Lyft

From OpenStreetMap Wiki
Jump to navigation Jump to search

 Lyft is a rideshare company based in San Francisco, operating in cities in the US and Canada. Lyft Mapping uses, tests[1], and invests in OSM data with an organized editing team[2] when errors are discovered.[3]

Lyft logo.svg

As of 2020, Lyft is not a Corporate Member of the OSM Foundation.

Mapping Focus Areas

The Lyft OSM team does not participate in bulk editing or importing data created by AI algorithms. Instead, the team is focused on improving the OSM map manually in ways that improve the rideshare experience. Lyft’s mapping team works with traffic roads (adding and validating missing ways, lanes, destinations), relations (mostly turn restrictions adjustments), barriers & access. Lyft is also taking a step towards enhancing the travel experience by expanding its focus beyond traffic roads to pedestrian paths.

Missing Roads: Identifying places where drivers can drive but there are no roads; reopening roads closed for construction; redesigning the road network after construction; changing road types after construction.

Road Lanes: Adding/editing the number of lanes on a particular part of the road; specifying possible directions from each lane if possible.

Highway Signs: Adding information from highway banners to road segments.

Turn Restrictions: Adding and removing relations that restrict turns according to special signs, road geometry, or specific rules.

Barrier & Access: Identifying the type of barrier and adding it to the map; adding different types of 'access' tags to point or linear objects, including conditional ones; adding other information to barriers (for example, for code gates).

Pathways: Adding pedestrian walkways in parks, colleges, stadiums, airports, bus and train stations, etc.

Complex Investigations: Improving the map without focusing on a specific problem, for example, investigating specific areas; analyzing OSM community's changes in case of vandalism or critical mistakes.

List of Projects we are working on

Sources for edits

Despite all available open sources, Lyft is also collecting ground truth and drivers telemetry data to update maps with real-time road constructions and environment changes. In addition our drivers also help us by reporting potential map issues or problems during rides.

Editing geographies

Lyft contributes edits to OSM in North America.

Mapping Guidelines

Lyft organized editing primarily concerns road network features and signage. Lyft collects valuable information such as missing roads, turn restrictions, one-way directions, destination signs, and other information that is wrong, missed or outdated in OSM. As sources in making edits, Lyft uses driver feedback, GPS traces, satellite and street-level imagery. The collected information is contributed to OSM with the help of our editing team. Edits are made to OSM according to Lyft-owned sources and are passed through internal quality control prior to being published. Lyft always follows the mapping guidelines mentioned in OSM wiki (where guidelines exist).

Process Flow

Here is the process followed by Lyft OSM team while editing:

1. Get generated area where GPS traces (internal data) differed from the existing road geometry or area where segments should be checked for missing or invalid data

2. Find the newest Lyft-owned street level imagery (internal data) and GPS traces to check the data

3. Use iD editor to fix issues according to the sources above

4. Audit compliance of edits to OSM's global and local policies by the Quality Control team

5. Upload changeset with appropriate comments

Quality Control

Lyft has a team of experts who audit edits made to OSM. Currently, our process is internal double-blind quality control. We use the tools listed in the wiki as well as manual analysis of OSM tags and geometries to ensure the quality of our work.

Lyft OSM team

Lyft has a team of trained OSM editors. The full list of the Lyft team is published on the GitHub page.

List of OSM Team Members

OSM Training Decks

Our training materials that we use to train OSM Lyft Team Members are publicly posted.

List of OSM Training Decks

Lyft impact

In the span of four years of mapping, the Lyft team has published over a half a million changesets with 5.5 million edits. Our contributions have included, but are not limited to:

  • 135k changesets to provide and maintain the actual road network and suggest the most optimal routes
  • 142k lanes editing changesets to help drivers better navigate through interchanges and multi-lane intersections
  • 67k edits based on destination signs to make guidance on motorways as useful as possible
  • 37k turn restrictions adjustments to ensure drivers are offered the best possible route without any illegal turns
  • To enhance the driver experience in pick-up and drop-off zones, we have published 37k changesets in areas with gates so that the route went through the appropriate gates

How to contact us

Lyft welcomes feedback and suggestions on improving the editing process. When members of the editing team receive feedback or comments on a changeset, it is considered a top priority to respond and make corrective action on their OSM edits if required. If you have any questions about the editing process, or a specific edit done by the Lyft team, please reach out by sending an email to dct-osm(at)lyft.com. Responses to inquiries received are typically sent within two business days. skudrashou_lyft and technician_lyft are acting as a primary contact person for the team.

See also

  • Grab Another taxi like company editing OSM
  • Uber Another taxi like company

References