My Google Summer of Code 2026 project was Category Support in Nominatim.

The goal was to replace Nominatim’s single class/type representation with a proper categories field so that one OSM object can carry multiple categories. For eg, an object tagged as both a hotel and a restaurant can now be represented in one row with categories such as:

osm.tourism.hotel
osm.amenity.restaurant

What was completed

• Added the PostgreSQL ltree[] categories column to the main Nominatim tables.
• Generated multiple categories from OSM tags during import while keeping the existing class and type fields for compatibility.
• Updated ranking, triggers, migrations, and related database logic to use categories.
• Replaced the old place_classtype_* search tables with category-based search using indexed queries.
• Removed the old place_classtype_* table creation and maintenance code.
• Added category filters to the search API.
• Updated the SQLite adaptor and tests for category data.
• Updated the documentation for the category system and its API behaviour.

Pull requests

• #4106: Add ltree categories column for multi-tag OSM object support - merged
• #4146: Use ltree categories column instead of place_classtype tables - merged
• #4163: Remove the place_classtype tables - merged
• #4164: Add category filters to the search API - merged
• #4166: Update the documentation for the category series - documentation PR

Querying by category

The search API can now accept category filters. For example:

/search?q=restaurant+Berlin&include=osm.amenity.restaurant
/search?q=Berlin&include=osm.amenity
/search?q=hotel+Berlin&include=osm.tourism.hotel&include=osm.amenity.restaurant
/search?q=restaurant+Berlin&exclude=osm.amenity.fast_food

A parent category such as osm.amenity can match categories below it, so a query can include restaurants, cafes, bars, and other amenities without listing every child category separately.

Acknowledgements

… See full entry

I’ve just joined OpenStreetMap after using the CoMaps app for a month. I’ve been a frequent Google Maps Local Guide, but want to support open source efforts.

Today, I signed up and updated a shop that was labelled as disused to a cafe that recently opened local to me.

I need to vent for a minute.

Aligning aerial imagery terrifies me. I feel like no matter what I do, I will be off by a little bit. Whenever I come across a place where buildings are mapped and are a little bit off from the aerial imagery I’m using, I think to myself “I should probably align my imagery with those buildings,” but then I look at the primary road the buildings are right next to and realize that it is aligned with how my aerial imagery is right now. And then I see a cluster of features made 9+ years ago which is nowhere close to being aligned with my imagery and I question whether all of my contributions have been miles off of the correct location and if I’ve accidentally been vandalizing hundreds of places across Virginia without even knowing. But the worst feeling is when the same aerial imagery layer has its tiles slightly misaligned from the one right next to it, so I have to make the decision on which tile I should align all my changes with.

How can I achieve peace when it comes to aligning aerial imagery?

Viernes 28 de Agosto de 2026.

Se realizo:

1. Se modifica Vista Sur I a Villa Vista Sur I y se ubica en lugar que corresponde.

2. Se agrega Villas Vista Sur II, Villa Florencia 1-2-3-4-5 y Terrazas de Lourdes.

3. Se elimina Bosques de San Francisco Norte y Sur, a la par se agregan Villa Bosques de San Francisco I-II-III.

4. Se agrega Condominio José Gil de Castro.

5. Se agrega Condominio Escultor Samuel Román Rojas y se cambia nombre de vía a Calle 21 de Mayo.

6. Se agrega Población 21 de Mayo y se agrega Condominio Francia con sus vías.

7. Se mueve área de construcción y se agrega Condominio Santa Josefina y Condominio Don Nicolás Anich Asís.

GSoC 2026 with OpenStreetMap

---

Routing through pedestrian areas in Valhalla

During this Google Summer of Code, I’ve solved a very common problem that pedestrian routers have by adding the feature to route through open pedestrian areas, such as squares, plazas, pedestrian zones… instead of routing around the perimeter.

It’s now merged and running on the full planet build. Where pedestrians used to be sent all the way around the perimeter, Valhalla now creates a path straight across it.

Before :

… See full entry

Alam ko na hindi mahalaga ang ranggo at stats dito sa OSM

Alam mo kung ano ang matindi, Na dalawang buwan at isang linggo pa lang ang lumipas, mula nang nagsimula akong mag-ambag sa OpenStreetMap

Sa totoo lang, naniniwala akong isa ako sa pinakamabilis lumaking OSM contributors sa Pilipinas.

I plan to participate in State of the Map 2026 next week.

So, in preparation for it, I read the programme page on the SOTM 2026 website and manually picked out all the topics that I’m interested in.

I plan to use the list below as a guide to help me navigate the programme and decide which events I should attend.

---

… See full entry

I know rankings and stats don’t matter here in OSM

You know what’s crazy, That It’s only been 2 months and 1 week, Since I started contributing in OpenStreetMap

For real, I genuinely think, I am one of the fastest growing OSM contributors in the Philippines.

