OpenStreetMap logo OpenStreetMap

Users' Diaries

Recent diary entries

The goal of https://openclimbing.org is to offer a non-commercial alternative to traditional climbing apps. Instead of filling private databases, we decided to design a structure for mapping climbing routes directly into OpenStreetMap.

This way, all data in the app is fully open and can be freely used or edited by anyone.

Climbing guide for climbing crag Hlubočepy, Czechia Image: Crag with marked climbing routes in Hlubočepy, Czechia

Key Features of openclimbing.org

  • Both the database and the code are completely open, so anyone can contribute with content or code. Thanks to the open data approach, anyone can create new ways to use or visualize the data.

  • Each climbing route has defined GPS coordinates, making it easier to find in the terrain. This is big advantage but also a limitation – you cannot add routes without it.

See full entry

Posted by hopet on 26 June 2025 in English.

UM980 is relatively affordable chip allowing multi-constellation multi-frequency GNSS with capability to provide raw phase measurements and hence potentially suitable for use in OpenStreetMap environment for precise mapping susing RTK.

In this post (or potentially series of posts), I will document my experiences with using UM980 with RTKLIB and other open-source or free tools.

Hardware

  • UM980 RTK GNSS USB Dongle with SMA antenna - ELT0222 (from gnss.store)
  • EM-FD700 antenna (form AliExpress)

The module can easily be configured using UPrecise tool coming with UM980. It is advisable to have faster than default COM port rate ~~~ CONFIG COM1 230400 ~~~ and use the signal group 2 (note that this leads to reset of the UM980 module ~~~ CONFIG SIGNALGROUP 2 ~~~

If you need to stop the stream of measured data, you can use ~~~ UNLOG ~~~ You can also UNLOG just specific messages.

If you want to store your config of the module so that it survives unplugging and restarting the module, use ~~~ SAVECONFIG ~~~

Software setup on Android

RTKGPS+ - collecting raw phase measurements

This is to collect the raw data in real time. The last RTKGPS+ tool is release is unfortunately relatively old and hence does not support, so the native UniCore support, which is available in demo5_b34L version of RTKLIB, is not yet available in RTKGPS+. Hence it is mostly useful for collecting raw data, rather than providing real-time solutions. If you want to collect the data, choose BINEX as the format and just log the input stream from the USB. This is how to configure the module:

CONFIG PPP DISABLE

OBSVMB 1

GPSEPHB 120
BDSEPHB 120
BD3EPHB 120
GLOEPHB 120
GALEPHB 120

GPSIONB 120
BDSIONB 120
BD3IONB 120
GALIONB 120

If you want to have faster measurements, you can use ~~~ OBSVMB 0.2 ~~~ But it may be then advisable to use at least 240kbaud rate of the COM port (which is generally well supported by RTKGPS+).

See full entry

Posted by jacobwhall on 23 June 2025 in English. Last updated on 1 August 2025.

UPDATE: Overture has since addressed many of my concerns regarding the language on their website. See here for more.


Dear Overture Team and OpenStreetMap Community,

This diary entry is a follow-up to an in-person conversation I had with some of the Overture team at State of the Map U.S. To their credit, they were quick to admit to and apologize for some of these omissions and verbally committed to making appropriate changes. I am therefore posting this entry for the following reasons:

  • to fully document my concerns,
  • to invite others to (respectfully please!) share their perspectives,
  • to hold the Overture team accountable for making these changes.

Clarity on Attribution Page

In my opinion, Overture does not take care to spell out the necessity of OpenStreetMap attribution on their Attribution and Licensing page. After an optional citation comes the following language:

The Overture Maps Foundation does not require text attribution or an OMF logo on maps, visualizations, and graphics created with our datasets. If you would like to credit Overture, we suggest: © Overture Maps Foundation.

So far it seems pretty easy to use Overture data. We can attribute…but only if we want. Easy peasy! After that, the potential for additional attribution requirements is raised:

Some of the data sources we use in Overture datasets do require their own attribution, according to their licenses. For example, our Explore tool displays several Overture datasets that are made wholly or in part from OpenStreetMap data, so we added the following line of attribution to adhere to the requirements of the Open Database License: © OpenStreetMap contributors, Overture Maps Foundation.

See full entry

Posted by Nodariel on 22 June 2025 in English.

Just reached 1000 edits, so thought it would be good to dump what I have learnt so far.

I started using OSM during the COVID 19 lockdown living in Hampshire, there were many walks available but the routes were not always clear on google maps. OSM was far better and I used it for information about walks before proceding. The level of detail was great and really useful.

