Source: Analyze Boston (data.boston.gov) Platform: CKAN-powered open data hub Total Datasets: 451+ License: Open Data Commons Public Domain Dedication (PDDL) Last researched: 2026-03-07
All datasets are accessible via the CKAN API:
# List all datasets
GET https://data.boston.gov/api/3/action/package_list
# Get dataset metadata
GET https://data.boston.gov/api/3/action/package_show?id={dataset-name}
# Query data directly (SQL-like)
GET https://data.boston.gov/api/3/action/datastore_search?resource_id={resource_id}
# SQL queries against data
GET https://data.boston.gov/api/3/action/datastore_search_sql?sql=SELECT * FROM "{resource_id}" LIMIT 10
Common formats: CSV, GeoJSON, KML, SHP, ArcGIS GeoServices REST API, XLSX
- What: Every constituent service request since 2011 — potholes, streetlights, trash, trees, sidewalks, noise, code enforcement, etc.
- Resource IDs:
- NEW SYSTEM:
254adca6-64ab-4c5c-9fc0-a6da622be185 - 2026:
1a0b420d-99f1-4887-9851-990b2a5a6e17 - 2025:
9d7c2214-4709-478a-a2e8-fb2020a5bb94 - 2024:
dff4d804-5031-443a-8409-8344efd0e5c8
- NEW SYSTEM:
- Note: Transitioning to new backend system (Oct 2025 - mid 2026). Different case types in different datasets during migration.
- Hackathon potential: Response time analysis, neighborhood equity, predictive maintenance, seasonal pattern detection
- What: BPD incident reports — type, time, location
- System: Mark43 RMS Database
- Formats: CSV by year, offense code reference in XLSX
- Hackathon potential: Safety mapping, hotspot prediction, temporal pattern analysis, resource allocation optimization
- What: Traffic crashes requiring 911 response — date, time, location, type
- Goal: Eliminate fatal/serious crashes by 2030
- Format: CSV (~5.4 MB), updated monthly
- Contact: emailems@bostonems.org
- Hackathon potential: Dangerous intersection identification, pedestrian/cyclist safety tools, infrastructure improvement prioritization
- What: Property ownership, value, classification, and tax assessment data
- Fields: PID, CM_ID, GIS_ID, ZIPCODE, occupancy codes, assessed values
- Updated: Annually
- Hackathon potential: Housing affordability analysis, gentrification tracking, tax equity studies
- What: Health inspections, violations, and compliance records
- Updated: Daily
- Hackathon potential: Food safety transparency apps, restaurant health scores, violation pattern analysis
- Views: 517 (popular)
- Hackathon potential: Development tracking, neighborhood change analysis, construction impact prediction
- Views: 1,323 (most popular)
- Hackathon potential: Government transparency, salary equity analysis, department spending
| Dataset | Description |
|---|---|
| blue-bikes-system-data | Bikeshare trip data |
| blue-bike-stations | Station locations |
| existing-bike-network-2024 | Current bike infrastructure |
| parking-meters | Meter locations and data |
| traffic-signals | Signal locations |
| street-sweeping-schedules | Sweeping routes/times |
| snow-emergency-parking | Snow parking rules |
| snow-emergency-routes | Emergency snow routes |
| moving-truck-permits | Moving permit data |
| Dataset | Description |
|---|---|
| greenhouse-gas-emissions | City emissions data |
| tree-canopy-change-assessment | Urban tree coverage changes |
| bprd-trees | Parks & street tree inventory |
| rainfall-data | Precipitation records |
| climate-ready-boston-sea-level-rise-inundation | Flood risk maps |
| 9/21/36inch-sea-level-rise-* | Multiple sea level rise scenarios |
| climate-ready-boston-social-vulnerability | Vulnerable populations + climate |
| our-progress-toward-carbon-neutrality | Carbon goals tracking |
| berdo-reporting-* | Building emissions disclosure data |
| Dataset | Description |
|---|---|
| income-restricted-housing | Affordable housing inventory |
| short-term-rental-eligibility | Airbnb/rental eligibility |
| rentsmart-boston | Rental market data |
| article80-development-projects | Major development pipeline |
| building-and-property-violations1 | Code violations |
| residential-displacement-risk-map-scores | Displacement risk scoring |
| zoning-reform-impact-tracker | Zoning change impacts |
| Dataset | Description |
|---|---|
| shootings | Shooting incident data |
| shots-fired | Shots fired reports |
| boston-police-department-fio | Field interrogation/observation |
| fire-incident-reporting | Fire incident data |
| homicide-clearance-rate | Case clearance tracking |
| boston-emergency-services-team-best | Emergency services |
| Dataset | Description |
|---|---|
| 2024-digital-equity-survey | Digital access/equity |
| neighborhood-demographics | Neighborhood population data |
| certified-business-directory | Minority/women-owned businesses |
| women-owned-businesses | Women-owned business registry |
| wicked-free-wi-fi-daily-connections | Public WiFi usage |
| wicked-free-wifi-locations | Free WiFi locations |
| cityscore | Overall city performance metrics |
| Dataset | Description |
|---|---|
| streetlight-locations | All streetlight locations |
| fire-hydrants | Hydrant locations |
| sidewalk-inventory | Sidewalk conditions |
| pedestrian-ramp-inventory | ADA ramp inventory |
| public-works-active-work-zones | Active construction zones |
| charging-stations | EV charging locations |
| community-centers | Community center locations |
| public-libraries | Library locations |
Boston has plenty of dashboards. What's missing is cross-dataset intelligence — combining multiple data sources to predict problems before they happen and recommend specific actions.
