
Chaos Computer Club - recent events feed
4,903 episodes — Page 56 of 99
Mapping for Impact: Introducing Open Mapping to Civil Society in Sri Lanka (sotm2025)
Journey to the Center of the Planet (sotm2025)
Making OpenStreetMap More Accessible for New Mappers (sotm2025)
OpenStreetMap (OSM) has become one of the most valuable open mapping platforms, yet many aspiring contributors struggle with the initial learning curve. Technical complexities, lack of structured on boarding, and limited localized training resources often prevent new mappers from fully engaging with OSM. This talk will highlight key challenges faced by beginners and propose practical solutions to make OSM more accessible and inclusive. Drawing from my experience as an Open Mapping Guru, a YouthMappers leader, and a trainer with the Open Mapping Hub - Asia Pacific, I will discuss effective strategies for on boarding new contributors. These include localized training resources, mentorship programs, and simplified tool introductions. By addressing these gaps, we can empower a more diverse community to participate actively in open mapping. OpenStreetMap (OSM) is a powerful tool for open geographic data, yet many new contributors struggle with onboarding due to technical complexities, lack of structured training, and limited mentorship opportunities. These challenges hinder OSM’s potential to grow into a truly inclusive and diverse mapping platform. In this talk, I will address key barriers faced by beginners and present actionable solutions to enhance accessibility in OSM. Drawing from my experiences as an Open Mapping Guru, a YouthMappers leader, and a trainer with Open Mapping Hub - Asia Pacific, I will share strategies such as localized training materials, mentorship programs, gamification techniques, and social media engagement. Through case studies of successful community-driven initiatives, including YouthMappers projects and grassroots mapping efforts in Bangladesh, I will demonstrate how we can empower new mappers and improve retention. By fostering a more beginner-friendly environment, we can ensure OSM remains a valuable resource for humanitarian efforts, disaster response, and everyday navigation. Attendees will gain insights on practical ways to support new contributors in their communities and help make OSM more accessible, engaging, and sustainable. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/TX88WD/
World-Map-Explorer – Explore the world with ease (sotm2025)
World-Map-Explorer is a web application that helps the visually impaired learn about the world and understand maps. It is a great pleasure to note that the application was developed by a group of students from the Computer Science Department of Sri Krishnapuram Government Engineering College in collaboration with Zendalona with support of OpenStreetMap Kerala Community. This app, which can be used by both the visually impaired and the sighted, is based on the free software OpenStreetMap. This app describes the world based on audio cues with the help of a screen reader. Existing applications like Google Maps cannot be used with a keyboard. Therefore, the visually impaired cannot use map applications. They understand the structure of areas by swiping their hands through the lines and dots that stand out on the tactile map. They understand where a place is by waving their hands over the place name written in Braille. World-Map-Explorer is a web application that helps the visually impaired learn about the world and understand maps. It is a great pleasure to note that the application was developed by a group of students from the Computer Science Department of Sri Krishnapuram Government Engineering College in collaboration with Zendalona with support of OpenStreetMap Kerala Community. This app, which can be used by both the visually impaired and the sighted, is based on the free software OpenStreetMap. This app describes the world based on audio cues with the help of a screen reader. Existing applications like Google Maps cannot be used with a keyboard. Therefore, the visually impaired cannot use map applications. They understand the structure of areas by swiping their hands through the lines and dots that stand out on the tactile map. They understand where a place is by waving their hands over the place name written in Braille. But this has many limitations. It only provides a few details about the place. And it is difficult to carry it anywhere. That is why the Map World-Map-Explorer, which can be easily used from anywhere, is opening up new possibilities for visually impaired peoples. https://map.zendalona.com/ This Project Launched by Dr. R Bindu, incumbent Minister for Higher Education and Social Justice, Government of Kerala. https://www.facebook.com/drrbindhu/posts/pfbid02vPrXmPFuG6UE2p8yymnqxVKWwUTh2nA5FkDfw37qCNgrntCqiczPNoc8hJK81H7kl Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/ABLRCG/
Making OpenStreetMap More Accessible for New Mappers (sotm2025)
World-Map-Explorer – Explore the world with ease (sotm2025)
~Translation and Mapathons~ Building Open Mapping Education by Students; A YouthMappers Student Project from Furuhashi Lab, Japan (sotm2025)
fAIrSwipe (sotm2025)
fAIrSwipe (sotm2025)
This project aims to combine AI‐generated building predictions from fAIr with crowdsourced validation and conflation workflows in MapSwipe, ultimately pushing high‐confidence conflated map data into OpenStreetMap (OSM). fAIr is an AI‐powered mapping assistant by HOT that helps users map smarter, faster, and more accurately. MapSwipe is a crowdsourcing app that lets volunteers validate or identify features (e.g. buildings) quickly on satellite imagery. Objectives - Automate the creation of MapSwipe projects from fAIr (with building predictions, TMS layers, etc.). - Validate AI‐predicted features via MapSwipe’s volunteer workflow via redundancy. - Conflate validated features with existing OSM data. - Upload conflated data back to OSM in a controlled, conflict‐aware manner. - Provide feedback to improve fAIr's AI models based on volunteer validation results and possibly task the tasking manager to manually map the parts that are difficult to validate via MapSwipe. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/3HRTVP/
~Translation and Mapathons~ Building Open Mapping Education by Students; A YouthMappers Student Project from Furuhashi Lab, Japan (sotm2025)
This talk introduces a student-led initiative from the Furuhashi Lab at Aoyama Gakuin University in Japan, focusing on the localization of global open mapping knowledge. As a YouthMappers chapter, we are currently translating the open-access book Open Mapping towards Sustainable Development Goals into Japanese. The aim of this translation project is to make critical mapping knowledge accessible to non-English-speaking students and to encourage greater participation in the open mapping community. Recognizing the importance of engaging peers through action, we organized our first online Mapathon in early 2025. This event helped beginner mappers gain hands-on experience and fostered a sense of community among participants. Building on its success, we are now planning a collaborative inter-university Mapathon scheduled for July 2025, with the goal of expanding our network and impact across academic institutions in Japan. In this talk, we will share insights from managing the translation workflow, including how we divided chapters, ensured consistency in technical vocabulary, and utilized collaborative tools such as GitHub for version control. We will also reflect on our outreach efforts—how we communicated the value of open mapping to our peers and engaged them in both the translation and mapping processes. Through this project, we aim to empower students with the knowledge and tools needed to contribute to global development goals through geospatial technologies. Ultimately, our initiative highlights the potential of youth-driven efforts to strengthen open mapping education and build inclusive communities, even in regions where English is not the primary language. Since October 2023, YouthMappers AGU—based at the Furuhashi Laboratory of Aoyama Gakuin University in Japan—has been leading a student-driven translation project of the open-access book “Open Mapping towards Sustainable Development Goals.” This book, published by Springer, collects diverse case studies from YouthMappers chapters across the globe, demonstrating how young people address social and environmental challenges through open mapping. In Japan, awareness of YouthMappers, OpenStreetMap (OSM), and humanitarian mapping remains relatively low, particularly among students. To bridge this gap, our seminar group proposed translating the book into Japanese—not only to make the content accessible to non-English speakers, but also to inspire local mapping activities based on global examples. This translation effort also serves as an opportunity to cultivate academic and digital literacy skills within the seminar. The project started during our Fall 2023 semester. We divided the 370-page book among 12 seminar members, assigning roughly 20 pages each. To ensure quality, we created shared translation rules covering tone (formal), terminology consistency (e.g., “map” rendered consistently), format alignment with the original layout, and reference handling. All translations are stored collaboratively on Google Drive and GitHub. As of April 2025, 140 pages have been completed. Our goal is to finalize the full Japanese version by September 2025. In parallel, our team also initiated hands-on engagement through a student-led online Mapathon held in January 2025. The event involved preparation of posters (via Canva), registration forms, and slide decks. Participants, mostly undergraduates, were introduced to creating OSM accounts, navigating the Tasking Manager, and mapping buildings in Yoronjima, Japan. Post-event surveys showed that while time was limited, many valued the experience of learning the basics of humanitarian mapping. Key reflections from this event include the importance of communication setup: for instance, participants using smartphones for Zoom and laptops for mapping caused screen-sharing issues, hindering real-time support. Some were unfamiliar with Zoom’s screen-sharing function. These experiences highlighted the need for clearer technical preparation, which we plan to implement in future events. Looking ahead, we are planning a collaborative Mapathon with another university in July 2025, aiming to expand inter-university cooperation in humanitarian mapping and involve more Japanese students in the global YouthMappers movement. This presentation will showcase our dual-track initiative: (1) localizing global YouthMappers knowledge through structured translation work, and (2) fostering peer learning and community engagement through Mapathon events. As students in the Furuhashi Lab, we will also reflect on how being part of an OSM-active academic lab has supported our growth. We hope our efforts inspire other YouthMappers chapters—especially in non-English-speaking contexts—to pursue similar educational and participatory projects. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/8VMYYW/
How to Complete Japan’s Building Mapping in One Year? Strategies for Integrating PLATEAU Data into OSM (sotm2025)
Since 2020, Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT) has led Project PLATEAU, releasing 3D city models in CityGML format. By 2025, over 236 municipalities have published building datasets compatible with the Open Database License (ODbL), suitable for integration into OpenStreetMap (OSM). Since 2022, OpenStreetMap Japan volunteers have imported PLATEAU’s Level of Detail 1 (LOD1) building data into OSM, completing imports for 13 cities as of May 2025. Despite these efforts, only about 62% of Japan’s estimated 38 million building polygons are mapped in OSM. This presentation proposes a roadmap to complete Japan’s building mapping within one year, focusing on optimizing PLATEAU data imports, ensuring data consistency, and strengthening community collaboration. — Strategies for Integrating PLATEAU Data into OpenStreetMap Since 2020, Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT) has spearheaded Project PLATEAU, a national initiative to develop and openly release 3D city models in CityGML format. As of 2025, over 236 municipalities have published detailed building datasets, which are compatible with the Open Database License (ODbL) and suitable for integration into OpenStreetMap (OSM). These semantically rich and high-precision datasets hold significant potential for applications in urban planning, academic research, and civic technology. Building upon previous presentations at State of the Map conferences, this talk outlines the progress made since 2022 by volunteers from OpenStreetMap Japan in importing PLATEAU’s Level of Detail 1 (LOD1) building data into OSM. As of May 2025, imports have been completed for 13 cities. In addition to these imports, numerous volunteer mappers continue to enhance OSM’s building data through aerial imagery tracing and other methods. Despite these efforts, only approximately 62% of Japan’s estimated 38 million building polygons are currently mapped in OSM, leaving about 14 million polygons—38%—yet to be integrated. Addressing this gap necessitates a clear numerical target and a strategic approach. This presentation sets forth an ambitious yet achievable goal: to complete the mapping of Japan’s building data within one year. We will provide a quantitative analysis of current progress and delineate a roadmap to achieve this objective. Key strategies include optimizing the importation of PLATEAU data, ensuring data consistency between PLATEAU and OSM, and strengthening collaboration within the mapping community. By sharing these insights and methodologies, we aim to accelerate the integration of PLATEAU data into OSM and to serve as a model for other countries seeking to enhance their building mapping efforts. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/N89YSU/
Community-Driven Data: How Local Knowledge is Reshaping Urban Waste Systems (sotm2025)
Effective urban solid waste management in African cities remains a critical challenge characterized by rapid urbanization, limited infrastructure, and insufficient data. Traditional top-down approaches often fail to address the unique, hyperlocal realities of waste generation, disposal behaviors, and service gaps. This presentation explores how OpenMap Development Tanzania (OMDTZ) with the funds from the World Bank, has implemented community-driven data collection approaches in Tanzania, Mozambique, and Kenya to fill this knowledge gap and build more responsive, inclusive waste management systems. Through the integration of OpenStreetMap (OSM), mobile data tools like OpenDataKit, OsmAnd, and participatory mapping techniques, communities have been engaged in mapping informal waste collection points, illegal dumping sites, transfer stations, and service coverage areas. By centering local knowledge, these projects have produced actionable geospatial data that informs city-level planning, supports environmental health interventions, and amplifies citizen voices in waste governance. Participants will gain insights into how grassroots data collection not only improves the quality of open data but also reshapes power dynamics in urban planning, placing communities at the forefront of building cleaner, more sustainable cities. Urban waste management systems across many African cities face severe capacity challenges. Waste collection services are often inconsistent, infrastructure is unevenly distributed, and data on where and how waste is generated and handled is largely missing or outdated. This session shares lessons from community-centered mapping initiatives implemented by OMDTZ and partners in Dar es Salaam, Nacala- Mozambique, and Mombasa -Kenya, where local residents have played a central role in generating critical waste-related datasets. Instead of relying solely on municipal audits or high-tech surveys, the approach engages community members, youth groups, and local organizations to map and monitor solid waste dynamics in their own neighborhoods. Equipped with mobile tools, participatory maps, and training, local data collectors document: 1. Community perception and attitudes on the solid waste management 2. Waste handling modes, dispose ways and payment behavior 3. Informal dumpsites and hotspots for illegal dumping 4. Waste collection service routes and gaps 5. Types and volumes of waste generated in residential areas 6. Proximity of vulnerable communities to unmanaged waste zones These datasets are then analyzed and shared with the responsible authority, allowing governments and funders to create the city-level strategic plans for solid waste management. While researchers access real-time, location-specific information for designing targeted interventions, from optimizing collection routes to deploying behavior change campaigns. This talk will cover: 1. The methodology for engaging communities and collecting waste-related geospatial data 2. Case studies from Tanzania, Mozambique, and Kenya that demonstrate impact 3. Tools and platforms used for data collection, visualization, and decision-making 4. How local knowledge has influenced waste policies and infrastructure planning 5. Challenges and solutions for sustaining community engagement and ensuring institutional uptake of data By leveraging community knowledge and open technologies, these projects have helped cities move toward more resilient, inclusive, and evidence-based solid waste systems. The session will also highlight how this approach contributes to climate resilience, youth empowerment, and the democratization of urban data ecosystems. Ideal for practitioners, urban planners, open data advocates, and local government officials, this session will offer replicable strategies for harnessing community-driven data to transform waste governance and sustainability in rapidly growing urban areas. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/VTGCCY/
How to Complete Japan’s Building Mapping in One Year? Strategies for Integrating PLATEAU Data into OSM (sotm2025)
Lightning Talks II (sotm2025)
## Utilizing OSM with QGIS _by Ariel Dome_ ## Open Mapping potential in Jakarta Metropolitan: Case Study of Mapillary Capture in Depok City _by Yabes Butar Butar_ ## People data tracing movement with GPS _by CJ Capuli_ ## All of five km _by Ewen Hill_ Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/NVHF3U/
Lightning Talks II (sotm2025)
