administration funding (thank you!), we are so proud to have released versions 3.X of uMap since last April. Since then, we made a couple of adjustments to ease the deployment of that new version.
The main feature of that version is the long awaited real-time collaboration! Thanks to Alexis for the development and NLnet for the funding 👏
We decided for now not to activate the real-time collaboration on the French server of OpenStreetMap for two reasons:
- The instance is very big and we want to encourage smaller instances
- We preferred to first deploy it on smaller instances to test the load
It’s activated on these instances:
Map templates
We also introduced the template feature which allows to avoid copy-pasting maps to reproduce similar styles and (optionally) data. It also allows to share templates with others (useful for friends, workshops and so on).
Cet article de blog a été publié à l’origine sur ce site web et a déjà été mentionné par WeeklyOSM.
EN translation below
Renforcement de l’approche méthodologique
Dans le billet précédent consacré à la cartographie des dommages du bâti à Mayotte, j’annonçais vouloir comparer ces dommages évalués dans OSM à partir d’imageries Pléiades post-désastre avec d’autres jeux de données et de promouvoir la réplication de l’exercice cartographique, ce qui implique de présenter la démarche et ses résultats.
Dans cette double perspective, je me suis rendu compte qu’il serait utile de disposer de statistiques plus détaillées que celles montrées dans le tableau de bord produit jusqu’à présent en intégrant à la fois un nouvel indicateur, un rapprochement méthodologique et une autre échelle territoriale. Dans la base PostgreSQL de l’IFL consacrée à Mayotte, j’ai ainsi décidé de :
- calculer la fraction des bâtiments évaluables sur les imageries Pléiades post-désastre, c’est-à-dire ceux qui sont visibles sur au moins l’une d’entre elles, ce qui exclut ceux qui sont malheureusement toujours masqués par la couverture nuageuse
- considérer les bâtiments OSM sur les zones d’évaluation Copernicus qui sont sans intersection avec un point EMSR780 comme étant non endommagés, afin de pouvoir faire une comparaison directe classe à classe avec la méthodologie BAR adaptée à OpenStreetMap. Ce n’est certes pas toujours vrai, lorsque les points EMR780 sont décentrés au point de ne pas intersecter le bâtiment OSM qui leur correspond, mais ces cas sont rares.
- désagréger les indicateurs au niveau des 17 communes de Mayotte, afin de visualiser l’hétérogénéité des dommages sur le territoire
De nouveaux indicateurs
Today is a special day—my birthday. I woke up feeling grateful and hopeful. Another year of life is not something to take for granted. I feel happy, not just because it’s a celebration, but because I’ve come a long way, and I’m ready for more. Birthdays are not just about cake, candles, or greetings. They are about reflection, growth, and new beginnings.
As I look ahead, I’m deeply motivated to grow—not just in age, but in wisdom, character, and capacity. I want to build myself professionally and personally. My journey so far has been full of lessons, and I know there is much more to learn. I believe in continuous improvement, and that belief fuels my ambition to keep pushing forward.
This year, my focus is on professional development—gaining new skills, strengthening what I already know, and becoming more confident in my work. I plan to seek opportunities for learning, whether through training, mentorship, or real-life experience. I want to become more capable, resourceful, and impactful in everything I do.
At the same time, I also want to grow emotionally and mentally. Maturity isn’t just about work—it’s also about how we treat people, how we handle challenges, and how we understand ourselves.
So, on this birthday, I’m not just celebrating the past—I’m preparing for the future. I feel proud of the person I am becoming. I will keep dreaming, working hard, and building the life I envision.
Here’s to a new year of growth, purpose, and becoming the best version of myself.
The situation
This is something I encountered mapping bike infrastructure in the Austrian town of Neuhofen an der Krems: what one might assume is a continuation of a shared bike/pedestrian path over a pretty unimportant side road is actually a legal trap for cyclists and people walking. View it on Mapillary and look closely: a sign indicates that the cycling/walking path ends, only for it to continue right after the crossing. This means that pedestrians and cyclists no longer have right of way and have to yield to cars coming out of the residential road.
A safety nightmare
It is unfortunately a bit hard to make out using imagery, but the sight lines at this crossing are horrible. Not only is it basically impossible for cars from the south to see anyone cycling by before it’s too late, it’s also pretty much impossible to see any cars coming as a cyclist, even when slowing down. This is exactly the sort of situation one should avoid as a planner, but it is strikingly obvious that the bike path here was just an afterthought.