The city publishes 451+ datasets but treats them as isolated silos. The hackathon opportunity is in the connective tissue between datasets.
| Layer | Status | Example |
|---|---|---|
| Descriptive (what happened) | SATURATED | CityScore, Vision Zero map, 311 data |
| Diagnostic (why it happened) | PARTIAL | Some journalist analysis, academic papers |
| Predictive (what will happen) | SPARSE | Fire risk model (Open Data Challenge), 311 ML routing |
| Prescriptive (what to do about it) | ALMOST NONEXISTENT | No public tools |
The 311 schema is the key linkage point. Each record has: neighborhood, city_council_district,
police_district, fire_district, public_works_district, ward, precinct, latitude,
longitude, zip_code. This means 311 data can be joined to virtually every other dataset.
| Dataset A | Dataset B | Join Key | Prediction Unlocked |
|---|---|---|---|
| 311 requests | Property assessment | address/parcel ID | Property condition → service demand |
| 311 requests | Building violations | address/sam_id | Violation patterns → future complaints |
| 311 requests | Crime incidents | lat/lng + time | Safety conditions → service demand |
| Vision Zero crashes | Bike network + sidewalks | lat/lng | Infrastructure gaps → crash prediction |
| Building violations | Fire incidents | property_id/sam_id | Code violations → fire risk (proven in Open Data Challenge) |
| Property assessment | Displacement risk | census block group | Value trends → displacement prediction |
| 311 requests | Weather/rainfall | date | Weather → infrastructure failure prediction |
| Food inspections | 311 complaints | address | Complaint patterns → inspection failure prediction |
| Building permits | 311 requests | neighborhood + time | Development activity → neighborhood change prediction |
| BERDO emissions | Property assessment | address | Energy profile → building health prediction |
The strongest hackathon concepts follow this pattern:
- Cross-reference 2+ datasets that the city treats as separate
- Predict a specific, measurable outcome
- Recommend a concrete action to a specific audience (resident, city dept, council member)
- Validate with historical data (train on 2020-2024, test on 2025)
The prediction: Given a neighborhood + time of year + recent weather + property age profile + recent building permits, predict the volume and type of 311 requests for the next 30 days.
Cross-referenced datasets:
- 311 historical requests (2011-2026) — target variable
- Property assessment — building age, type, value as features
- Building violations — leading indicator of future complaints
- Rainfall data — weather-driven infrastructure stress
- Building permits — construction activity disruption
- Public works active work zones — ongoing disruption context
The recommendation engine:
- For city departments: "Dorchester 02121 will see a 40% spike in pothole reports in the next 3 weeks based on freeze-thaw patterns and road age. Pre-deploy crews."
- For residents: "Your neighborhood historically sees [X] issues at this time of year. Here's what to watch for and how to report early."
- For council members: "Your district's top 3 emerging issues based on leading indicators."
Why it's novel: The city's 311 ML classifies and routes existing requests. Nobody predicts future requests from cross-dataset signals. The Open Data Challenge fire risk projects proved multi-dataset prediction works — this extends the approach to all 311 categories.