The Ugandan Geo Quests: Mapping Libraries and Museums into the Knowledge Commons (sotm2025)
Across Uganda, countless community libraries and local museums hold untold stories, cultural memory, and untapped knowledge—but remain invisible on the digital map. The Ugandan Geo Quests project aims to change that. This session introduces two complementary national mapping initiatives: The Ugandan Libraries Geo Quest and The Ugandan Museums Geo Quest, which leverage OpenStreetMap, Wikidata, and Wikipedia to bring local knowledge institutions into the global digital commons. We’ll explore how we: Mobilized local communities, Wikimedians, and mappers for participatory documentation Used tools like uMap, ODK Collect and KoboToolbox to ensure data quality and visibility Linked OSM data with Wikidata and Wikipedia for multi-platform discoverability Addressed challenges around verification, tagging, and sustainability Attendees will gain insights into designing thematic mapping campaigns that go beyond infrastructure—to elevate culture, education, and heritage in open data ecosystems. Whether you're a mapper, librarian, or advocate for digital equity, this session offers inspiration on how maps can preserve identity and empower communities. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/MTS9TS/
The Ugandan Geo Quests: Mapping Libraries and Museums into the Knowledge Commons (sotm2025)
OpenStreetMap x Wikidata Collaboration: Taiwan Case (sotm2025)
Lessons from teaching OpenStreetMap in a masters program (sotm2025)
Teaching with OpenStreetMap has mostly been focused at the high school and undergraduate levels. In this presentation, I will share my experience of integrating OpenStreetMap into a professional graduate program, the Master of Geomatics for Environmental Management at the University of British Columbia. We will look at some of the opportunities for teaching with OpenStreetMap at the graduate-level, identify some of the challenges with teaching natural resource management with OpenStreetMap, and consider some broader implications for incorporating OpenStreetMap into graduate education. Finally, I will share some open educational resource lesson plans for teaching with OpenStreetMap that can be adapted for your classroom. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/P8MPHS/
Lessons from teaching OpenStreetMap in a masters program (sotm2025)
OpenStreetMap x Wikidata Collaboration: Taiwan Case (sotm2025)
Not only NSI, but Wikidata can be integrated with OpenStreetMap, and vice versa! Many years ago, with the help of Wikidata Taiwan, OpenStreetMap Taiwan has mapped all 7,000 villages, and also cross-linked to Wikidata. We also have a similar river project of mapping all rivers in Taiwan and crosslinking to Wikidata with the help of the river code published by the Taiwan government. In this talk, we want to further talk about the school, mountain, church, temple, and hiking trail mapping projects, which also have the corresponding external 3rd-party Wikidata property. We will also describe the process like documenting and tools involved NSI is quite an important project to cross-link OpenStreetMap and Wikidata, but the story can go further. With the help of the Wikidata Taiwan community, there are some achievements, like the rivers crosslinked (P9170), and the villages mapping and cross-linked (P5020). We also use Wikidata information to double check if the school's basic information, like address or website is correct or not, and adding its coordinates from OpenStreetMap geolocation. The OpenStreetMap Taiwan continues to integrated more Wikidata and external identifier like religious related Chinese Church and Organization Dictionary ID(P13272) and Platform for Taiwan Religion and Folk Culture ID(P13349); hiking related identifier: Hiking Note Trail identifier(P13406), Hiking Note mountain identifier(P13407). We also use automatic approach to make sure both OpenStreetMap and Wikidata records all those entries and information. By the help of Listeria powered by OpenStreetMap, and the datamining tools Overpass Turobo, to doing some documents and data check. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/JUSRXF/
Platinum Sponsor Session: From Edits to Impact - TomTom’s Journey with OpenStreetMap Communities (sotm2025)
TomTom partners with OSM communities worldwide to build richer maps through collective action. In this SotM Global event 2025, we share how we support university mapathons, YouthMappers, and humanitarian mapping in 189 countries, reflecting on learnings and challenges in scaling community engagement while advancing OSM’s mission through meaningful, local partnerships. TomTom and the OpenStreetMap Community share a common vision: building a map that reflects and serves the world, powered by local knowledge and collective action. In this session, We will be taking you behind the scenes of TomTom’s partnerships with OSM communities worldwide, demonstrating how collaborative efforts are strengthening the quality, freshness, and inclusiveness of OpenStreetMap data. From hosting mapathons with universities and YouthMappers chapters to supporting humanitarian mapping during crises, TomTom’s initiatives have connected 3,397+ OSM contributors across 189 countries. We will share real examples of how these collaborations are transforming local mapping ecosystems while providing technical resources, training, and data that support contributors in making impactful edits. We will also discuss scaling corporate-community collaborations responsibly, challenges faced in diverse regions, and opportunities for co-creating projects that benefit both OSM and local communities. Participants will leave with ideas and inspiration on how to replicate or adapt these approaches to engage more mappers, foster skill-building, and deepen OSM’s reach. Join us to discover how mapping together can empower communities and create a richer, more inclusive map of the world. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/TZNMJG/
Addressing Participation Gaps in Nepal’s OSM Ecosystem: Strategies for Long-Term Community Retention (sotm2025)
Addressing Participation Gaps in Nepal’s OSM Ecosystem: Strategies for Long-Term Community Retention (sotm2025)
Nepal’s OpenStreetMap (OSM) community is currently experiencing a noticeable decline in active participation and long-term engagement. While the ecosystem once thrived on regionally supported events and student-led enthusiasm, it now faces a rise in passive contributors and a lack of continuity in mapping efforts. This trend is particularly concerning in a context like Nepal, where access to reliable and up-to-date geospatial data remains limited, and OSM presents a valuable, open-source alternative for inclusive mapping and data democratisation. A key factor contributing to this decline is the narrow framing of OSM as a repetitive editing platform, often reinforced by conventional training sessions that overlook the broader creative and applied potential of the tool. Innovative and accessible platforms such as Mapillary, MapSwipe, and OSM-based design tools—which can make mapping more engaging and relevant to everyday challenges—are rarely introduced or integrated into learning experiences. Drawing on my own experience organising initiatives such as the OSM Hackfest and thematic map design based on OSM dataset competitions, I have witnessed how student engagement can be restored when OSM is presented not just as an editing tool but as a gateway to problem-solving in urban planning, disaster resilience, and many more potentials of the OSM datasets. More importantly, there is a need for advocacy for a shift in narrative: from simply contributing data to understanding the importance of OSM and utilising it. By embedding OSM awareness from the student level to policymaking stakeholders, Nepal can foster a resilient and self-sustaining mapping ecosystem. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/RUCRPS/
Resilience Starts with a Map: Community-Led OSM Action in Dhaka’s Climate-Vulnerable Settlements (sotm2025)
Platinum Sponsor Session: OSM Data for Two-Wheelers and Safety (sotm2025)
Platinum Sponsor Session: OSM Data for Two-Wheelers and Safety (sotm2025)
This talk delves into the unique challenges and opportunities of enhancing OpenStreetMap (OSM) data for two-wheeler routing and safety. Two-wheelers are a vital mode of transportation within SEA, yet their safety needs are often overlooked. Drawing from our experience in building and contributing enriched safety data back to OSM, we explore how attributes like lighting conditions, speed bumps, and pothole assessments can make navigation safer and more efficient for two-wheelers. Through this talk, we aim to inspire the OSM community to prioritize safety attributes for improving map data quality for two-wheelers. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/8KQRCY/
Resilience Starts with a Map: Community-Led OSM Action in Dhaka’s Climate-Vulnerable Settlements (sotm2025)
