No, you’re fine, I just figured you might find it more convenient to work with 6-inch imagery than the blurrier global layers when realigning things. However, I forgot that OGRIP blocks overseas connections for some reason. Sorry for the confusion.

Hi, thanks for your attention to this area. Please use a high-resolution imagery layer when realigning roads and other features. Most parts of Ohio have OSIP (Ohio Statewide Imagery Program) imagery that’s both high-resolution and well-aligned, albeit sometimes a little older than the global layers.

Hi, can you elaborate on the reason that this changeset deleted the boundary of the San Manuel Reservation? This changeset also removed part of the San Bernardino city limits. Did you intend for your road edits to have this effect?

It wasn’t a particularly serious endeavor. The locations of the highway=traffic_signals nodes didn’t quite describe the complex layout of this bridge and intersection, so I had a bit of fun micromapping it all. As far as I can tell, the more practical aspects of this bridge are already in good shape for rendering and routing (as best as can be done for a SPUI).

Not quite. highway=traffic_signals is for the point along a roadway that is controlled by traffic signals, whereas I was attempting to also mark the location of the physical signals. I avoided highway=traffic_signals for this purpose because of the likelihood that it would generate validator errors and surprise data consumers.

This usage is quite experimental and has not yet been documented, but I may propose it in the future if no one else comes up with an alternative.

A lot of the discussions so far have focused on edge cases, hence the hedging and weasel words. You aren’t alone in favoring DOT functional classification as a factor in OSM highway classification because they seem more cut and dry. I’d be interested in hearing everyone’s thoughts about how the Caltrans FC criteria would best translate to OSM tags, so we can identify any shortcomings in that approach and find ways to keep the weasel words to a minimum. Just… not here in this tiny sidebar please. :-)

To participate in the talk-us mailing list, go to https://lists.openstreetmap.org/listinfo/talk-us/ and enter your e-mail address and a password. You’ll be able to post to the list and reply to others’ messages using whatever software or service you normally use for e-mail. However, you won’t be able to easily respond to existing messages from before you joined the list, so make sure to join before someone starts the thread about the California classification proposal.

A local community of one does not own any part of the map; that’s not how OpenStreetMap works. You may also recall that I’ve previously defended your right to make edits in another part of the country that one local mapper disagreed with, in a changeset discussion that got heated very quickly. Decisions about how to map should be based on the merits, not the identity of the mapper. A local mapper can inform the discussion with much-needed local knowledge, but that’s different than being a final authority.

There isn’t much riding on the classification of these downtown streets other than a typical rendered map being able to communicate that 299 connects to 44. At some zoom levels, a renderer other than osm-carto could quite reasonably include trunk roads but omit primary roads. That’s a stylistic decision OSM should be able to accommodate; to assume otherwise is to cater to osm-carto specifically.

Even if forming a continuous path of trunk roads along Eureka/Market and Shasta/Tehama looks arbitrary when zoomed way into the downtown area, I think it’s a reasonable tradeoff compared to leaving a gap within the city limits at a lower zoom level.

The questions about route membership or motorist preference are just starting points for discussion, not determining factors, and they only really matter at a county or state scale, not this hyperlocally. A router might prefer Market over California due to the road classification, but traffic conditions would probably have a bigger impact anyways.

(Typo: I meant it makes a lot of sense to give this segment of Highway **152** the same level of prominence as the segment east of Highway 156.)

Thank you for your attention to this highway. It makes a lot of sense to give this segment of Highway 156 the same level of prominence as the segment east of Highway 156. In case you haven’t seen it yet, several of us California mappers are collaborating on a standard for determining which highways to classify as highway=trunk:

osm.wiki/California/Draft_Highway_Classification_Guidelines#Trunk

Fortunately, your change seems to be consistent with this draft. Regardless, I encourage you to review the draft and provide feedback on it. As you upgrade roads to highway=trunk based to ensure connectivity, please consider explicitly tagging any expressway segments with expressway=yes for additional clarity. Thanks again!

This is good feedback. In the passage that you quoted from the guidelines, the operative word is “usually”. It’s referring to the character of the overall road from end to end as a means to keep every long-distance highway from being upgraded to trunk, but it isn’t intended to be a rubric for classifying, say, a block-long stretch of road. I attempted to clarify this point (for consistency with the national guidelines) in osm.wiki/Special:Diff/2251291 , though it’s frustrating that “road” and “highway” are such overloaded terms.