Validation approach: Train on 2020-2023, validate on 2024, test on 2025. 311 data has 15 years of history — excellent for seasonal and trend modeling.
The prediction: A composite, forward-looking "health score" for every census block group that predicts quality-of-life trajectory over the next 12 months.
Cross-referenced datasets:
- 311 request trends (volume, type, response time) — service delivery signal
- Crime incident trends — safety trajectory
- Building violations — physical environment degradation
- Property assessment year-over-year — economic trajectory
- Building permits — investment/development signal
- Displacement risk scores — housing stability
- Food establishment inspections — commercial corridor health
- Tree canopy data — environmental quality
The recommendation engine:
- For residents: "Your block's health score is trending down. The leading indicators are [building violations up 30%, tree canopy declining]. Here are 3 things you can do: [report issues, attend council meeting on X date, connect with neighborhood association]."
- For city planners: "These 15 block groups show early-warning signals of decline. Intervention now costs X; intervention in 2 years costs 5X."
- For community organizations: "Your service area's biggest emerging needs are [X, Y, Z] based on leading indicators."
Why it's novel: CityScore tracks city-wide performance. The Displacement Risk Map tracks one dimension (housing). Nobody combines ALL signals into a block-level trajectory with forward-looking recommendations. This is the "credit score for neighborhoods" — a single number backed by transparent, multi-source data.
The prediction: For any walking/cycling route in Boston, predict the real-time risk score based on historical crashes + infrastructure condition + time of day + seasonal patterns.
Cross-referenced datasets:
- Vision Zero crash records — historical risk by location
- Sidewalk inventory — infrastructure quality
- Pedestrian ramp inventory — ADA accessibility
- Streetlight locations — nighttime visibility
- Traffic signals — intersection control
- Public works active work zones — temporary hazards
- 311 requests (sidewalk, streetlight categories) — maintenance status
- Bike network — cycling infrastructure availability
The recommendation engine:
- "Your route from A to B passes through 2 high-risk intersections. Here's an alternative that adds 3 minutes but reduces risk by 60%."
- "This intersection has had 12 crashes in 3 years, mostly at dusk. If you must cross here, use the crosswalk on the north side (signalized) rather than the south (unsignalized)."
- For city transportation dept: "These 10 intersections have the highest predicted crash rates for the coming season based on historical patterns + current infrastructure condition."
Why it's novel: Vision Zero maps where crashes happened. This predicts where they're likely to happen and routes around them. Adding 311 sidewalk/streetlight complaints as a real-time infrastructure condition layer is something nobody does.
The prediction: Given a restaurant's inspection history + violation patterns + 311 complaint trajectory + neighborhood food safety trends, predict the probability of a serious violation at the next inspection.
Cross-referenced datasets:
- Food establishment inspections (2006-present) — historical outcomes
- 311 complaints (food/sanitation categories) — early warning signals
- Active food establishment licenses — business status
- Building violations — physical plant condition
- Neighborhood demographics — contextual factors
The recommendation engine:
- For diners: "This restaurant's predicted safety score is 87/100. Trend: improving. Last 3 inspections show declining violations. Similar restaurants in this area average 79."
- For inspectors: "These 20 establishments have the highest predicted risk of failing their next inspection. Prioritize these for proactive visits."
- For restaurant owners: "Based on your inspection history and comparable businesses, your top 3 risk areas are [X, Y, Z]. Addressing these would improve your predicted score by N points."
Why it's novel: The city grades restaurants A/B/C after inspection. DrivenData ran a prediction competition, but the results weren't productized. Nobody provides a forward-looking risk score or gives restaurants prescriptive guidance to improve.
The prediction: When one infrastructure system degrades (e.g., water main, road surface, tree damage), predict the cascade of secondary failures and 311 requests that will follow.
Cross-referenced datasets:
- 311 historical requests — temporal sequences of related complaints
- Building violations — structural degradation signals
- Fire incidents — infrastructure failure outcomes
- Public works active work zones — current disruptions
- Streetlight locations + 311 streetlight complaints — electrical infrastructure
- Tree data + 311 tree complaints — urban canopy stress
- Rainfall data — weather triggers
- Property assessment — building age and condition proxy
The recommendation engine:
- "A water main break on X Street historically triggers: 3 pothole reports within 2 weeks, 1 building violation within 30 days, and a 15% increase in rodent complaints within 60 days in a 2-block radius. Recommended: pre-inspect adjacent properties and schedule pavement assessment."