Dhaka, one of the most climate-affected cities in South Asia, is home to millions of migrants living in informal urban settlements. These densely populated areas are increasingly exposed to environmental and health hazards, particularly seasonal waterlogging and recurring dengue outbreaks. Despite the severity of these issues, many of these settlements remain underrepresented in official datasets, limiting the ability to implement targeted and effective resilience measures. As part of our Capstone Project under the Climate Resilience Fellowship (CRF), this initiative leveraged OpenStreetMap (OSM) to conduct a community-led, GIS-based multi-hazard assessment in some of Dhaka’s most climate-vulnerable settlements. Our work followed a two-phase approach: Field-based data collection and mapping, where youth and women from local communities actively identified water accumulation zones and dengue breeding hotspots using OSM tools. Awareness building and local preparedness activities, using the mapped data to facilitate community dialogues and promote action for health and disaster resilience. Through this participatory mapping effort, we not only generated critical geospatial data but also strengthened local capacity to respond to climate and health risks. OSM served as both a data platform and a tool for empowerment, enabling residents to visualize their vulnerabilities and advocate for solutions. This talk will showcase how open mapping, when integrated with community engagement and local knowledge, can effectively support disaster risk reduction, public health planning, and climate adaptation in marginalized urban areas. We will also reflect on the potential for replicating this model in other cities facing similar climate and health challenges. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/8KKLCS/
What opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)
What opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)
In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project. The work described in this abstract is motivated by the rising interest in studying the human-like traits of Large Language Models (LLM). Since LLMs are pretrained on vast amounts of human data, it is reasonable to assume that LLMs can reflect the AOVs (Attitudes Opinions and Values) embedded in the data [1]. LLMs are increasingly being used in open-ended contexts, where the opinions they reflect in response to subjective queries can have a profound impact, both on user satisfaction, and shaping the views of society at large [2]. While LLMs can never have opinions in the same way a person does these systems can be prompted to generate opinion-like text. For example, in Malleson et. al [3] LLMs are used to give opinions on neighbourhood change. Chatbots can output opinionated language, but service providers usually purposely describe these outputs as "generated perspectives" rather than actual beliefs [4]. There is usually topic-specific or subject-specific safeguards implemented—especially around political discourse and advice involving risk to human life, health, and so on. Despite this, the huge popularity of chatbots mean that opinionated text produced by chatbots and LLMs can have the potential to be very influential on the reader or user[4,5]. The attraction of these outputs emerge from the fact that these opinions arrive instantly, they are written in very confident, certain, and reassuring language, and often include citations and evidence to support claims. When LLMs are integrated into search engines, social platforms, and other interactive systems, their outputs when the user is seeking opinions or viewpoints can become a user’s first—or only—exposure to a topic[5]. There are positives and negatives to these situations. On the one hand, the systems can provide well-sourced perspectives to a wide audience but they also risks amplifying hidden biases in training data or reflecting other ideological priorities embedded in deployment of these models. Ultimately, their persuasive power arises less from any individual (correct,incorrect,biased) answer and more from their persistent and seamless presence in everyday workflows where over time they can shape public discourse [1]. With this we decided to investigate the opinions or views expressed by some of the major LLMs when prompted or questioned about their opinions on OpenStreetMap (OSM). Given that LLMs will have been trained on data and information related to OSM - papers, blogs, social media, presentations, etc., can we understand the Attitudes Opinions and Values (AOVs) of LLMs to OSM? To the best of out knowledge, there is no research currently reported about this topic. As all major LLMs have been trained on huge volumes of data from the Internet spanning many years, even decades, we make the assumption that it is very likely that LLMs will have consumed papers, articles, presentations, social media, and so on regarding all aspects of OSM and in particular legacy topics such as: data quality, comparison with official or authoritative sources, accuracy, bias in crowdsourced data, and so on. In related work, Santurkar et. al. [2] developed an extensive framework to study the opinions reflected by LLMs and their alignment with different existing human population opinion-based surveys. These surveys were structured multiple choice-type surveys. Everyone involved in the OSM community has their own AOVs about OSM, the OSM ecosystem, OSM strategic direction, and so on. For example, there are often differences in OSM mapper behavior "explained by clashing values and opinions within and across different mapper subgroups" [6]. As mentioned above, we are not aware of any similar work having been carried out so far. The majority of work published about LLMs and OSM revolves around using the LLMs as mapping assistants [7,8] or using LLMs to enrich existing data or allow more accessible approaches to mapping [9]. Other works have considered the ability of chatbots such as ChatGPT to take a GIS exam [10] or exploit the deep contextual understandings in LLMs for building function classification in OSM [11]. Currently, at the time of writing, we are beginning the implementation of this work. However, a high-level description of our methodology is outlined as follows: 1. Selection of the target group of LLMs/chatbots: ChatGPT (GPT-4o / GPT-4-turbo), Microsoft Copilo, Anthro
Open Mapping for Urban Resilience: A Routing Model to Nearest Safe Spaces in Earthquake-Vulnerable Dhaka (sotm2025)
Dhaka, one of the most densely populated cities in the world, faces a high risk of catastrophic damage in the event of a major earthquake. The lack of accessible open spaces poses a serious challenge for emergency evacuation and survival. This project utilizes OpenStreetMap (OSM) road and building layers, combined with satellite imagery, to identify existing open spaces across the city. Using ArcGIS, a routing model was developed to guide individuals to their nearest safe zone during an earthquake. This approach demonstrates how open-source geospatial data and GIS tools can be leveraged for disaster preparedness in high-risk areas. 1. Title: Open Mapping for Urban Resilience: A Routing Model to Nearest Safe Spaces in Earthquake-Vulnerable Dhaka 2. Submission Type: 20-minute talk 3. Abstract: This presentation introduces a GIS-based routing model developed for earthquake preparedness in Dhaka, Bangladesh. By integrating OpenStreetMap data with satellite imagery, the project identifies open spaces and directs urban populations to the nearest safe zones during seismic emergencies. 4. Description: Dhaka is one of the most densely populated cities in the world, with over 20 million residents living in an area of approximately 306 square kilometers. Despite its rapid urbanization, the city faces a severe shortage of open spaces, such as parks, playgrounds, and public squares, which are critical in times of disaster. The urban landscape is dominated by high-rise buildings and commercial complexes, with little consideration for creating large, accessible public spaces. This makes it increasingly difficult to accommodate the population in emergency situations, particularly during a major earthquake. According to the World Bank, Dhaka’s vulnerability to earthquakes is exacerbated by the lack of proper urban planning and outdated infrastructure [6]. A study by Rahman et al. (2015) highlights that most of Dhaka’s infrastructure, including residential buildings and roads, was not designed with seismic safety in mind, making the city highly susceptible to catastrophic damage in the event of an earthquake [2]. In addition to the lack of open spaces, the city's rapid urbanization has led to insufficient planning for evacuation routes and safety measures. Public awareness regarding earthquake preparedness is low, and emergency response systems are not adequately equipped to handle large-scale disasters. Without well-documented open spaces and efficient evacuation routes, the challenge of evacuating residents to safety during an earthquake becomes significantly more complex. The objective of this research is to develop a spatial framework for earthquake emergency evacuation planning by integrating multiple data sources: OpenStreetMap (OSM) road and building layers, satellite-derived open space data, and spatial analysis tools in ArcGIS. The methodology involves three major steps: 1. Open Space Identification: Open spaces were delineated by classifying high-resolution satellite imagery, verified and cross-referenced with existing land use data and field knowledge. Parks, playgrounds, schoolyards, open fields, and other publicly accessible open areas were identified as potential evacuation zones [2]. 