Legal trouble
Intersections like this are common all over (Upper) Austria and often “solved” by applying this “interrupted cycle path” treatment. This means that if a cyclist and car were to collide here, the cyclist (person far more likely to die) would have to be the one explaining why they didn’t yield, and not the car driver (person driving a multiple-ton death machine) - just adding insult to (literal) injury. This is especially bitter because the bike path connects a fairly sizeable (by Austrian standards) part of town with the town center, while the side road only serves a handful of houses.
What should be done
My journey as an OpenStreetMap contributor began in 2022, with a humble yet impactful project: mapping roads in Mugu, Humla, and Jumla—three of Nepal’s most remote and mountainous districts. Since then, I’ve grown into an active mapper, dedicated validator, and proud member of the global OSM community. In 2023, I was recognized as an OSM Guru and listed among the top contributors, a reflection of my deep passion and consistency in open geodata contribution.
Areas of Focus My primary interest lies in mapping buildings and roadways, especially in critical and under-mapped areas. I’ve also actively contributed to tools like MapRoulette and MapSwipe, which help bring micro-edits and mobile contributions to the mapping ecosystem.
Highlighted Contributions “Map Roads, Make Your Way” Project Recognized as a top-quality mapper, I contributed significantly to this project—led by HOT, Open Asia Pacific Hub, KIRDARC, and OSM Nepal—focusing on mapping roads in Humla, Jumla, and Dolpa. This initiative was a pivotal moment for me, helping improve accessibility in some of the most remote regions of Nepal.
Digital and Spatial Technologies for Anticipatory Action Volunteered in this 5-day event organized by NAXA Nepal, where we mapped roads, waterways, buildings, and land use across six municipalities. Using ESRI imagery on the HOT Tasking Manager, we enhanced maps to support community resilience and disaster preparedness. We focused on: Mapping open spaces critical for evacuation and aid Digitizing health posts, schools, power lines Improving road networks and land use classifications
Global Solidarity Through Mapping Solidarity Mapathon for Myanmar (HOT) Climate Change Challenge with HOT, TOMTOM, Open Asia Hub Participated in multiple Kathmandu University mapathons—both as a mapper and organizer, including events under NEPGEOM
Testing, Heading
Subheading, Subheading
- List
- List continued
Hello World!
I just completed my first university GIS course and I absolutely love mapping. The course enabled me to contribute to the Humanitarian OpenStreetMap team. This experience is excellent and I am thankful to sharpen my digitizing skills while helping others.
This Diary feature is new to me and I’m not sure what it does. Good to experiment with! :)
When I first received the email notifying me of my selection for the YouthMappers Leadership Fellowship 2024 in Thailand, it was a surreal moment. I was doing my assignments all frustrated but suddenly when the notification popped up, and I couldn’t contain my emotions. I jumped with joy, overwhelmed by the thought of being chosen for such an incredible opportunity. That email was the start of an unforgettable journey—one that would take me to new places, connect me with passionate individuals, and leave me with memories I’ll cherish forever. The dream of the dreamer started since then. The preparation for the fellowship began months in advance. YouthMappers, a global community of students, researchers, educators, and scholars that use public geospatial technologies to highlight and directly address development and environmental challenges worldwide ensured we were well-prepared with pre-departure sessions and constant communication through emails and WhatsApp. They guided us every step of the way, taking care of us with unmatched warmth and care. From learning how to say “Sawadike” (hello) and “Khapunka” (thank you) in Thai, to planning and packing for the trip, every moment was filled with excitement. The day of departure was momentous—my first international flight, passport in hand, and butterflies in my stomach. Along with my Nepali peers, we clicked countless photos at the airport, thrilled to embark on this journey. As our flight landed in Bangkok, the sparkling city lights welcomed us to a country that truly never sleeps. Angela, our warm and wonderful guide, greeted us at the airport, setting the tone for an amazing experience.
First Impressions and New Connections
If anyone who want to edit Indonesian maps, I recommend using BHUMI ATR/BPN aerial imagery data as source. You may see the links from my last 2 months of renewed editing career of OpenStreetMap.
The primary reason I use BHUMI is their data quality is crisper and detail than Bing/Esri data (but as crisp as OpenAerialMap data), more extensive (OAM has limited coverage; in Indonesian context, their coverage is tinier than a grain of rice) and as a work of Ministry of Agrarian and Spatial Planning (Kementerian Agraria & Tata Ruang/Badan Pertanahan Negara), based from open data and their BHUMI website does not mention the copyright, it is in public domain (art. 43 point B of 2014 Copyright law/UU No. 28 Tahun 2024).
