Impact
Adoption and outreach of the Urban Toolkit, by year. Each year runs from September to August. The numbers are collected again every time the site is published.
| Metric | Y1 Sep 2024 to Aug 2025 | Y2 Sep 2025 to Aug 2026 | Y3 (so far) Sep 1 to Oct 6, 2026 |
|---|---|---|---|
| CI Software Ecosystem | |||
| [1] | 13 | 4 | 1 |
| [2] | 415 | 2,275 | 7,416 |
| [3] | 92 | 134 | 140 |
| Cloud Environment | |||
| Users at large[4] | 0 | 0 | 0 |
| External cloud deployments[5] | 0 | 0 | 0 |
| Urban Data | |||
| Curated datasets[6] | 1 | 4 | 0 |
| Community metrics | |||
| Hackathon attendance (per hackathon)[7] | 0 | 30Interdisciplinary transportation workshop | 0 |
| Workshop attendance (per workshop)[8] | 0 | 30Interdisciplinary transportation workshop | 0 |
| Use cases[9] | 5 | 20 | 1 |
| Internship projects[10] | 0 | 5 | 0 |
| Tutorials[11] | 0 | 1 | 0 |
| Courses[12] | 0 | 2 | 0 |
| Scientific Impact | |||
| Publications[13] | 3 | 5 | 0 |
What is counted
[1] External GitHub contributors
People with at least one commit that year, on any branch, to the 9 repositories listed under the row. They are external when no commit email, team listing or GitHub profile places them at University of Illinois Chicago, New York University and University of Washington. Bots are not counted.
Sources: Git history of the repositories, GitHub REST API: commits, GitHub REST API: users, Team page
[2] Package downloads (average per month)
PyPI downloads of utk and utk-curio, from ClickHouse's public PyPI dataset, and npm downloads of the 7 Autark packages. The average per month divides a year's downloads by its months, counting the elapsed days of a partial month.
Sources: ClickHouse's public PyPI dataset, npm download counts API
[3] GitHub stars (total)
All the stars the 9 repositories had at the end of the year, by the dates GitHub gives them.
Sources: GitHub GraphQL API: stargazers
[4] Users at large
Accounts registered on Curio's hosted instances (flow.urbantk.org and dev.urbantk.org), without the shared guest account, as each instance's public monitor reports them. A year shows the last count taken in it.
Sources: Curio monitor, flow.urbantk.org, Curio monitor, dev.urbantk.org
Y3 (so far)
- dev.urbantk.org0 registered accounts on 2026-10-06
[5] External cloud deployments
Reported by the team.
Sources: The team's figures, in impact.yaml
[6] Curated datasets
Data the team created or curated for its papers, not data downloaded as is: the data releases linked from the project pages, in the year of their paper, and the Data Catalog datasets and data files read by the Curio examples that reproduce a paper's use case, in the year they were added to Curio. Boundaries, samples and OpenStreetMap extracts are not counted.
Sources: Data releases on the project pages, Curio's Data Catalog, Curio's example data, Curio's example gallery
Y1
- Global Shadow DatasetDeep Umbra, Deep Umbra paper
Y2
- Milan Census Polygons (over 65)Curio Data Catalog, read by example 09
- Milan ERA5 Hourly WeatherCurio Data Catalog, read by example 09
- Milan Mean Radiant TemperatureCurio Data Catalog, read by example 09
- niteroi_lst_verao_2001_2024.tifCurio example data, read by example 08
[7] Hackathon attendance (per hackathon)
Reported by the team. Attendance is the mean per hackathon.
Sources: The team's figures, in impact.yaml
Y2
- Interdisciplinary transportation workshopHackathons on real-world transportation and urban data, 30 attendees
[8] Workshop attendance (per workshop)
Reported by the team. Attendance is the mean per workshop.
Sources: The team's figures, in impact.yaml
Y2
- Interdisciplinary transportation workshopUIC with the University of Toronto and the Chicago Metropolitan Agency for Planning, 30 attendees
[9] Use cases
The urban use cases in the usage-scenario or case-study sections of the papers, in the year of the paper, and the examples in Curio's gallery that are neither feature demos nor one of those use cases, in the year their walkthrough was added.