2. Urban Feature Mapping with OSM: OSM was used as the primary source for road networks and building footprints. The completeness and accessibility of OSM data enabled detailed modelling of the urban structure [3], and the data was further cleaned and enriched using local knowledge and field validation. 3. Routing Model Development: Using ArcGIS Network Analyst, a routing model was developed to calculate the shortest and most accessible path from any given building or cluster of buildings to the nearest open space [4]. This model considers actual street networks and real-world constraints. Future versions of the model could also include blocked roads or damaged areas during earthquakes to make the routes more realistic. The significance of this work lies in its practical application: in a high-density city like Dhaka, where open spaces are scarce and poorly documented, a pre-disaster preparedness model like this can be integrated into municipal planning, community training, and mobile-based alert systems [1]. By leveraging open-source geospatial data such as OSM and combining it with remote sensing and network analysis, this approach offers a low-cost, scalable, and replicable solution for other vulnerable cities in the Global South [3]. Moreover, this project highlights the power of community mapping and open data in disaster risk reduction. YouthMappers and local GIS volunteers can play a role in validating OSM features and verifying open space locations [3]. Community participation in data validation enhances the credibility of the model and promotes local ownership of the disaster preparedness process. The proposed presentation will: • Demonstrate the technical workflow from satellite image classification to OSM integratio
Open Mapping for Urban Resilience: A Routing Model to Nearest Safe Spaces in Earthquake-Vulnerable Dhaka (sotm2025)
Behaviour-Based Quality Assessment of OpenStreetMap Data in Data Scarce Area Using Unsupervised Machine Learning (sotm2025)
This study introduces a behavior-dependent, unsupervised machine learning approach to assess the intrinsic quality of OpenStreetMap (OSM) data in Dhaka, which is both data-starved and urbanizing rapidly urbanizing area. Leveraging enriched contributor metadata and Principal Component Analysis (PCA), latent behavioral patterns and segmented contributors identified using KMeans and HDBSCAN. The silhouette score for PCA-based clustering was 0.951. The results show superior interpretability of KMeans over HDBSCAN. This repeatable methodology provides a scalable and reference-free solution to take quality assurance of VGI datasets to the front-line, in cases of limited or no authoritative data. OpenStreetMap (OSM) is an important source of geospatial information in data-starved urban areas, where official geospatial data are scarce, outdated, or are not readily available. Increasing need for current and accurate geospatial data in fast urbanizing and under surveyed regions makes the use of OpenStreetMap (OSM) an essential resource. As one of the most representative Volunteered Geographic Information (VGI), OSM offers a free world map that is editable and can be contributed by millions of people [1]. The tool is an essential component for urban analytics, transport planning, disaster risk reduction, and spatial modeling in the world [2], [3], [4]. Although widely used, the quality of OSM data varies greatly across regions and contributor skill level, and there is no unified, system level quality assurance mechanism [5]. This heterogeneity can be risk inducing for users making use of this data for precision tasks (e.g., routing, land use modeling and infrastructure design) [2], [6]. Traditional OSM quality assessments rely on extrinsic comparisons with satellite imagery or authoritative datasets, which are often unavailable in the very regions that need the data the most [7], [8]. To overcome this challenge, a reproducible, unsupervised machine learning framework propose to assess OSM data quality intrinsically, based on contributor behavior metadata alone. Specifically, Dhaka —a data-scarce and fast-growing megacity in Bangladesh select as a study area—using the hypothesis that distinct contributor behavioral patterns correlate with different levels of data reliability. This behavior-centric perspective leverages the insight that contributor frequency, recency, thematic focus, and spatial editing behavior can serve as meaningful proxies for feature quality [5], [9]. Roads and buildings for Dhaka extracts by using by a.osm.pbf with the Pyrosm library. Then enriched feature vector creates for each unique contributor, composed of (total_edits, edit_rate, active_days, spatial_extent, pct_road, pct_building, weekday_activity, days_since_last_edit). Principal Component Analysis (PCA) applies for dimensionality reduction and shows that PC1 roughly represents global mapping activity, while PC2 corresponds to thematic attention (road versus building), and PC3 represents the geographical coverage of contributions. These observations are supported by a feature contribution heatmap (Figure 1.(a)), which indicates that it is reasonable to consider the behavioral features to be interpretable and highly separable in the component-reduced space. PCA has also the purpose of reducing noise and gets the data ready for clustering [10]. Next, KMeans clustering (with k = 4) and HDBSCAN, a density-based clustering is performed on the PCA-transformed feature set. The silhouette score of the KMeans model was 0.951, suggesting high cohesion within the clusters and good separation between the clusters of behaviors. The PCA cluster scatterplot (Figure 1.(c)) indicates four separated clusters: (1) most participants (Figure 1. (b)) fall in cluster 0, which mainly encompasses casual or one-hit contributors who probably participate in sporadic mapathons, or make large scale imports, (2) cluster 1 and 2 consist of moderate to heavy contributors, who are relatively more or less stable, with richer semantic tagging, and whose edits are spatially distributed, (3) cluster 3 is composed of a small group of “power users,” who are characterized by high activity volume and a large geographical distribution. HDBSCAN also use on the same dataset in order to analyze its capability of separating varies densities in clusters and noise. HDBSCAN found small, dense clusters, and labeled a large percentage of contributors as noise. Although helpful for identifying anomalousness and potential vandalism, HDBSCAN was unable to produce as clear clusters for the main contributors as KMeans, likely because the extreme imbalance in contributor engagement. This benchmarking demonstrated that KMeans comes with a better interpretability and cluster stability, and is therefore preferred for behavioral segmentation at the high volumes of OSM dataset. To further verify the clustering, the changes in edit volume over time per cluster investigated, and calculated feature distr
Behaviour-Based Quality Assessment of OpenStreetMap Data in Data Scarce Area Using Unsupervised Machine Learning (sotm2025)
EUthMappers - learning by teaching mapping (sotm2025)
EUthMappers is an ERASMUS+ initiative that promotes STEAM education in secondary schools throughout the European Union, enhancing students' digital skills and fostering environmental civic engagement. The project includes three universities, students and teachers from five European schools, working for two years. The project has three main phases: development of training materials, local mapping projects and humanitarian mapping collaboration. This presentation outlines the project's implementation steps and showcases the remarkable results achieved by not only participating students but also organizers. EUthMappers is an ERASMUS+ initiative that promotes STEAM education in secondary schools throughout the European Union, enhancing students' digital skills and fostering environmental civic engagement. The project bridges theoretical learning with practical applications using open-source geospatial tools and collaborative mapping on the OpenStreetMap (OSM) platform. The idea is to establish an European mapping network similar to YouthMappers. This presentation outlines the project's implementation steps and showcases the remarkable results achieved by participating students. The project involves three universities (Politecnico di Milano, Universidad Politécnica de Madrid, and Presovska Univerzita V Presov) and five secondary schools located in Italy, Spain, Slovakia, Romania, and Portugal, engaging approximately 160 students in total. The management of the project is done by Euronike, which is an association expert in capacity building activities and EU policies. Until now, the project has been running for two years, and was implemented in three phases. 