(You may object the last one, as I judge it solely on their website’s statement and my limited understanding to Copyright Law. In their website, the term of use is only limited to general responsibility using BHUMI ATR/BPN data. If you consider their data as copyrighted, that means 2 months’ worth of cleanup. Not an easy task, I hope don’t let it happen!)
However, some of their data might be not up to date or broken, so please cross-check with less than detail Bing/Esri data and zooming in and out, because you could get detail out of it.
I have big wish about this: ATR/BPN keep making high-quality imagery, and Indonesian OAM contributors to do the same, not just waking up when they do a mapathon!
Load testing vector tiles
As part of bringing the new vector tile servers into production, I had to benchmark their performance. Since there’s a cache in front of the servers, it’s challenging to benchmark them accurately. Although we’ve never had a heavy load on the vector tile servers, we’ve been running raster tile servers for years.
All tile requests on the standard layer are logged, and from those logs, I can generate a list of tiles to benchmark the vector tile servers. The logs are stored as Parquet files, which I query using Amazon Athena, a hosted Presto database.
Vector tiles and raster tiles typically have different scales at the same zoom level. To convert raster tile requests to equivalent vector tile requests, I divide the x and y coordinates by 2 and decrease the zoom level by 1. I also skip zoom 0 raster tile requests to simplify the process, as these don’t affect performance since zoom 0 is always cached.
The OSMF shortbread tiles have a maximum zoom of 14. Lower scales (higher zoom levels) are achieved by overzooming on the client side. Requests from zoom 1 to 15 should have their zoom level lowered by 1. Requests from zoom 16 to 19 need their zoom level decreased by the difference between their level and 14. I divide the x and y coordinates by 2 the appropriate number of times to match the new zoom level.
Filtering to have only cache misses gets me a request list on the backend servers.
SELECT
CASE WHEN z > 15 THEN 14 ELSE z - 1 END AS v_z,
bitwise_right_shift(x, CASE WHEN z > 15 THEN z-14 ELSE 1 END) AS v_x,
bitwise_right_shift(x, CASE WHEN z > 15 THEN z-14 ELSE 1 END) AS v_y
z, x, y
FROM fastly_success_logs_v1
WHERE year=2025 AND month=5 AND day = 1 AND hour = 1
AND z >= 1
AND cachehit = 'MISS';
Unfortunately, this is the wrong list.
I’ve undertaken the task of mapping Inveralmond Industrial Estate, which lies close to my location but is currently not mapped in any great detail. While most buildings were drawn onto the map fifteen years ago, only about 25 businesses have been added since then—and I contributed half a dozen of those entries just a few months ago. At least two businesses that were added years ago have since closed permanently. Last Sunday, I conducted a preliminary survey of the area, capturing dashcam footage that helped me compile a list of fifty additional unmapped businesses. This number will likely grow substantially, as I left out the minor roads in the estate during this initial pass. My estimate is that the final count could easily reach one hundred businesses. This pattern reflects a broader issue: industrial areas consistently suffer from poor mapping coverage. That’s my take on it anyway from the areas I’ve looked at. I’m currently testing the iOS app Every Door, and plan to conduct comprehensive field mapping in the coming weeks. My goal is to document everything from the largest manufacturing facilities down to the most modest infrastructure, including grit bins. osm.org/#map=16/56.41782/-3.47651
The Community Bonding Period officially wrapped up on June 1, and the coding phase is already off to a solid start.
During CBP, we successfully migrated the legacy 3DMR repo from GitLab to the new official GitHub repo. This cleanup made collaboration and discussions way easier. I also dove into pygltflib to handle glTF files, experimented with the obj2gltf converter.
I spent a good chunk of time mapping with the iD and JOSM editors, focusing on features from my home village. It’s oddly satisfying to see buildings you’ve walked past for years show up as polygons and tags. I plan to keep adding more local features whenever time allows. I experimented rendering both my mapped data and some iconic Indian landmarks using OSM2World, let’s just say it made my village look unexpectedly fancy.
With coding underway, my first priority was building out the test suite. The project didn’t have one before, which would’ve made every upgrade feel like defusing a bomb in the dark. I’ve opened a PR for it, currently under review. While working on this, I found a few minor bugs lurking quietly in the codebase for years, filed them, fixed them, and thankfully had those PRs merged.
