# Open for business: the Silicon Valley POI import

Proposed talk for [State of the Map 2021](https://2021.stateofthemap.org/calls/general/)

## Abstract

_(10–75 words)_

In Silicon Valley, we are undertaking one of the largest imports of points of interest in OpenStreetMap’s history, adding tens of thousands of businesses to the map as the local economy emerges from the pandemic. Find out how we obtained comprehensive, reliable POI data and made it easy for newcomers to help – and why we think other local communities would benefit from a similar approach.

## Description

_(200–800 words)_

Santa Clara County, California, is one of the most populous counties in the United States. Home to Silicon Valley, it is an important barometer for OpenStreetMap’s suitability as a data source. Over the years, volunteers and corporate mappers have built up local coverage of buildings, addresses, and sidewalks through large-scale imports. However, until recently, there has been little focus on points of interest. Our POI coverage has been uneven, with significant gaps in many industries and in lower-income, predominantly nonwhite neighborhoods.

When the COVID-19 pandemic began, Santa Clara County was the first county in the country to impose a lockdown. Since then, the public health department has permitted nonessential businesses to reopen after submitting a “social distancing protocol” that declares adherence to safety measures such as mask-wearing. Tens of thousands of these SDPs are published on a searchable website, which is effectively a comprehensive, continuously updated business directory.

Our team of volunteers at Code for San José geocoded addresses from the SDP dataset and created a workflow for adding each business one at a time using MapRoulette and iD. We decided against an automated bulk import due to conflation and geocoding challenges, frequent typos in SDPs, and privacy considerations. In a series of online meetups, we introduced the user-friendly workflow to new mappers, growing our relatively small volunteer community without the need for extensive training.

Though still underway, this import has already had a significant impact on OSM’s local coverage, especially of POI types that are less efficient to survey. POIs are distributed more evenly throughout the county, with particular improvement in underserved neighborhoods. The POI import, along with a simultaneous building import, adds precise addresses to improve the “last-mile” navigation experience. The new POI data enables us to attract mappers from a broader segment of the population, for example by making StreetComplete’s quests more compelling.

Our experience in Santa Clara County shows that it is feasible to transform piecemeal POI coverage through creative use of government regulatory data. We believe other communities across the United States can readily take advantage of a similar approach. Our source dataset exists due to the pandemic and fortuitous health regulations, but eventually, SDP datasets will decline in relevance as the pandemic subsides. Fortunately, local governments in the U.S. have long published licensing and inspection data for various industries. These datasets have different characteristics but could potentially be imported automatically with greater confidence. In Santa Clara County, we anticipate relying on these sources going forward to supplement ad-hoc edits.

Just as the first imports of buildings and addresses raised mappers’ and users’ expectations of OSM, POI imports can challenge the popular notion that OSM is merely a basemap to place pins atop. These imports also help us achieve a balance that better serves the communities in which we live. As local economies struggle to reopen from a pandemic that has exacerbated inequality, this is an opportunity for OSM to affirm the relevance of open data.

## Notes for the organizing committee

https://wiki.openstreetmap.org/wiki/Santa_Clara_County,_California/Social_distancing_protocol_import
https://lists.openstreetmap.org/pipermail/imports/2020-November/006425.html