1. Development of training materials: a comprehensive training package on open geospatial tools and data analysis was written and made available in six languages (English, Italian, Spanish, Portuguese, Slovakian and Romanian). This training was first delivered to teachers, who then transferred the knowledge to their students. 2. Local mapping projects: under the guidance of their teachers and with university collaboration, students developed local mapping initiatives from ideation to implementation, focusing on data acquisition and visualization techniques. Across five European cities, pupils carried out mapping-focused projects that combined local engagement with practical outcomes. In Rovereto (Italy), students identified drainage channels on forest roads in the “Bosco della Città” as a central issue and mapped them using GPS and drone imagery. Since such features had never been mapped on OSM, they engaged with the OSM community to propose and share a new mapping methodology. The resulting maps and database now support both routine and extraordinary maintenance by the Rovereto Forestry Service, helping mitigate hydrogeological risks. In Madrid (Spain), pupils mapped key elements related to mobility, leisure spaces, and climate shelters in the Arganzuela neighborhood. Working in groups, they identified and analyzed public space features, and then proposed concrete improvements to enhance young people’s quality of life, which they presented at local events and forums. In Prešov (Slovakia), students focused on tree mapping in central city parks. After ecological training, they collected detailed data—such as species, trunk width, and height—using practical tools and mobile apps. Their results contributed to urban ecological databases and were showcased to the public during their school’s open-door day. In Lisbon (Portugal), pupils explored how graffiti and street art reflect local identity by mapping urban artworks around their school. Through documentation and critical analysis, they created a record of artistic interventions with social and historical value, culminating in a community event promoting dialogue on youth expression in public space. In Pitești (Romania), students diagnosed low public awareness of recycling facilities and mapped the location, accessibility, and types of waste accepted at collection points. Their interactive map aimed to make recycling easier and more visible, encouraging environmental responsibility across the city. These projects show how pupil-led mapping activities can generate innovative tools, influence local planning, and foster active citizenship. 3. Humanitarian mapping collaboration: as the final activity, students participated in a humanitarian mapping project issued by the United Nations Global Service Centre (UNGSC). There were two workshops about humanitarian mapping and Sustainable Development Goals (SDGs) in order to provide students basic knowledge about humanitarian mapping and their contribution to common goods. Students are then trained on humanitarian mapping simulation projects before actually participating in a real-world scenario proposed by UN Maps, a programme in UNGSC to enhance UN peacekeeping missions operational capabilities through open geospatial information. The training project on
Lightning Talks I (sotm2025)
Lightning Talks I (sotm2025)
## OpenStreetMap Sound Demo _by Taro Matsuzawa (@smellman)_ ## ArcGIS Living Atlas of the World: Open Data Content _by Deniz Karagulle_ Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/P3FWCW/
EUthMappers - learning by teaching mapping (sotm2025)
Intelligent Enough? Evaluating Collective Action in HOT Tasking Manager Mapping Projects (sotm2025)
Intelligent Enough? Evaluating Collective Action in HOT Tasking Manager Mapping Projects (sotm2025)
This talk examines the dynamics of collective intelligence in humanitarian mapping projects coordinated through the HOT Tasking Manager, using a dataset of 746 projects and 312,289 tasks to evaluate participation, collaboration, and evidence of intelligent group behavior. Humanitarian mapping projects coordinated through the Humanitarian OpenStreetMap Team Tasking Manager (HOT-TM) represent a paradigmatic case of large-scale digital collaboration. Yet, while their practical utility in disaster response and preparedness is increasingly evident, the underlying collective dynamics that allow these efforts to function effectively remain under-explored. This talk builds on our recent article published in ACM Transactions on Computer-Human Interaction (TOCHI) [1] that presents a comprehensive analysis of HOT-TM projects through the lens of collective intelligence. Collective Intelligence is defined as “groups of individuals acting collectively in ways that seem intelligent [2].” Following this definition, we structured our analysis around three guiding research questions: (RQ1) What characterizes the group of individuals collaborating in HOT-TM mapping projects?, (RQ2) How is collective action organized within these projects?, and (RQ3) What evidence of intelligent action can be identified in this setting? To answer these questions, we constructed and analyzed a dataset encompassing 746 HOT-TM projects executed between December 2021 and November 2023. The dataset includes 312,289 mapping tasks performed by 38,893 contributors, as well as detailed records of over 1.8 million task states. Additionally, we incorporated spatial information on the area of mapped buildings using data extracted from the OpenStreetMap database. Our analysis proceeds in three stages. First, we profile the mapping community. Results show that the vast majority of contributors are beginners, who typically participate in a single project. However, a small group of highly experienced mappers—classified by HOT-TM as "advanced"—contribute to dozens of projects and assume more complex tasks. Notably, only 29% of contributors declare their country, but among those who do, the majority are based outside the regions affected by the mapping projects (see Figure 1). This reinforces existing concerns about the limited presence of local knowledge in humanitarian Volunteered Geographic Information (VGI) initiatives [3]. Second, we investigate the organization of collective action using process mining techniques applied to task state logs. Most mapping tasks follow a simple trajectory: a task is mapped and then validated without being split or invalidated (see Figure 2). However, tasks that involve higher complexity or suffer errors require more contributors and longer processing times. Roles within the mapping system are clearly stratified: while beginners dominate mapping in simpler projects, advanced mappers take the lead in complex cases and are responsible for nearly all validations. Despite the potential for collaboration in the mapping phase—through the sequential editing of tasks—true interdependence among contributors is limited. Most tasks are executed by a single mapper, and where collaboration occurs, it is often sequential and uncoordinated. This suggests that HOT-TM's microtasking design promotes a form of "collection" rather than "collaboration" [4]. Third, we assess the presence of intelligent group behavior by analyzing task validation outcomes through logistic regression. We find that advanced contributors are significantly more likely to produce validated outputs, especially when working alone. However, involving multiple contributors in a task—especially when they are less experienced—decreases the probability of successful validation. Furthermore, tasks with larger building areas or those requiring extensive validation times are less likely to be validated, suggesting that complexity and ambiguity remain major challenges. These findings highlight a paradox at the heart of humanitarian mapping: although the system succeeds in rapidly mobilizing volunteers to produce useful geographic data, its collective intelligence is unevenly distributed and relies heavily on a small core of experienced contributors. The wisdom of the crowd is therefore not uniformly distributed; rather, it is the wisdom of a few that sustains the productivity and reliability of the system. Moreover, the absence of strong collaborative mechanisms and the limited engagement of local mappers constrain the potential for adaptive and context-aware mapping. We conclude by reflecting on possible design improvements for platforms like HOT-TM. These include: (1) enhancing onboarding and mentorship to accelerate the transition from beginner to advanced contributor; (2) incentivizing meaningful collaboration beyond sequential task handovers; and (3) integrating local knowledge more effectively by prioritizing and rewarding contributions from
Mapping workflows in iD for new, intermediate and advanced mappers (sotm2025)