Fast forward to 2023 and I decide to start contributing using StreetComplete and MapComplete. I drive an Electric vehicle and the information on charge points is so obscure I wanted to contribute. This led me down a rabbit hole of StreetComplete, filling in addresses, road surfaces, bus stop shelter. Anything and everything. I learnt the importance of information that would be useful to users.

Only in the last year, have I picked up mapping, this time using ID editor, chnaging tags was a little daunting but I slowly got there. I noticed some areas around me were missing house numbers. I learnt the importance of doing a survey on foot. There’s so much to learn from looking around IRL.

The last few weeks, I have been mapping out areas that had no buildings, only Residential Areas. A few thousand terraced houses later (feels like it) and quite a few places are on the map around Yarmouth (yeah I moved from Hampshire in this time!). Most imporatntly, check the alignment of background layers before starting a new edit! It is also useful to turn off layers not used, really makes things easier to see.

Latest update, I own a drone and can produce aerial images, very useful for areas not on Bing maps. This also gives me access to a 3D view of the area, very useful for double checking my survey notes!

Maybe I’ll do another one of these at 2000 edits?

Posted by Atirut W on 21 June 2025 in English.

Hi everyone. I recently started contributing to the map around my university’s campus, and I thought I should get in touch with the community a little. I’ve only labeled major buildings in the campus plus some road fixups, but I hope to continue contributing more to the map around here.

Oh yeah, this account is old, but I don’t even remember having it until I tried to sign up lol.

 韓国の百科事典・ナムウィキのオープンストリートマップの記事には以下のような記述があります。

【ネット右翼が独島(韓国名)の関連するすべての情報を消したこともあり、独島を日本が主張する行政区域である島根県沖ノ島に名前に変更する事例もあり、鬱陵島から独島へ行く航路の名前を「不正入国海路」に変更しながら独島関連の地物を日本語に変えるなど、島を水没させる編集もあった。 独島を竹島と表記して日本領土に編集する場合が多く起きるので、関心を持ってみるのも良い。 独島削除と竹島で作ったり消したりする問題は現在進行形だ。 独島の衛星写真はうまく出ていないので、すべて消してしまえばJOSMのような高度なツールでない限り、再編集するのが難しい点がある。 】

 まあ、そうだろうなあと思い適当な地物をクリックしてみました。私は東アジア地域で地名の日本語表記を書くのに四苦八苦しているので、竹島は興味深い場所です。

 いくつかのポイントは実効支配を行っている韓国側が定めたと考えられる日本語名称が書かれているのですが、肝心な『独島里』の日本語名称が削除されているではありませんか!

 ちなみに、韓国での『里』は日本では小字に該当します。

 削除した理由は『日本ではそう呼称されていないから』だそうです。しかも、政治的意図はない←?????

 この理論が罷り通るなら北方領土では同様な編集が行われていないのも疑問です。

 実効支配を行っている国の呼称を削除するのは、日本人の私ですら甚だ理解できません。当該編集は私が修正しましたが、おそらく、今後もこのような編集が行われると推測されます。

 日本の立場を表明することは重要ですが、【政治的意図のない編集】や【竹島は日本の領土】という編集をここで行って何になるというのでしょうか?

 少し愚痴りたくなったので書かせて頂きました。

Posted by Nicolelaine on 21 June 2025 in English. Last updated on 19 November 2025.

Hello OSM community,

We (myself and the MapSwipe community!) want to hear from you!

My name is Nicole Siggins and I’m currently on the governance team at MapSwipe a humanitarian mobile phone app and web app where volunteers can contribute to geospatial data projects.

I presented a talk, “MapSwipe Today and its Future in OSM” along with Benjamin Herfort from HeiGIT at State of the Map Europe 2025 in Dundee, Scotland. (recording coming soon…)

The talk has two main purposes:

  1. Let people know more about MapSwipe, its history, and what one can do with it today!

  2. Begin a discussion within the OSM community about what it might mean for MapSwipe to become its own OSM editor, especially if AI or machine learning models are involved.

The MapSwipe community wants to ensure that if we are going to move forward with any ideas and planning around becoming an OSM editor, and maybe even one that has AI in the mix, that we also make sure the OSM community is present in our discussions.