- For public works: "This area has 3 concurrent infrastructure stress signals. Historical pattern suggests a cascade event is likely. Here's the predicted sequence and recommended prevention plan."
Why it's novel: Everyone treats each 311 category independently. Nobody models the cascade relationships between infrastructure systems. This is genuinely new analytical territory — the 15 years of 311 data make temporal sequence mining feasible.
| Rank | Idea | Datasets Crossed | Prediction Clarity | Recommendation Value | Feasibility |
|---|---|---|---|---|---|
| 1 | NextCall (311 prediction) | 6 | HIGH — clear target variable | HIGH — actionable for 3 audiences | HIGH — rich historical data |
| 2 | BlockHealth (neighborhood score) | 8 | MEDIUM — composite score | VERY HIGH — broadest impact | MEDIUM — complex aggregation |
| 3 | PlateScore (restaurant risk) | 5 | HIGH — clear target variable | HIGH — consumer + inspector value | HIGH — well-defined problem |
| 4 | SafeStep (pedestrian routing) | 8 | MEDIUM — route-level risk | HIGH — directly usable | MEDIUM — routing engine needed |
| 5 | Infrastructure Cascade | 8 | AMBITIOUS — sequence prediction | VERY HIGH if it works | LOW-MEDIUM — needs careful temporal analysis |
Problem: Are city services equitably distributed across neighborhoods? Data: 311 response times + neighborhood demographics + property assessment + cityscore Approach: Cross-reference 311 response times by neighborhood with demographic and income data to surface systemic inequities.
WHAT ALREADY EXISTS:
- Boston Globe analyzed 311 response times by neighborhood (Jan 2026), finding South End fastest at 2.2hr median
- CityScore dashboard tracks overall city performance metrics but does NOT break down equity across neighborhoods
- The city's Analytics Team builds internal dashboards but no public-facing equity comparison tool exists
- The Open Data Challenge had a "Making Open Data Local" track for neighborhood-level tools
NOVELTY ASSESSMENT: MEDIUM-HIGH — Response time data is public and has been journalistically analyzed, but no persistent, interactive equity dashboard exists that cross-references service delivery with demographics. The gap is the systematic, ongoing equity lens rather than one-off reporting.
DIFFERENTIATION ANGLE: Go beyond 311 response times. Combine CityScore, 311, property assessment, and demographic data into a composite "neighborhood equity index" that updates automatically. Frame around actionable policy recommendations, not just visualization.
Problem: Pedestrians and cyclists don't know which intersections are most dangerous. Data: Vision Zero crashes + bike network + traffic signals + sidewalk inventory + pedestrian ramps Approach: Route-planning tool that factors in crash history, infrastructure quality, and real-time construction zones.
WHAT ALREADY EXISTS:
- Vision Zero Boston — city program with crash data, safety concern reporting, and an interactive ArcGIS map
- Boston Region MPO Vision Zero Action Plan — regional crash data dashboard showing where crashes occur, who's involved, and causes
- Smart Streets program — city infrastructure redesign initiative
- Crowdsourced safety concern map had 11,000+ citizen reports within 6 months of launch
- Neighborhood Slow Streets program for residential speed reduction
NOVELTY ASSESSMENT: LOW-MEDIUM — This space is well-covered by the city. Vision Zero already has mapping, reporting, and data dashboards. The MPO has a regional data dashboard. The missing piece is route planning — existing tools show where crashes happen but don't help you plan a safer route from A to B.
DIFFERENTIATION ANGLE: If pursuing this, focus narrowly on the routing aspect — a Google Maps-like tool that suggests the safest walking/cycling route, not just the fastest. Integrate crash history, sidewalk quality, ADA ramp availability, lighting (streetlight data), and active construction zones into a route score. This doesn't exist yet.
Problem: Residents don't understand their personal flood/climate risk. Data: Sea level rise projections (9/21/36 inch scenarios) + social vulnerability + property assessment + tree canopy Approach: Address-level climate risk scoring combining flood maps, social vulnerability, and property data.