Mapping in iD is designed to be a welcoming experience that should require little to no required knowledge to get started. Additionally, the editor does also have some additional features up its sleave for more advanced mapping tasks. Regardless if you are a new, intermediate or advanced mapper: it can never hurt to know a trick or two that make your mapping more efficient! This talk will go through common mapping workflows in iD and how they can elevate your mapping experience. There will be sections dedicated for beginners, intermediate and advanced mappers. For starters, it will be shown how one can get up to speed most efficiently with iD through its built-in walkthough and other help functionality. For intermediate mappers, the talk will cover topics such as photo-mapping, efficient use of keyboard shortcuts, integrated quality assurance tools, useful browser extensions, etc. The talk concludes with advanced techniques such as adding custom background imagery or map data. The showcased mapping workflows shall give a good overview of the spectrum of different mapping tasks one might encounter as a mapper on an regular basis, and will also highlight tools other than iD that allow to dive even further into the respective topics. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/H3NM7X/
User testing of AI-assisted mapping tool fAIr (sotm2025)
Mapping workflows in iD for new, intermediate and advanced mappers (sotm2025)
User testing of AI-assisted mapping tool fAIr (sotm2025)
Humanitarian OpenStreetMap Team proposed the fAIr project (https://fair.hotosm.org/) - an open-source AI-assisted mapping tool. This study describes our user testing organized to compare AI-assisted mapping of buildings in the fAIr tool and classic manual mapping of buildings in the JOSM editor without AI assistance. 26 participants took part in the experiment. Efficiency (number of buildings mapped per minute), effectiveness (proportion of buildings mapped correctly), and satisfaction (feedback from participants) were analyzed. Introduction Humanitarian organizations need maps for their activities. However, there is a lack of quality maps in many countries. The Humanitarian OpenStreetMap Team (HOT) is coordinating a global effort for humanitarian mapping, where volunteers create maps of remote locations. They use satellite imagery to search for roads and buildings and draw them into the OpenStreetMap. Despite all the efforts of volunteers, many regions remain insufficiently mapped (Herfort et al., 2021). Artificial intelligence (AI) is already used to analyze satellite imagery. There is thus an opportunity to use AI for humanitarian mapping as well. HOT proposed the fAIr project (https://fair.hotosm.org/) – an open-source AI-assisted mapping tool (HOT, 2025). In a presentation from SOTM 2024, Anna Zanchetta (Zanchetta, 2024) analyzed the performance of fAIr in detecting buildings in different conditions. Our group, located in the Department of Geography, Faculty of Science, Masaryk University, Czechia, is part of the humanitarian mapping community Missing Maps Czechia and Slovakia. We were in contact with the fAIr developers from HOT – Omran Najjar, Kshitij Sharma, and Anna Zanchetta. They asked us if we could organize user testing of the first version of fAIr. The goal was to compare AI-assisted mapping buildings with the fAIr tool and classic manual mapping buildings without AI assistance. The popular JOSM editor, which offers experienced tools for manual building mapping, was used for comparison. Methodology The experiment took place in an online environment during November 2024. The experiment took place in four different sessions, so each participant could choose the day and time that suited them best. Training datasets were prepared for 24 different localities around the globe. 26 participants took part in the experiment – 14 beginners and 12 experienced contributors. The participants were divided into groups A and B, so there were similar numbers of experienced contributors and beginners in both groups. Each participant was assigned one location for mapping in fAIr and one location for mapping in JOSM. Each session was structured as follows: -An explanation of the rules for correctly and incorrectly mapped buildings. -45 minutes: group A mapped in fAIr, group B mapped in JOSM -5 minutes: break -45 minutes: group B mapped in fAIr, group A mapped in JOSM -Participants filled in the feedback questionnaire. Validators evaluated the mapped buildings, and the results were analyzed. Efficiency (number of buildings mapped per minute), effectiveness (proportion of buildings mapped correctly), and satisfaction (feedback from participants) were analyzed. Results Mapping in JOSM (86.08 %) was significantly more accurate than the fAIr tool (78.22 %). Mapping in JOSM (4.23 buildings/min) was significantly faster compared to the fAIr tool (1.97 buildings/min). Results of experienced users and beginners were also compared. Beginners mapped faster in JOSM (2.89 buildings/min) than in fAIr (2.08 buildings/min), but the mapping quality was almost the same in JOSM (79.28 %) and fAIr (80.41 %). Experienced contributors mapped much faster (5.69 buildings/min) and with higher quality (93.45 %) in JOSM than in fAIr (1.84 buildings/min and 75.38 %). Interestingly, it means that beginners have better results of mapping quality (80.41%) and mapping speed in fAIr (2.08 buildings/min) than experienced contributors (75.38 % and 1.84 buildings/min). On the contrary, experienced contributors have much better results in mapping quality (93.45 %) and mapping speed (5.69 buildings/min) in JOSM than beginners (79.28 % and 2.89 buildings/min). We also analyzed the influence of the complexity of mapping locality, e.g., the density of buildings in the mapped site and the quality of the imagery. For easy localities, experienced contributors generally have higher mapped building counts, indicating better fAIr performance on simpler tasks. For complex localities, the number of buildings mapped by experienced contributors drops significantly, suggesting that complex localities pose a challenge even for experienced contributors. A higher number of errors was found in complex localities, where AI likely generated lower-quality building outline designs. Interestingly, for complex localities, beginners mapped more buildings than experienced contributors, which may suggest that they focus on quantity. This could indicate potential quality issues.
Keeping alive OSMTracker (sotm2025)
This is a story about the last 7 years of OSMTracker and how, from an academic space in a Costa Rican public university, we managed to keep alive the app with computer engineering undergrad students. We want to share what we have learned: the challenges and contributions during this process, and how we plan to continue. OSMTracker is one of the oldies free software data capture tools in the OSM ecosystem. Its simplicity, low technical requirements, easy customization, and ability to use the exported data in various applications make it a valuable resource for mappers. The app became part of the methodological resources used by the Laboratorio Experimental de Computación y Comunidades (LabComún) for development of university extension projects together with communities like Erizo Juan Santamaría Informal settlements and Alajuela en Cleta urban cycling collective. Naturally, the usage of OSMTracker in this context showed us improvements needed in the app, and consequently, we contributed with code that were incorporated in the tool. In 2018, nguillaumin, the original developer of the app, transferred the maintenance OSMTracker to LabComún. Since then, LabComún has been managing the challenge of developing free software from a (global south) public university: scarcity of resources, bureaucratic difficulties, and how to align the process of developing software while offering a sustainable educational experience for students. The ongoing focus of development of OSMTracker highlights the importance of engagement of undergrad computer science students in projects that are related to territories and people. This offers a real perspective on how their contributions to software could improve the quality of life. From a technical and academic perspective, the opportunity to be part of a bigger free software and open data community is unique. Finally, we want to open discussion on how to enhance the sustainability of OSMTracker, involving other actors in collaborations and keeping the app useful for thematic mapping projects. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/9HN9SX/
When Metadata Isn’t Enough: Extrinsic Quality Assessment of OSM Using Custom Reference Data (sotm2025)
Keeping alive OSMTracker (sotm2025)
When Metadata Isn’t Enough: Extrinsic Quality Assessment of OSM Using Custom Reference Data (sotm2025)