Some concrete items we’d like to discuss with the OSM community are:

  • Should MapSwipe become it’s own OSM editor to add attributes to existing OSM features, using it’s own built in validation process?
  • If 3+ people validate an attribute, who shows as the OSM editor? (One of the users, or a MapSwipe bot?) Any experience or feedback would be helpful!
  • If humans refine AI models when (if ever) AND HOW would it be ok to add AI generated data to OSM …
  • AND how could we make sure to validate the quality of that data?

The talk at State of the Map Europe was an extension of the same talk I gave at State of the Map US 2025 in June in Boston, Massachusetts.

You can view that talk on YouTube.

Looking forward to the discussion!

Nicolelaine

Posted by OG Alex on 20 June 2025 in English.

Today we at Unique Mappers did our weekly presentation were I made a presentation on Humanitarian Use of OpenStreetMap where i talked about OSM what it is why it is crucial in Humanitarian context how it provides critical data in poorly mapped or un mapped areas. I still talked about the basics( getting started with OSM), the intermediate OSM data use(Analysis and virtualization) and advanced used of OSM. I still talked about the tools and resources( software) used with OSM and example is the HOT tasking manager, QGIS [qgis.org], Umap and many other software, site, organisation using OpenStreetMap

Posted by BeardMD on 20 June 2025 in English.

As I write in my profile, I was really peeved off, that dozens of “Camino Apps” (apps used to track users along the Ways of St. James, the Caminos de Santiago) did ask their users for updates to opening hours or bed counts in albergues, but never contributed back to OSM what they’d discovered.

I launched a “test balloon” in 2023, renaming an accommodation to reflect its real name, and lo and behold, of the 23 apps I checked, 22 had suddenly also changed the name, meaning they used OSM data, but didn’t ever give back to the community.

So I changed it, wrote a Camino App that did contribute back. We don’t expect our users to have OSM accounts. Instead, we recruited a bunch of volunteers, who are served the changes, check them personally, and then contribute them back to OSM.

First step: fountain potability. Along the Caminos are thousands of fountains providing drinking water to the 500k pilgrims walking the Way every year. Spain is excellent, when it comes to drinking water quality, but sometimes fountains dry out, become undrinkable, or become drinkable again. Nothing sucks more than going 1000m out of your way to find out that fountain has been closed.

I’ll semi-manually start updating fountain “drinking_water=” tags over the next 30 days along the most famous route, the Camino Francés. Once that’s stable and enough volunteers have been trained and shown that they can do this without harming the dataset, we’ll expand to hostel data.

Location: Zabalguneak / Ensanches, Segundo Ensanche, Pamplona, Iruñerria / Comarca de Pamplona, Navarre, Spain

junction:cluster:radius - тэг для указания максимально возможного радиуса влияния/отношения пересечения на окружающие объекты


Синтаксис

node.tags {
   junction:cluster:radius: number[1..N]
}

Применяется для объектов

Когда этот тэг применяется для объектов типа node, эта эта точка должна являться пересечением - junction. См статью про это. Тэг указывает радиус окружности в которую может быть вписана функциональная зона для данного пересечения. Это означает, что в данной зоне другие объекты (парковки, переходы, стоп-линии и так далее) могут отображаться или интерпретироваться как-то иначе. В какой-то степени это понятие соотносится с понятием функциональная зона перекрестка, только в данном случае - простейшего пересечения.

Главной мотивацией введения данного тэга было обеспечить возможность группировки нод соседних пересечений в обобщающее понятие “Перекресток”. Для этого у нас есть несколько различный вариантов, как это может быть сделано:

  1. Relation: type:intersection, members[node1,...,nodeN, way1,..., wayM]
  2. Атрибут у ноды который является ключом обобщения(имя-идентификатор кластера) junction:cluster = name or id
  3. Радиус, который при наложении(union) окружностей даст общий полигон для некоторого множества нод junction:cluster = 5

Плюсы и минусы первых двух подходов очевидны. Опять relation, ссылочная целостность, генерация тэга. Все эти способы решают одну задачу - управляемая кластеризация нод пересечений в некую более сложную структуру данных.

Рассмотрим 3 способ.

  • Очень геометричный, отражает площадные/линейные характеристики пересчения
  • Не требует поддержки ссылочной целостности( как в 1) и контроля уникальности, для 2
  • Можно найти зависимость или корреляцию с другими свойствами ноды (число полос)
  • Просто числовое значение в метрах
  • Формально нового объекта типа перекресток не появляется, но он всегда может быть получен простейшей операцией buffer + union
  • То есть принцип бритвы Окамма - не плодим новых сущностей без необходимости

See full entry