WHAT ALREADY EXISTS:
- Climate Ready Boston — comprehensive city initiative with vulnerability assessment and resilience planning
- Boston Harbor Flood Risk Model (BH-FRM) — dynamic flood model with sea level rise + storm surge simulations
- MAPC Climate Vulnerability Index — regional mapping tool showing neighborhood vulnerability combining sociodemographic, health, housing, and climate data
- Green New Deal Data Dashboard — tracks progress on climate goals by neighborhood
- MA Sea Level Rise and Coastal Flooding Viewer — statewide flood projections
- BWSC has citywide flooding scenario models
- Our Progress Toward Carbon Neutrality — Analyze Boston showcase
NOVELTY ASSESSMENT: LOW — This is the most saturated space of all seven ideas. Multiple city, regional, and state tools already provide flood risk mapping, climate vulnerability analysis, and neighborhood-level climate data. The MAPC tool specifically combines social vulnerability with climate exposure.
DIFFERENTIATION ANGLE: The only gap is a personalized, address-level "climate report card" that combines ALL of these sources into one simple score with plain-language recommendations (e.g., "Your property has a 23% chance of flooding by 2050. Here's what you can do."). Existing tools are expert-oriented; a consumer-friendly synthesis could add value. But the bar is high given existing work.
Problem: Diners can't easily see restaurant inspection histories. Data: Food establishment inspections + active food licenses + 311 complaints Approach: Consumer-facing app with inspection scores, violation history, and trend analysis for any Boston restaurant.
WHAT ALREADY EXISTS:
- Boston's restaurant grading system — A/B/C letter grades (A=94-100, B=81-93, C=80 or below)
- Mayor's Food Court — online database of food service businesses and inspection results
- Grades are posted on boston.gov after inspection
- Inspectional Services conducts annual inspections + complaint-driven follow-ups
- A DrivenData competition previously ran a "Keeping it Fresh: Predict Restaurant Inspections" challenge using Boston data
NOVELTY ASSESSMENT: MEDIUM — Boston already has a grading system and online database, but the Mayor's Food Court interface is dated and not mobile-friendly. The raw inspection data on Analyze Boston is much richer than what's exposed through the city's consumer-facing tools. The DrivenData competition proves the ML angle has been explored.
DIFFERENTIATION ANGLE: Build a modern, mobile-first experience that goes beyond the current letter grade. Show violation history trends (is this restaurant getting better or worse?), comparison to neighborhood averages, and predictive risk scoring. Integrate 311 complaint data for a fuller picture. Think "Yelp health layer" rather than just a grade lookup.
Problem: Residents in gentrifying areas need advance warning of displacement risk. Data: Residential displacement risk scores + property assessment trends + building permits + income-restricted housing + short-term rental data Approach: Neighborhood-level displacement risk tracker combining multiple leading indicators.
WHAT ALREADY EXISTS:
- Anti-Displacement Action Plan — official city plan
- City's Displacement Risk Map — dynamic mapping tool at Census Block Group level, assessing risk based on renter vulnerability (race, poverty, cost burden, education), proximity to amenities (transit, businesses), and market changes (rent appreciation, development)
- Modeled after Seattle and Portland displacement risk tools
- Residential Displacement Risk Map Scores — the underlying data is on Analyze Boston
- RentSmart Boston — rental market data showcase on Analyze Boston
- Zoning Reform Impact Tracker — GIS tool showing zoning change impacts
- Academic research: Preis et al. (2021) published methodology comparing four displacement measurement approaches using Boston data
NOVELTY ASSESSMENT: LOW — The city has already built essentially this exact tool. The Displacement Risk Map provides block-group-level risk scoring with the same methodology our idea proposed. The underlying data is publicly available. Academic research has explored the methodology.
RECOMMENDATION: SKIP or SIGNIFICANTLY PIVOT. If interested in this space, focus on the tenant action side — what can a resident DO when they're in a high-risk area? Connect displacement risk data to actual resources: legal aid, housing programs, tenant organizing, income-restricted housing availability. The gap is actionable response, not risk identification.
Problem: City reacts to problems instead of preventing them. Data: 311 historical requests + weather/rainfall + streetlight locations + sidewalk inventory + tree data Approach: ML model predicting where 311 requests will spike, enabling proactive maintenance.
