Hey, just a heads up that this changeset removed a couple ways that in turn broke a number of relations (county boundaries, USFS boundaries, forest landcover). I'll have this fixed shortly, but in the future please do not delete ways unless you are certain what you are doing, especially when they belong to large relations.

Thanks,
Bradley

This sounds like a problem with the routing engine rather than the data. "Massaging" data to correct undesired behavior in third-party apps is generally considered bad practice, similar to "tagging for the renderer". If the tags are correct (in this case, foot/bicycle/horse=designated rather than *=yes, because these trails aren't just simply open for use via these travel modes but are designated specifically for use via these modes), then the problem is with app, not the data.

Hello,
I've noticed this is the second time you've changed tags for foot/bicycle=designated to foot/bicycle=yes on this stretch of the TRT. Why are you making this change? These trails are designated for foot, bicycle, and equestrian use and are signposted as such.

Bradley

Found a mobile-friendly way to look at the way history - there isn't a road across the meadow there, nor is there any evidence of one having been there in the past. Multiple visits to the area myself confirm this. This geometry is imported from old TIGER data, which is notoriously inaccurate with minor roads.

Hi ConnorWong,
I can confirm this way deletion specifically later tonight when I can get in front of JOSM, but as a matter of normal tagging practice I don't remove tracks unless there's no physical evidence of something existing or having once existed roughly along the way. If there's a discernible trail following an old road bed I'll tag 'highway=path', 'abandoned:highway=track'; old road beds with no discernable/navigable path just 'abandoned:highway=track', and old roads physically decommissioned as 'razed:highway=track'; 'motor_vehicle=*' as appropriate. I use multiple satellite imagery sources, USFS GIS data, and high-res USGS hillshade especially (allows you to "see" road grades through forest canopy) in absence of a manual survey to make these tagging decisions.

Bradley

I'll be importing the boundary line from USFS GIS data with the result that:
- The LTBMU/Eldorado NF protected_area boundary relations will contain all land owned by the federal government and administered by the LTBMU & Eldorado NF respectively (protect_class=6 lands)
- The separate Desolation Wilderness boundary relation (protect_class=1b area) will overlay the two previous relations as a proper subset, rather than as separate mutually exclusive sets.

Cool, sounds like there's buy-in on overlapping protected area boundaries. What'd need to be done at this point is to reimport the boundary line between the LTBMU and Eldorado NF within the Desolation boundary and then correct the relation members. Gimmie a couple hours and I should be able to have this done. Day off today - just got back from skiing Tallac this morning, speaking of Desolation!

This relation represents a protected area rather than an administrative boundary - it maps land that is physically owned (and therefore protected) by the USFS as opposed to federally designated administrative boundaries. Desolation is a separate relation because it is protected at a higher level than other nearby federally owned land. The wilderness here would be considered jointly operated by LTBMU and Eldorado, with the exact line where the jurisdiction changes not relevant to the relation since, again, that is a federal administrative boundary rather than a change of name or protection level "on the ground". Were these boundaries to be introduced, it should be as a 'boundary=administrative' rather than 'boundary=protected_area', but that would be a question to bring to the community via mailing list or slack/discord as to whether this information should be included in the database, and if so, what administrative level these administrative boundaries should be at. Having stood multiple times on the ridge that splits the LTBMU and Eldorado management of the wilderness area, I can assure you there is no way to know there is a change in managment outside of consulting official maps.
Bradley

Hello thehedgehog,
I'm reverting this changeset since it broke the LTMBU NF relation. I'm not sure why members were removed from the relation - the members removed define the land ownership boundary and were correct. ID is undergunned to deal with these complex relations in my experience - please consider learning to use JOSM instead if you wish to modify complex relations in the future.
Bradley

Please don't forget to add 'expressway=yes' tagging when moving roads off the trunk<=>expressway tagging scheme. Recovering this information is time-consuming after the classifications have been changed.