Following the launch of the OSM Skeleton methodology (see this post, featured in weeklyOSM #836), I wanted to illustrate it with an example by resolving the main reports in an area of Senegal where the demo is currently being deployed.

At first, I thought about mapping the Koungheul department, but given the scale of the task, I ultimately settled for one of its arrondissements, Ida Mouride. Arrondissements are the third level of administration in Senegal, after regions and departments.

The small size and limited number of Skeleton reports in this district allowed me to complete the task in about fifteen sessions of my daily OSM mapping routine.

The initial situation as of July 31 was as follows, for the entire department:

And zooming in on Koungheul:

… See full entry

Suite au lancement de la méthodologie OSM Skeleton (voir ce billet, repris dans hebdoOSM 836), j’ai voulu l’illustrer avec un exemple, en résolvant les principaux signalements sur une zone du Sénégal, où la démo est actuellement déployée.

J’ai pensé au départ faire le département de Koungheul, mais vu l’ampleur de la tâche, je me suis finalement contenté de l’un de ses arrondissements, celui d’Ida Mouride. Les arrondissements constituent le troisième échelon administratif du Sénégal, après les régions et les départements, arrondissements.

La petite taille et le nombre limité de signalements Skeleton que contient cet arrondissement m’ont permis d’en venir à bout en une quinzaine d’occurrences de ma petite séance quotidienne de cartographie dans OSM.

La situation de départ au 31 juillet était celle-ci, à l’échelle du département entier :

Et en zoomant sur Koungheul :

… See full entry

I keep getting questions about multilingual names in OpenStreetMap, so I want to state my position clearly in one place.

I understand the OSM data-model argument: name:xx should normally represent a name that actually exists in that language, not simply a machine-generated translation. My mapping practice has evolved accordingly.

However, as a product person, I care primarily about the user experience.

A map is a product used by real people. If I travel to a country whose writing system I cannot read, a map that shows me only the local script may be technically pure but practically much less useful.

For example, an English-, Russian-, or Georgian-speaking traveller should not need to read Arabic, Chinese, Thai, or another unfamiliar script just to understand what street, station, church, village, or other place they are looking at.

There are several ways to solve this:

• verified names in the user’s language;
• standardized transliteration;
• localized rendering;
• automatic transliteration or translation at the application layer;
• sensible fallback rules when no localized name exists.

I do not particularly care which layer solves the problem.

What I care about is the outcome:

the user should be able to understand the map.

Database purity is useful when it improves data quality. It should not become an excuse for poor UX.

OpenStreetMap is a geographic database, but the value of that database ultimately comes from people being able to use it.

For OSM naming guidance, see:

• Names
• Multilingual names

This post also reflects the lessons I took from my earlier discussion about AI-assisted multilingual tagging. I no longer treat automatically generated translations as sufficient evidence that a name:xx exists.

But my broader position has not changed:

when data modelling and usability pull in different directions, I will keep advocating for a solution that gives users both accurate data and a usable map.

… See full entry

Thanks to a note suggested adding a Via Crucis station, I checked how similar routes in Switzerland are mapped in OpenStreetMap.

OSM tags relations for these with worship=stations_of_the_cross (and type=route route=worship ).

These relations model outdoor paths, not stations located inside a single church building.

• They can be found in traditionally Catholic regions of Switzerland. The ones mapped in OSM are mainly in Ticino and the canton of Solothurn (Soleure).
• Traditionally there are 14 stations, sometimes 15 stations.
• The monuments at stations vary: stones, crosses, sculptures, shrines or chapels.
• Distances and ascent vary: The ones at Sant’Abbondio and Astano follow the walls of the churchyard, the one to Madonna del Sasso has an ascent of almost 200 m.
• Some paths are old (18th or 19th century), others are more recent.
• The state of preservation varies greatly.
• Many routes lead to a chapel.
• Some paths to chapels feature stations on other subjects. These have route=worship , but not worship=stations_of_the_cross .

RSI counts 54 in Italian-speaking Switzerland; the scope may not match OSM’s outdoor-route tagging. OSM has 10 Via Crucis in Ticino.

OSM has about 2000 wayside crosses and 1200 wayside “shrines” in Switzerland. Some could be part of Via Crucis that aren’t mapped yet.

Here is a short clip of the one above Biasca and a longer video about the one at Klingnau.