I only inquired about the status of 299 as an aside, because I thought you were claiming that 299 ends more or less at the city limits, which would’ve required correcting the ref tag and route relation. In California, state routes only consist of state highways (which are owned by the state). Unless Eureka Way and Market Street have been relinquished to Redding or Shasta County, 299 still goes through Redding uninterrupted, even if no one prefers to refer to it as 299.

Anyways, the point I was trying to make is that, according to the new guidelines, highway=trunk doesn’t necessarily mean “highway” any more than highway=primary or highway=secondary does. I don’t think anyone is claiming Eureka Way or Market Street to be a highway in the sense of a high-speed or limited-access highway. Perhaps that could be made clearer in the documentation?

Has the section of 299 within the city limits been relinquished to the city? Lately the state has relinquished a number of state highways within city limits, but I can’t find any mention of that happening to 299 in the legislative record. Have they removed the 299 shields along Eureka Way or something?

Besides, highway classification isn’t tied to route membership, and a major reason the community is bothering with this reclassification effort is to decouple trunk from construction quality and connect it through urban centers.

Here are some threads where the new national guidelines were discussed at length:

https://lists.openstreetmap.org/pipermail/talk-us/2021-May/thread.html#21018
https://lists.openstreetmap.org/pipermail/talk-us/2021-May/thread.html#21099

What hasn’t been discussed on talk-us so far is the *California edition* of the new guidelines. For your benefit and Steve’s, I’d be in favor of keeping the talk-us list in the loop as we develop the rest of that edition. However, there’s no way to keep at least some of the discussion from taking place on Slack, where there’s a much larger group of mappers assembled.

While there isn’t unanimous agreement everywhere about the new national guidelines, there’s clearly momentum behind them. In addition to considerable discussion on this topic on the mailing list and Slack, multiple data consumers are in the process of making changes to better accommodate the new tagging.

In particular, expressway=yes explicitly distinguishes the highways previously tagged highway=trunk under the old standard. This has freed up highway=trunk to be used in a manner more consistent with functional classification principles, which sometimes includes running trunk down surface streets downtown. That said, it’s quite reasonable to quibble over N. Market St. in Redding. This is an example of where we don’t have all the answers yet and are counting on locals like you to help us refine the guidelines.

The overall national reclassification project is documented at osm.wiki/United_States/2021_Highway_Classification_Guidance . It’s no longer in “draft” status, having been discussed extensively in multiple places including the talk-us mailing list. An important aspect of the project is that each state gets to adapt the guidelines to local conditions while adhering to the same overarching principles. Half the states have begun this process, and a few have already completed the reclassification.

With respect to California’s guidelines, the motorway and trunk guidelines have been relatively uncontroversial until now, compared to other states. However, the documentation is still marked as a draft because we haven’t filled out the guidance for classifications below trunk, which require considerably more discussion. That said, it isn’t too late to participate in crafting these local guidelines. It’s unfortunate that you’re unable to participate on the wiki right now, but most of us participating in this process are subscribers to the talk-us list, even if we happen to use Slack more heavily.

Bradley has moved ahead with the trunk reclassification process up here in part to clean up highly questionable classifications that remote mappers like Fluffy89502 implemented over the past year. (In the Bay Area, where I’ve followed suit, that resulted in returning highways like 17 to how they had been for years.) But it also included upgrading some roads to trunk based on the national guideline’s connectivity principle. Now would be the time to voice any objections to this overall national approach on the mailing list or somewhere similarly visible.

Just to be clear, I changed things back to how you had them. Whoever came by later on must’ve ignored the note about the Begin/End Freeway signs that you were aligning the classifications to.

This changeset was restored in changeset 116170303, but with additional traffic_sign nodes as justification so that hopefully it doesn’t get reverted again.

This changeset also removes a fictitious off-ramp in favor of existing turn lane tags at an at-grade intersection.

https://github.com/hotosm/tasking-manager/issues/4959 requests that the tasking manager tag the changeset with the project URL like MapRoulette does.

Here’s the tasking manager project: https://tasks.openstreetmap.us/projects/248