Bradley

The 'motorway' tagging here terminates at regulatory 'Begin Freeway' and 'End Freeway' signage. This is unique to California and has been discussed with other CA mappers as an on-the-ground way to tag motorway beginnings and endings (osm.wiki/California/Freeway_ends). Please revert these changes back to the posted beginnings and endings.
Bradley

Hey Joseph,
It isn't settled that CA 299 east of Redding should be tagged as 'trunk' per discussion on #local-california channel of OSM Slack - if it were to be promoted, it would be in service of a Redding <-> Boise route, but would involve roads that have such low traffic & "importance" (per feedback from Oregon mappers) that it's debatable whether this should be consider a 'trunk' routing at all. If the goal is to orthogonalize roadway construction standards from importance this short segment should probably remain 'primary' with 'expressway=yes' for the time being, especially in light of contentious discussions in this area recently.
Bradley

>I've literally said 299/44 could be kept as trunk roads, and made multiple recommendations where the best place to not tag 299 as a trunk anymore would be.

Okay, this is all that's being proposed. CA 44 through downtown as 'trunk', and CA 299 west of Market as 'trunk' as well. If this is acceptable to you, then the classification debate is resolved, and there are no other proposed changes in this area.

To be clear here, I didn't mean to say that I think SteveA is acting authoritatively necessarily. I've never edited in Santa Cruz nor do I know anything about the Adamant1/SteveA drama arc. Just that, if A1 felt like they weren't treated fairly, that they would use that as motivation to act better rather than use it as an excuse to do the same to others.

It doesn't seem like this is going to be a productive discussion at this point. Leaving the tagging as is since it was the closest thing to an agreement that could be reached so far. While again I am sorry I didn't reach out to you first before making these changes, and I do respect and appreciate your work in the Redding area (one of the better mapped areas in the US imo), I strongly disagree with the sentiment that any area "belongs" to any one mapper in the sense of having an authoritative/final say, be it you *or* SteveA.

(Lots of "quoting" words here - I don't mean to imply anything when I do this aside from that these words are a bit fuzzy and could be replaced with a handful of other similar terms and therefore shouldn't be read strictly)

I feel like this analysis of traffic patterns through Redding is hyper-fixated on local Redding traffic, which is missing the point of classifying 'trunk' roads as "main inter-regional long-haul routes" (or some other equivalent language). It's true of every city and town that the "best" way to get in/out of town could involve many different roads depending on where you're headed. What's being proposed here is that, the way you would take to travel between Eureka and Reno (specifically for this discussion) should be tagged 'trunk', in general with more weight being given to higher quality, higher traffic roadways, as well as towards roadways that are part of a state or national highway network (to discourage tagging local arterial "shortcut" roads that are technically faster but not part of a major route). This is a world-wide mapping project and I'm really not understanding what this language regarding who the "target audience is here" is supposed to imply.

Traffic counts are useful for determining regionally what roads are "more important" relative to others but don't necessarily mean anything outside of that context. It's true of many major highways - certainly true in urbanized areas - that the majority of the traffic carried on them is for local or regional trips. This doesn't take away the importance of a route for serving as the "best" or "main" way between two significant population centers. As an example edge case, US 95 is a very quiet conventional highway that a solid portion of the daily traffic is related to local agriculture. However, it is unambiguously the "best" way between Reno/Carson and Boise, so it is promoted to 'trunk' under this tagging scheme. If we classified it based on the trips that the majority or plurality of the traffic on the road is using it for, it'd barely qualify for 'secondary'.

I understand there are many routing idiosyncrasies that are unique to Redding, but this situation looks similar to that of Salem, OR, which has the connection between OR 22 and Interstate 5 following a sequence of urban surface roads promoted to 'trunk', which may not necessarily be the "best" way for trips from Salem to the west depending on where you're at, but *is* the way you would recommend out-of-town traffic passing west through Salem to take.

I dropped the surface roads east of Buenaventura back down to 'primary'. I would like to try to reach a resolution however on what the "main" way is connecting the interstate/freeway system to CA 299 west of Buenaventura, since the goal of this tagging project is to develop a coherent network of 'trunk' roads in CA by figuring out these gaps. If you were taking a trip between Chico or Reno through Redding to Eureka, what is the way would generally consider the best way to get through Redding on to CA 299? I understand that downtown Redding is a mess and exactly what way is "best" depends on where you're coming from and what you're driving, but it seems fairly clear to me that, speaking very generally, one would recommend traffic to take Shasta/Tehama and Pine/Market one-way couplets to Eureka Way (CA 44 route).