List

Here are a few mapped:

… See full entry

好的關於這破電源我是非常生氣，一個貼牌電源把我一條記憶體打壞了，下次不用二手的了再用我是小狗。

反正這不是重點，壞了就壞了，誰不會壞嘛。but那個破回收場，明知道裡面都是銅線，還用廢鐵價跟我收，純把我當豬仔宰是吧，還只有十塊錢，十塊錢欸，掛社團我都至少還有一杯手搖，好歹也給我15 20我去買冰水喝，以後不去了，黑。

反正賣完我就嘎拉嘎拉的騎著我的破腳踏車四處閒晃了，三個小時大概騎了20公里吧，沒很累，但到底是誰會把鐵皮屋標廟宇啊，我想說我怎麼沒聽過埔里有這個廟，騎過去一看，還真的鐵皮屋，不過應該是別人家，這人真無聊。還有不知道誰把荒地裡標一個廟宇，我 跑進去被蚊子叮慘了。

今天主要是跑郊區，市區太多要標的了標不完，明天跑田裡的廟吧，車少也好騎。

I remember when I first heard about these traffic cameras this guy paid me to listen to apb and manually log the codes to later look up and learn. Thinking about it these people were dangerous but it was normal for me being a military brat who joined the Marines after high school. I reminded the highest ranking ones of my grandfather who served 30 years. I guess it stirred up emotions making them uncomfortable cause I was clueless of my early departure resulting me in an unforseen path unaligned with my preset destiny.

This contribution is what Google identified as alpr I have a pic of it is just amazing how it doesn’t catch the license plate of a vehicle sitting in the middle of the intersection to prove truth of the occasion. Truth shall set us free

Several weeks ago, my bicycle finally got fixed, after being abandoned in a broken state for a long time.

Before that, I didn’t even know that my bicycle was actually broken.

All I knew was that this bicycle was really, really, really tiring to use. Even a short-distance trip was enough to tire me out for the whole day.

I didn’t know that this was caused by something being broken, which meant I had to put in (approximately) twice the effort to make it move (citation needed).

At that time, I simply blamed my own weak body. “Maybe I’m not strong enough to wield this bicycle,” I thought, dejected, considering giving up my whole cycling career prematurely.

Then, one random day, one of my family members used that bicycle, noticed something wrong with it, fixed it, and then it was finally ridable without needing much energy and force.

Alright. Now I can use my bicycle again, I guess.

---

For short trips, I usually prefer a bicycle to a motorcycle, for several reasons, you name it : hiking gas prices, the need to reduce our own carbon footprint to save the Earth (back then I took the mandatory environmental college course and thanks to that I’m still brainwashed so hard about the need to reduce my own carbon footprint), a nice opportunity to exercise my body, and the previous trauma of using a motorcycle and then suddenly having the engine not start so I had to walk the motorcycle to the nearest repair shop which turned out to be not near at all. Yeah, plenty of reasons.

So this afternoon, I decided to take a long trip to several places in my town, using my recently fixed bicycle.

I started around 4 PM and went home again around 6 PM.

Now it’s night, I’m sitting in front of my laptop, and getting curious : how many kilometers did I traverse this afternoon while cycling?

---

I didn’t bring a proper GPS device to record my trip this afternoon, but I remember each path that I took today. So, what should I do to calculate the distance?

At first, I tried BRouter.

… See full entry

Oh, where do I start?

It was around October when I had recently imported the barangay boundaries of my hometown, Cauayan City. I had just asked the GIS team at the local Planning and Development Office to adjust the boundaries again, as many issues were identified in the dataset. A while later, I had an idea to try to help the GIS team adjust the boundaries, because I figured, “Why not?” and they’re probably doing a lot of work already, given so many developments in the area.

A little back and forth, but then they stopped responding. I mean, it is basically a 50/50 chance that they respond or not. A little later, I dove into further research and stuff, and I thought of asking an institution to collaborate with me on making a dataset for the PDO to use and merge with theirs, so it can be as accurate as possible (I like things accurate).

… See full entry

… See full entry

呃好吧，我是新來的，可以叫我kiwi，熱愛刷機的kiwi。 先說說我為啥來這好了，我是為了因應「去谷歌化」與「自製BOB包」（好吧我根本就怪人），在網上爬啊爬，並在F-Droid上下載了這個開源地圖。 然後我在我刷了crdroid 8.6的Sony Xperia xz1上安裝了這個軟體，挺讓我驚奇的是，意外的好用，絲滑流暢，甚至在我的低端機上都運行的非常順利。 於是今天我就開始跑點了，我居住的小鎮上有很多無名小店，跑點這個過程讓我感到愉快。 希望各位能多多包涵我這怪人：） My telegram: https://t.me/taro_7421

One thing that usually worries us as OpenStreetMap mappers is having our hard work deleted by someone else, either accidentally or intentionally.

Thankfully, our edits are still in the database.

The delete operation in OpenStreetMap is not really a delete operation. It simply flips the object’s visible flag from true to false, while retaining its entire history.

And you can resurrect them using JOSM’s “Undelete Object” feature, as long as you have the OSM object’s ID.

… See full entry

New flock alprs are appearing in neighborhoods near where I go to college. This is scary, do your part.

• ᵈⁱˢᵗᵃⁿᶜᵉᵈ