Sources: Curio's example gallery, Papers, read from their HTML versions on arXiv
Y1
- Expert-in-the-loop urban accessibility analysesCurio paper
- What-if scenario planningCurio paper
- Visual analytics of heterogeneous dataCurio paper
- Sunlight access in Parisian parksDeep Umbra paper
- Analyzing land use-level street shadowsDeep Umbra paper
Y2
- Sidewalk accessibility (Project Sidewalk, Chicago)Curio example 01, Vega-Lite chained transforms
- Chicago green roofsCurio example 02, Vega-Lite spatial density
- Chicago speed-camera violationsCurio example 03, Vega-Lite linked temporal charts
- Chicago red-light violationsCurio example 04, Vega-Lite multi-flow dashboard
- Chicago building energy useCurio example 05, Vega-Lite multi-view drilldown
- Chicago Lincoln Park greenery auditCurio example 10, Street-level computer vision
- Cross-height spatial comparison in building facadesOcclusion-Free Conformal Lensing for Spatiotemporal Visualization in 3D Urban Analytics paper
- Cross-season temporal comparison in building facadesOcclusion-Free Conformal Lensing for Spatiotemporal Visualization in 3D Urban Analytics paper
- Exploration of thematic data using direct queriesA Neural Field-Based Approach for View Computation & Data Exploration in 3D Urban Environments paper
- Viewpoint discovery using inverse queriesA Neural Field-Based Approach for View Computation & Data Exploration in 3D Urban Environments paper
- Quantifying visibility from buildingsA Neural Field-Based Approach for View Computation & Data Exploration in 3D Urban Environments paper
- Analyzing access with Project SidewalkUrbanite paper
- Topology exploration of social media dataUrbanite paper
- Impacted houses in flooding simulationUrbanite paper
- Analyzing sidewalk accessibility dataStreetWeave paper
- Detailed sidewalk accessibility analysisStreetWeave paper
- Exploring crime and 311 service dataStreetWeave paper
- Serverless remake of Urbane Autark paper
- Accumulated shadows in ChicagoAutark paper
- Heat islands in NiteróiAutark paper
Y3 (so far)
- Summer shadows in SCOUT's Chicago Loop example, with SCOUT's Deep Umbra modelCurio example 24, SCOUT building rasters
[10] Internship projects
Reported by the team.
Sources: The team's figures, in impact.yaml
Y2
- Curio for street-level data collection (SideSeeing)Undergraduate intern, University of São Paulo
- Interaction provenance in CurioPhD intern, Universidade Federal Fluminense
- Sidewalk Stewards evaluationMaster's intern, Federal University of Pernambuco
- Research visitPhD intern, University of Birmingham
- Curio redevelopment and deploymentThree undergraduates, UIC Early Access to Paid Internships Program
[11] Tutorials
Reported by the team.
Sources: The team's figures, in impact.yaml
Y2
- Autark in Action: Building Serverless Prototypes of Urban Visual Analytics SystemsSIBGRAPI 2026, Tutorials Track
[12] Courses
Reported by the team.
Sources: The team's figures, in impact.yaml
Y2
- CS 424: Visualization & Visual AnalyticsUIC, Fall 2025, 61 students, with Curio
- CS 524: Big Data Visualization & AnalyticsUIC, Spring 2026, 22 students, with Curio
[13] Publications
Papers on the Papers page, in the year of their issue. Preprints are not counted.
Sources: Papers
Y1
- Curio: A Dataflow-Based Framework for Collaborative Urban Visual AnalyticsIEEE VIS 2024
- Deep Umbra: A Generative Approach for Sunlight Access Computation in Urban SpacesIEEE Transactions on Big Data
- Assessing the landscape of toolkits, frameworks, and authoring tools for urban visual analytics systemsComputers & Graphics
Y2
- Occlusion-Free Conformal Lensing for Spatiotemporal Visualization in 3D Urban AnalyticsIEEE VR 2026
- A Neural Field-Based Approach for View Computation & Data Exploration in 3D Urban EnvironmentsIEEE Transactions on Visualization and Computer Graphics
- Urbanite: A Dataflow-Based Framework for Human-AI Interactive Alignment in Urban Visual AnalyticsIEEE VIS 2025
- VA-Blueprint: Uncovering Building Blocks for Visual Analytics System DesignIEEE VIS 2025
- StreetWeave: A Declarative Grammar for Street-Overlaid Visualization of Multivariate DataIEEE VIS 2025
Notes
- Years. Each column counts its year on its own, except GitHub stars and users at large, which are the totals at the end of the year. The current year runs to the last update.
- Code. Every deploy of the site collects and computes all of these numbers again with these two files. Collector: scripts/impact/collect.mjs, Computation: site/.vitepress/theme/node/impact.ts