The Django 5.2 upgrade is nearly done and is clinging on my forked repo as i am writting this… Most dependency updates and refactoring are complete. Now I’m exploring edge cases, making sure nothing’s hiding in the shadows ready to throw a 500 error when nobody’s looking. Once the test suite merges, we’ll be in a much safer spot to modernize the stack.
On the personal side: I’m a long-distance runner and have been using summer break to improve my endurance. I run 5 to 7 kilometers daily, a good counterbalance to all the screen time, long debugging sessions, and occasional existential dread that come with working through old code 🙃. I also have a few books lined up, both fiction and non-fiction, hoping to get through them—unless I get distracted again.
Recently, I needed to open my OpenStreetMap profile—just to right-click and save my own profile picture for use on another platform.
Thanks to the newly redesigned OSM profile layout, I was greeted by a few new statistics—one of which showed how many comments my last diary post had received. While I was busy grabbing my avatar, I couldn’t help but notice that my recent diary post had garnered quite a bit of discussion.
To my surprise, at least two commenters pointed out the same thing : they suggested it would be more intuitive if the value reflected actual months, rather than “something that roughly represents the progress of the year in base-10.”
That got me thinking—how hard would it be to convert that base-10 year-progress value into something closer to a conventional month (base-12)?
Step 1: Extract the Year
To compute the year from an OSM timestamp (Unix time), we start by offsetting it from a known reference point—specifically, the Unix timestamp for the start of the year 2000.
(osm_timestamp - 946692127) / 31556952
- 946692127 is the Unix timestamp for Sat Jan 01 2000 02:02:07 GMT+0000. This value was arbitrarily chosen by me (high accuracy isn’t necessary; I just needed a reference point roughly around the year 2000).
- 31556952 is the average number of seconds in a year (365.2425 days).
This gives us a floating-point number: the integer part is the year offset from 2000.
To extract the integer part (representing the year), we can substring the first 2 character:
substring(divided_by(osm_timestamp - 946692127, 31556952), 0, 2)
Oh wait!
While writing this post, I stumbled upon a small but interesting bug from the initial release.
Originally, the code extracted the first three characters of the computed year value. This worked fine for double-digit years like 2010 and beyond—10.5 would yield “10.”, which was sufficient for identifying the year and using the decimal as a makeshift separator between year and month.
OVERVIEW
I stumbled across a website which hosts every single Soviet military map 1:200k from 1985.
I plan to analyze each quadrant starting from the top left corner to the western side of russia, and map any unmapped settlements, buildings or other features.
I will update the post as time goes on, this is a VERY-long-term project.
QUADRANTS
CURRENTLY W.I.P.
- R60-05 to R6-35
PROGRESS SO FAR
-
R60-05 (~80%)
-
R60-11 (~30%)
STATS
Day 1
Changes - 167
Changes/day -167
Changesets - 2
Changesets/day - 2
Changes/changeset - 83,5
Quadrants - 1,1/3520 (3,13%)
Quadrants/day - 1,1
Km^2 - 2955.04/17 100 000 (0.017%)
DIARY
DAY 1
Added all the peaks, ridges, survey points which were on land in quadrants R60-05 and R60-011. Added capes, fixed lake names and added missing homes (how anyone can live there is beyond me). Tommorrow I will focus on adding rivers, the survey points along the rivers, adding the singular road and the rest of the capes
As OpenStreetMap becomes a central part of global mapping infrastructure, it’s worth asking:
Are corporations offering as much or more as they are getting from their OSM involvement?
Some corporations benefit enormously from OSM using it to build commercial services, run logistics, and analyze market data. But these same companies often have access to large, rich datasets, especially from telemetry and user behavior, that are far beyond what the average contributor can collect.
If shared, such data could dramatically improve OSM’s accuracy, completeness, and utility. But how much of this data is actually being shared?
There’s a tension here between open knowledge and proprietary advantage. While some companies do contribute back through edits, funding, or tools, the scale of these contributions doesn’t always match the value extracted.
This leads to deeper questions: - Is the current model of corporate engagement with OSM sustainable? - Are the benefits of the open model being eroded when sharing isn’t reciprocal? - Could new norms or frameworks encourage deeper, more equitable collaboration?
The OSM community thrives on openness, transparency, and volunteerism. As more private actors join, the long-term health of the ecosystem may depend on whether those values are upheld not just by individuals, but by corporations as well.