WHAT ALREADY EXISTS:
- Boston's new 311 platform is AI-native with ML-powered request classification and "next best action" recommendations
- The system uses ML to detect patterns (e.g., rodent infestations near garbage complaints, snow complaints triggering removal offers)
- 311 Technology Modernization Project — active modernization underway (Spring 2025 - mid 2026)
- Cartegraph — Boston's asset management platform tracking hundreds of thousands of city assets (signs, streetlights, pavement, trees) for maintenance planning
- Harvard Data-Smart City Solutions case study on Boston's data-driven approaches
- The Open Data Challenge had a "Identifying Fire Risks" track using predictive modeling
NOVELTY ASSESSMENT: LOW-MEDIUM — The city is actively investing in AI/ML for 311 and already has an asset management platform. However, their ML focuses on classifying and routing requests, not on predicting where future requests will originate. The predictive maintenance angle specifically — forecasting infrastructure failures before residents complain — may still have room.
DIFFERENTIATION ANGLE: Don't compete with the city's 311 modernization. Instead, build a public-facing predictive model that shows residents and city planners where infrastructure is likely to fail next, based on historical patterns, weather, asset age, and seasonal trends. Think "weather forecast for city infrastructure." The city's tools are internal and operational; a transparent, public model would be novel.
Problem: Digital divide affects access to city services and opportunities. Data: Digital equity survey + WiFi locations/usage + neighborhood demographics + library locations + community centers Approach: Map digital access gaps and recommend infrastructure investments.
WHAT ALREADY EXISTS:
- 2024 Digital Equity Assessment — comprehensive city assessment informing the 2025 Digital Equity Plan
- State of Broadband Equity in Boston report — commissioned study
- Key findings: 133,000+ housing units lack fiber broadband; 51,000 households lost ACP broadband subsidies
- City launched an internet speed test website for crowdsourced data collection
- MAPC Digital Equity data — regional data resources
- Wicked Free WiFi network data available on Analyze Boston
- The 2024 Digital Equity Survey raw data is on Analyze Boston
NOVELTY ASSESSMENT: LOW-MEDIUM — The city has invested significantly in digital equity assessment and planning. The 2024 assessment and 2025 plan cover much of this ground. However, the data is largely in PDF reports and static assessments, not in an interactive tool.
DIFFERENTIATION ANGLE: Build the interactive layer that the city's reports lack. Take the survey data, WiFi usage patterns, and demographic data and create a live, neighborhood-level dashboard that tracks digital equity over time. Add a "digital access score" per block group. The city has the data and the plans — what's missing is a dynamic, public-facing tool that residents and advocates can use to hold the city accountable on its own goals.
| Rank | Idea | Novelty | Why |
|---|---|---|---|
| 1 | Neighborhood Equity Dashboard | MEDIUM-HIGH | No persistent equity lens exists across city services |
| 2 | Food Safety Transparency | MEDIUM | City has grades but UX is dated; rich data underexploited |
| 3 | Smart 311 (Predictive) | MEDIUM | City does ML for routing, not public-facing prediction |
| 4 | Safe Routes (Routing) | MEDIUM | Maps exist, but no safety-optimized routing tool |
| 5 | Digital Equity Mapper | LOW-MEDIUM | Heavy city investment, but no interactive public tool |
| 6 | Climate Resilience Explorer | LOW | Most saturated space; multiple tools already exist |
| 7 | Displacement Warning | LOW | City already built this exact tool |
Official City Apps:
- BOS:311 — report non-emergency issues (iOS/Android)
- TrashDay — waste collection schedules and reminders
- ParkBoston — mobile parking meter payment
- Where's My School Bus — real-time BPS bus tracking
Dashboards & Tools:
- CityScore — overall city performance metrics
- Green New Deal Data Dashboard — climate progress by neighborhood
- Vision Zero Interactive Map — crash data and safety concerns (ArcGIS)
- MAPC Climate Vulnerability Index — regional climate + social vulnerability
- Mayor's Food Court — restaurant inspection database
- Displacement Risk Map — Census Block Group level risk scoring
- Economic Indicators Dashboard — labor, mobility, real estate data
Analyze Boston Showcases:
- BuildBPS Dashboard, Zoning Reform Impact Tracker, Boston Globe Pothole Tracker, Carbon Neutrality Progress, Beantown Solar, RentSmart Boston, Boston Garbage Atlas, Trash City, Fire-Proof Boston Housing, Vision Zero Boston, Boston Tax Parcel Viewer
- CKAN datastore API supports SQL-like queries for filtering, aggregation, and joins
- GeoJSON datasets can be directly rendered on maps (Leaflet, Mapbox, etc.)
- Most datasets update daily or monthly — good for live dashboards
- ArcGIS GeoServices REST API available for many spatial datasets
- All data is PDDL licensed — no restrictions on use