Let’s Build an Open Zoning Data Standard

Zoning – unsexy, unglamorous zoning – is likely the most important tool that government agencies and urban planners have in their arsenal to shape the urban environment. For those blissfully unaware of zoning’s existence – it’s the mechanism by which almost all US cities (with one notable exception) regulate what you can or can’t do with your property. Typical zoning regulations include things like land use (residential, commercial, industrial, etc), a measure of density called Floor Area Ratio (or FAR, shown in the map below), and design regulations like height limits and setback requirements that determine the position of a building on a lot. Things can – and do – get more complicated than this, but these are the essentials of zoning that have been in place since the early 20th century.

Zoning Allowed Density and Transit Stations in Chicago

Zoning Allowed Density and Transit Stations in Chicago

Zoning regulations are not static – in principle, they’re designed to change to reflect a city’s vision of future development, making them one tool by which master planning principles are applied in practice. Changing zoning codes to allow for higher densities in areas that are well served by public transport is, for example, a critical element of the concept of Transit-Oriented Development (or TOD).

Zoning is powerful tool, but one that’s rarely discussed in detail except by planners, city officials, or those who suddenly realize that the zoning code affects what they can do with their property. Given how critical zoning is to the future of our cities, this conversation deserves to be broader and more inclusive.

Part of the reason for a lack of interest in zoning is the difficulty in actually visualizing and understanding your zoning code. The traditional way to share the zoning code is in static maps or, more recently, data portals that often suffer from all the challenges described here. Neither of these tools are particularly easy, user friendly, or inviting.

Luckily more cities are doing better and opening up their zoning data in geospatial formats for bulk download. I’ve listed the cities I’ve found that do this in Chicago, Washington, DC and New York at the Civic Commons wiki. This is a great step – and I hope more cities are already following or will in the near future (please add to this list if you know of any).

But simply providing the zoning district geospatial data is only half the battle. Actually interpreting what’s in the zoning code – how individual zones regulate the nature of what, where, and how much you can build – is just as critical, and generally more complicated. To get a sense of this – block out an afternoon and read through a typical zoning code (just kidding). What’s needed is a simplified key to decipher the zoning code and breakdown its key elements. In the most basic terms, what we need is a lookup table to link the spatial map of zoning districts to their most critical regulatory elements – permitted land use, density, setbacks, etc. This is exactly what the folks at Open City Apps in Chicago did with Second City Zoning, a project designed to make zoning understandable to the average SimCity player. You can see the simplified zoning district table they created here.

Their zoning district table is fantastic – and their hard work deciphering the data is what allowed me to create quick mashup map from their data to show zoning density near transit. What we need now is to start moving beyond deciphering the individual zoning codes of a given city to building a way to share zoning data in an easily sharable, standardized format. We’ve already seen the power of the open data standard for public transit that I’ve written about before, GTFS. GTFS has in fact been so successful that it’s becoming shorthand for ‘good data standard’ in the broader sense, as evidenced by a recent tweet from Mark Headd, chief data officer for the City of Philadelphia.

So why build an open zoning data standard? A zoning data standard would allow for zoning visualization apps, like those developed for NYC, to be used across jurisdictions. In cities that sprawl across zoning jurisdictions (or even states, as in DC), a zoning standard would allow for an easy analysis and comparison of how, for example, different cities in the DC metro area are supporting transit oriented development. More broadly, standardizing the way zoning data is shared would allow for it to be mashed up and analyzed with things like public transport accessibility analyses like those for NYC done here, possibly providing an analytical input to the way zoning densities are ultimately set.

So what would an open zoning data standard look like? Unfortunately, zoning data is a fair bit more complicated than transit schedule data. Every city defines land use categories with some local idiosyncrasies built in. In addition, things like density controls can actually be limited through overlapping regulations on setbacks, minimum lot size per dwelling unit, height, etc – and which one of these governs may well vary from individual lot to individual lot. In addition, ‘overlay’ zones are increasingly used by cities to supercede an existing zone and allow for different uses and densities, including in DC. So a zoning standard would need a way to not only communicate information, but a method for interpreting the end result of the different rules and processes in a given zone. More broadly, zoning is an evolving practice, and while what I’m describing is principally the Euclidean type of zoning codified in the 1920s, new forms of zoning that focus on form, performance, and other metrics have started to emerge.

It’ll be complicated, but I don’t know any better way to do this than to start. I’ve laid out a simple table for DC’s zoning code, and by the end of tomorrow’s Open Data Day event in DC I’d like to be able to make a map like this for DC. I’ll be at Open Data Day tomorrow in DC (check the projects list) working on the specifics of DC but eager to talk to anyone interested in the broader issue of open zoning data standards. Should be fun.

Transport, Accessibility, and Open Data

Open data and tools are making an old idea more powerful, democratic, useful, and sustainable.