This talk explores the extrinsic quality of OpenStreetMap data in Brno by comparing the city’s most frequently mapped amenities against a custom, field-collected reference dataset. The findings highlight relatively high attribute accuracy in OSM but reveal gaps in feature completeness, with only about 34.94% of features matched with the reference dataset. OpenStreetMap is a notable example of a database created by volunteers. Due to its open approach to data collection, establishing trust in the data is essential. Three key factors must be considered to evaluate this trust: completeness, correctness, and positional accuracy. The most common method in recent years for assessing large datasets like OpenStreetMap is to examine intrinsic data quality [1-3], which relies on metadata. However, this approach does not allow for a thorough analysis of the mapped features, resulting in only a rough estimate of the data's trustworthiness. To provide a more detailed understanding of the data, extrinsic data quality is evaluated by comparing OpenStreetMap with a reference dataset. This method is effective for evaluating feature completeness. However, when assessing attribute accuracy, a similarly detailed dataset for comparison is often unavailable. In these instances, the evaluator must gather their own reference dataset, which can be expensive and time-consuming. Because of that, previous studies mainly focused on assessing attribute accuracy by utilizing intrinsic data quality. In our study, we focused on assessing the extrinsic data quality of the city of Brno in the Czech Republic. We wanted to know how much we can trust OpenStreetMap in our city and if there is some correlation between attribute accuracy and metadata of the features. A secondary objective was to determine how well ISO 19157 can be used to assess the attribute quality of the OpenStreetMap. We chose the city’s ten most mapped amenity features and gathered a reference dataset for these features. Since we knew what we would be evaluating, we have acquired a dataset perfect for evaluating OpenStreetMap. Thus, there was no need for major compromises in data evaluation. Over the course of several months, we traveled over 1,000 km on foot and gathered a few thousand reference features using the Locus GIS app. Each feature contained a list of evaluated attributes together with a photo of the object for further evaluation. Evaluated amenities were bench, waste_basket, recycling, restaurant, bicycle_parking, cafe, vending_machine, post_box, pub and fast_food – the most mapped node features in Brno. Our assessment of OpenStreetMap's attribute accuracy revealed generally positive results. Several attributes in our sample achieved 100% accuracy, particularly those with boolean values. However, the most significant issues arose with string attributes that lack defined value lists, such as opening_hours. Ultimately, the primary concern identified was the completeness of the data. We found that the completeness of feature occurrence is inadequate; we were only able to match 34.94% of all reference features with those in OpenStreetMap. This completeness varied significantly across evaluated amenities, with waste_baskets and benches being notably underrepresented. We also assessed the positional accuracy of the data. Each amenity was evaluated separately, revealing average positional errors ranging from 2.63 meters to 4.02 meters. The median error was found to be between 1.83 meters and 3.13 meters. Overall, OpenStreetMap appears to be a relatively accurate positional database for the city of Brno, despite some isolated deviations (outliers). When examining the relationship between attribute accuracy and feature metadata, we assumed that more users editing a feature would lead to more accurate data. This concept is known as the “many eyes principle” [4, 5]. However, the correlations between metadata (such as the number of contributors, versions, and days since the last edit) and attribute correctness are typically not statistically significant. As a result, no explicit dependency can be determined, and no clear patterns emerge from the statistically significant values. Our work also shows that evaluating OpenStreetMap using ISO 19157 can be problematic because this standard does not consider multiple correct values or the varying degrees of attribute correctness (“level of detail” of attributes). Additionally, automating the evaluation of specific attributes, such as opening_hours, is challenging since these attributes can contain different yet correct values. Furthermore, missing or incomplete documentation significantly impacts evaluation, as there should be clear rules indicating which values are correct or not. However, achieving this is difficult for projects that rely on the folksonomy principle, which encourages users to create new values that often lack documentation. OpenStreetMap offers a comprehensive database, but its data quality varies signific
OSMlanduse: A dataset of European Union land use at 10 m resolution derived from OpenStreetMap and Sentinel-2 (sotm2025)
OSMlanduse is the first EU-wide 10 m land-use map integrating 3.2 million OpenStreetMap geometries with Sentinel-2 imagery through an open deep-learning workflow. Delivering CORINE-level thematic detail with finer spatial resolution, it achieves 89% accuracy, providinng wall-to-wall coverage while retaining OSM’s sub-metre detail where available. Released under open licences with reproducible scripts, it supports applications from climate modelling and biodiversity surveys to urban planning and policy monitoring. By uniting crowdsourced mapping and Earth observation, OSMlanduse demonstrates a scalable, transparent approach to producing reliable, high-resolution land-use information at continental scale. Spatially resolved information on present-day land use (LU) is fundamental for climate-mitigation tracking, food-system monitoring and spatial planning, yet Europe still relies on inventories such as CORINE Land Cover (CLC) that are updated quinquennially at 100 m and under-represent the urban micro-mosaic. Meanwhile, new 10 m remote-sensing maps excel at spectral land-cover separation but lack the thematic richness contributed by citizens through OpenStreetMap (OSM). We introduce OSMlanduse, the first European Union-wide LU map at 10 m resolution that fuses 3.2 million OSM geometries with Sentinel-2 multispectral composites by means of a completely open workflow. The product supplies CLC-level thematic detail while matching the spatial grain of Copernicus imagery. Our objective was to demonstrate that globally recurrent, freely licensed data streams can be combined through deep learning to overcome the spatial incompleteness of volunteered geographic information. We converted the March 2020 OSM planet snapshot into 13 CLC classes, directly labelling 61.8 % of EU28 territory. For the unlabeled remainder we trained country-specific residual convolutional neural networks (ResNets) on medoid composites of cloud-masked Sentinel-2 top-of-atmosphere reflectance, then mosaicked predictions and original labels into a seamless raster–vector hybrid. Accuracy was assessed with 4 616 stratified reference points interpreted independently on sub-metre Bing and Google imagery. The map attains 89 % overall accuracy (95 % CI ± 2 %); producer’s accuracies range from 77 % (shrub/herbaceous vegetation) to 99 % (water bodies), while user’s accuracies exceed 93 % for agricultural strata. Most confusion arises between spectrally similar urban greens and semi-natural grasslands, or between early construction sites and bare soil, reflecting both spectral ambiguity and occasional tag noise. OSMlanduse inherits sub-metre geometric detail wherever OSM mapping is dense—Dutch canal parcels, German allotment gardens, Romanian farmyards—yet guarantees wall-to-wall coverage through 10 m raster infill. GeoTIFF tiles, training rasters and the OSM-to-CLC translation table are released under CC-BY 4.0 (DOI: 10.11588/data/IUTCDN) and visualised at https://osmlanduse.org; GPL-3.0 scripts ensure full reproducibility. Three methodological insights emerge. First, the current density and thematic granularity of OSM LU tags suffice to train deep networks that generalise across divergent biogeographic regions without external annotation campaigns. Second, multi-temporal medoid compositing plus per-country modelling dampens atmospheric noise and phenological divergence, enabling continental consistency from uncalibrated top-of-atmosphere data. Third, open-science principles—public code, permissive licences, cloud execution—place high-resolution mapping within reach of resource-constrained institutions. The dataset opens new avenues for investigating LU dynamics, from crop-rotation detection and peri-urban sprawl quantification (SDG 11.3.1) to habitat-specific sampling frames for biodiversity surveys and downscaling of economic statistics. Moreover, its billions of labelled pixels address the chronic scarcity of public training corpora highlighted by recent computer-vision studies. Limitations persist. Crowdsourced tags remain temporally asynchronous; the March 2020 snapshot necessarily precedes pandemic-era peri-urban expansion. Spectral ambiguity endures in arid shrublands and gravel pits, and the 10 m grid misses features narrower than one pixel such as hedgerows. Nevertheless, by releasing not only the final map but all processing scripts under GPL-3.0 we invite replication, auditing and regional adaptation. Training was performed on the FAO SEPAL cloud using Nesterov-accelerated Adam, batch size 64 and early stopping on 20 % held-out OSM tiles per country. Managing the 1.2 TB Sentinel-2 archive through orbit-wise partitioning and raster caching allowed execution of the full continental workflow within 96 GPU-hours on commodity instances. Sampling 5 000 patches per class and tile balanced the abundant yet noisy training labels while preserving rare categories such as mines and wetlands. While earlier studies either harvested OSM tags to create fragme