TIGER/Line Contamination Breaking Transit Routing in Hamilton County MetroNow Zones
Posted by krassy513 on 28 April 2026 in English.1,934 Unreviewed TIGER Road Segments Are Breaking Public Transit in Hamilton County
SORTA operates MetroNow, a microtransit service covering four zones in Hamilton County: Blue Ash/Montgomery, Springdale/Sharonville, Northgate/Mt. Healthy, and Forest Park/Pleasant Run. The service is powered by Via Transportation, whose routing engine consumes OpenStreetMap as its base map layer.
I am a daily MetroNow rider. A persistent routing failure at my address led me to inspect the OSM data that Via’s routing engine depends on. What I found on one street turned into a zone-wide audit. The results are significant enough that I want to share them with the local OSM community and ask for help.
The Street-Level Problem
A residential street in the Blue Ash zone carries two defects from the TIGER/Line 2008 Census import. Both carry tiger:reviewed=no. No human has verified this data since import.
-
False
oneway=yestag on a two-way residential dead-end street. The false one-way tag makes the routing engine calculate entry with no legal exit. The engine marks the street as unreachable. Drivers get redirected to a neighboring street. Riders get circuitous routes that bypass the correct address entirely. -
Disconnected node gap at a cross-street intersection. Two ways representing the same street do not share a node. The routing graph treats the street as two separate, disconnected segments.
For a transit-dependent rider with mobility limitations, a routing failure is not an inconvenience. It is a service denial.
The Zone-Wide Data
I ran the following Overpass query across the Blue Ash/Montgomery zone:
[out:json][timeout:90];
way["highway"]["tiger:reviewed"="no"]
(39.16,-84.44,39.24,-84.33);
out tags;
Results:
- 1,934 road segments carrying
tiger:reviewed=no - 1,163 residential streets never reviewed since the 2008 import
- 57 residential segments with
oneway=yes(false one-way candidates) - 164 named streets with 2+ disconnected segments (node gap candidates)
- 20 residential streets matching the exact compound defect pattern: false one-way tag combined with multi-segment node disconnection
The full dataset (JSON, 1,934 elements) and a classified defect inventory with Way IDs are available on request.
Why This Matters Beyond Map Accuracy
Via charges SORTA for this routing platform. SORTA funds MetroNow with Hamilton County sales tax revenue from the 2020 Issue 7 levy at 0.8%. Via markets a proprietary layer called ViaMapping as the answer to OSM’s limitations. ViaMapping adds speed data and points of interest on top of OSM. ViaMapping does not audit the underlying road graph. It does not verify oneway tags. It does not detect disconnected nodes. Every TIGER artifact passes through to the routing engine that transit riders depend on.
Our tax dollars fund a routing platform built on map data that nobody audited. The OSM community created the data through a mass import nearly two decades ago. The commercial platform profiting from that data performs no quality assurance on it. The transit authority purchasing the platform does not know the base layer contains unverified data. The riders absorb the consequences.
What I Am Asking For
A collaborative TIGER cleanup of the MetroNow service zones. Starting with Blue Ash/Montgomery. Expanding to the other three zones.
The work breaks into two operations:
-
Defect Class A: False one-way tags. Verify ground truth for 57 residential segments carrying
oneway=yes+tiger:reviewed=no. Remove false tags. Each edit takes 3 to 5 minutes with satellite or street-level imagery verification. -
Defect Class B: Disconnected nodes. Inspect segment boundaries on 164 multi-segment streets. Merge disconnected endpoint nodes where two ways of the same street fail to share a node at an intersection.
I have documented remediation procedures for both defect classes with step-by-step instructions, changeset comment templates, and acceptance criteria.
Who Can Help
Anyone with an OSM account and familiarity with the iD editor can contribute to this effort. Local knowledge of Hamilton County streets is valuable for ground-truthing the one-way candidates. If you have experience with bulk or mechanical edits, I would like to discuss whether a scripted approach is appropriate for the tag removal portion and what the community review process looks like.
Anyone who rides the bus in Hamilton County or pays the levy that funds it has a stake in whether the map underneath the transit system works.
I have an active service complaint with SORTA establishing the transit impact through their Service Planning office. The goal is to fix the map data, get the corrections propagated to Via’s production systems, and prevent this class of failure from affecting other riders on other streets.
Comment here or message me directly if you are interested. I will share the full dataset and remediation blueprint with anyone who wants to participate.
Discussion
Comment from Minh Nguyen on 5 May 2026 at 22:52
Hello,
Thank you for your interest in our coverage of the Cincinnati area. By now the flaws of the old TIGER import are well-known in this community to the point of being a cliché. However,
tiger:reviewed=nodoes not by itself indicate that there’s a problem. It just means the roadway originated from the TIGER import. Each roadway in that import carries this tag unless someone has gone out of their way to delete it. Yet most mappers do not delete it even after they completely correct the data.To know whether a way has truly been reviewed, you’d need to dig deeper. We’ve compiled some queries for Overpass and QLever that check the last-edited metadata on each way and any node along it. These queries return nothing anymore because everything has been touched over the past couple decades. However, a more thorough analysis of the revision history would still yield many opportunities for improvement. Perhaps this is something your audit pipeline could do in order to generate more meaningful statistics.
I reviewed your first changeset to find out how the errant one-way and disconnection occurred. These appear to have been introduced, respectively, three years ago by a novice mapper and one week ago by a more experienced mapper. I can only imagine that both cases were honest mistakes that they overlooked. Both errors appear in existing OSM validators, but we have quite a large local backlog.
These cases offer a glimpse at some of the one-off problems that naturally occur in the raw data that OSM is collecting, because we rely to some extent on after-the-fact quality assurance. Besides your service complaint to SORTA, you may consider reaching out to Via about potential quality checks they could apply before ingesting updates, to ensure that these improvements reach customers in other cities as well. The local community would be happy to work with you, Via, and SORTA to ensure that riders can depend on the road and route network that OSM is powering behind the scenes.