Cities are a tool designed to bring people, goods, and ideas closer together. Transport systems – sometimes successfully, sometimes less so – are what makes these interactions possible. Despite this, most tools in use by planners today focus on measuring things like faster travel speeds, rather than directly measuring how well a region’s constituent parts are connected to one another.

That’s not to say that no tools exist to analyze what’s often referred to as as a city or region’s ‘accessibility’. The maps below show the most straightforward approach – ‘isochrones’ showing how far you can get in a certain number of hours from London on the British Rail system (these ones made by the brilliant folks at MySociety). Similar maps for public transport systems are also available. Clearly, an improvement in the transport system in an area will expand the ‘contour’ for a given time band. When combined with sources of data on the locations of jobs and population – these maps can reveal which areas are most and least accessible to jobs, activities, and population.

Image

Although this type of work has been carried out in many contexts for years, until recently the data required to make this type of analysis possible, particularly in developing countries, was hard to come by. Where it was collected, it was often stored in expensive, proprietary software systems, used for the purposes of a specific study, and rarely (if ever) shared. Clearly, despite sometimes vast expense, this is not an approach likely to have an impact on furthering the understanding of urban transport challenges.

Luckily, two data sets have been powering a change in the way this type of work is carried out. OpenStreetMap, the global free map of the world, is the most prominent. OpenStreetMap data is collected by a global community of users and is available under a license that allows for open access and use by third parties. OpenStreetMap can be used to map almost anything you can see, including roads, pedestrian footpaths, and building footprints and heights. As well as providing a solid base data set for those starting out on new projects, OpenStreetMap can also be used as a way of collecting new data (as demonstrated by HOT), helping to share the fruits of data collection beyond the uses they were intended for.

The second is the General Transit Feed Specification (GTFS), the emerging data format that many transit providers (particularly in the US) are using to publish line and schedule information to their users. This is what powers the back-end of many rider tools, including google map’s transit app, and the OpenTripPlanner project. Some – though not all – transit agencies that prepare this data share it freely with the world, here.

Neither of these datasets was designed primarily to support ‘back end’ or ‘planning’ applications. They both now have strong communities of users and developers that suggest they’ll both be around for a while serving their original purposes.  But now – and here’s where things get interesting – they’re both starting to be used in projects that look more and more like tools like urban and transport decision makers can use. One example is the recent work by OpenPlans experimenting with an analyst module built on top of the OpenTripPlanner project. An early demo allowed users to visualize the changes in travel time when the DC metro area’s new light rail line goes into operation, and, even more interesting, was used to visualize changes in accessibility after super storm Sandy shut down much of New York’s transit system.

sandyBefore copysandyAfter copy

At institutions like the World Bank, where I’ve spent most of my working days as a consultant over the past three years, these tools and others like them have clear applicability. By using accessibility metrics, it’s possible to directly measure the effect of a new public transport investment on bringing, for example, low income residents a greater ease of accessing a wider variety of employment opportunities. For cities considering the siting of affordable housing, such tools can help prioritize locations according to their transit accessibility. More broadly, the volume of economic activity accessible in a given area has been shown to have an important impact on stimulating economic growth, highlighting the connections between transportation systems and the overall health of the economy.

Just as critically, the goal of studies examining the issues above is never to do something just once – it’s to support analyses that become an ingrained part and parcel of work done locally, in countries and cities around the world. In the old days, with costly project specific data collection and proprietary tools, this was usually unlikely. Now, with open source tools like OpenTripPlanner Analyst building on open data sets like OpenStreetMap and GTFS, there looks to be a much stronger chance that local agencies can learn, master, and adopt the tools to carry out these accessibility analyses themselves.

So that’s part of what’s exciting: open data and tools mean it’s a lot easier for places all over the world, especially where resources are limited, to have access to amazing planning tools and data sets. But maybe more interesting is the fact that these tools, built with and for the wider public, will enable access to this type of transport-land use analysis for a much broader swath of the public than would previously have been engaged.

This is critical, since people in planning offices, and particularly those (like me) who work in one country but support projects in others, often have to make informed guesses about the most crucial issues to analyze. Even with strong public outreach and engagement, there is always likely to be some bias to focus on understanding ‘big’ issues we know tend to be important: ie access by public transport to jobs. But what if the most critical local issue isn’t the number of jobs accessible from a given location, but the number of people within walking distance of a toilet? This, notably, has been a big issue in Mumbai in recent years.  A city wide map of Mumbai showing how many residents are within walking distance of a decent toilet could highlight areas of clear overcrowding. There are doubtless other critical issues planners havent even thought of – but that are out there in the minds of someone confronting the issue right now. The power of the new generation of tools and data is that any local tech developer can help communities adapt available tools and data sets to share these issues both locally and internationally.

There will never be a better informed source on critical planning issues than local residents themselves.  And, by building on global data sets like OSM and GTFS using open source tools, there’s no reason a tool modified by someone in Mumbai couldn’t help a local resident or planner understand the issues being confronted in Beijing, Manila, or Washington, DC. That’s worth getting excited about.