W.R.T. to Market St/etc north of Eureka, I bumped these up to trunk based on signage EB at Eureka and Market indicating destination NB I-5 traffic to take Market north. I'm happy to defer to your local knowledge here and will bump these back down as soon as I can (currently only at my computer a couple hours a week). My intent wasn't to trample on your knowledge of the region and I'm sorry I didn't reach out to you before making some of these changes. With that being said, they are easily reversible and ultimately prompted the discussion that needed to be had. If we're in agreement that CA 299<->44 is acceptable as 'trunk' then there's no disagreement as far as tagging goes, with the understanding that roads north of Eureka need to be bumped back down to primary.

These are "drafts" in the sense that they are incomplete and that many routes and areas are still under discussion, not in the sense that every road or area needs to undergo a formal voting process. The linked documentation is the summarization of months of discussion and debate involving more US mappers than I have seen in any debate over this on the mailing lists in years past. While I am the one who ultimately changed the 'highway' tag in this region, these changes were not undiscussed nor unilateral. It isn't a mass, automated re-tagging if I'm also local to the region and am familiar with many of these roads. CA 299 is the road I am least familiar with of my proposed routes, but my justifications for thinking this should be a 'trunk' route are further down.

The vast majority of the roads I personally proposed are north of I-80 and Sacramento, and most are roads I have driven personally. I put them up towards the largest active discussion I could find regarding road classification with one issue north of Alturas that was resolved after some discussion. I think it would be a bit absurd to put this to a wiki vote; you'd be lucky to have more than two mappers local to northern CA proper participate in a wiki vote. Establishing consensus in areas where there are only a low single-digit number of active, local mappers is going to be difficult and I wouldn't consider a wiki vote of three people authoritative either. Likewise, sending a list of 8 or so roads I propose to bump up to 'trunk' to the mailing list doesn't seem to be appropriate either. The only other CA mapper that is active there but not on Slack is SteveA, who I wouldn't necessarily consider "local" to northern CA.

Most mappers who are concerned with improving US highway classification and are trying to productively reach a consensus have been discussing this on Slack, which is why my "proposal" was posted there. It seems to me that you're upset you haven't been involved with discussions, but the options I've seen are:
1. Private DM, not useful for public discussion - this is a long-contentious topic that concerns more than just you and I.
2. Changeset comments, too decentralized and impossible to keep track of to be of much usefulness for public discussion and consensus building. I will cross-link this changeset to the local-california Slack channel so they are aware of your objection and can participate here.
3. Mailing lists, which have an unfortunate extended history of being unproductive regarding highway class discussion.
4. Wiki, which is fairly useful for discussion, but you cannot participate on by admission.
5. Slack, where there has been active daily discussion about road classes with mostly productive results.

I'm not on the Discord, but if there are similar discussions going on there that are active and productive I'd be more than happy to participate.

My reasoning for this highway being 'trunk' is that it is more important than neighboring primary roads. It is the best route connecting Eureka (greater CSA of pop 45k) to Redding and through to Reno/Carson. This road is an ex-US highway, has more open geometry than nearby roads supporting faster traffic, and has frequent passing lanes which other nearby highways to my knowledge don't have nearly as frequently. If this road is to be 'primary' (only regionally important), I think CA 3, 36, and 96 would need to be discussed as well since it seems these play a less important role in the highway network than CA 299. This is corroborated with Caltrans AADT counts (https://dot.ca.gov/programs/traffic-operations/census), which shows higher minimum counts on CA 299 than on other nearby highways.

I'm happy to revert this if discussion pans out to determine that this road ought to be 'primary', but it should be done so on the merit of discussion rather than turf. If you have "zero urge" to participate in discussion I'm not sure what to say here. Why specifically do you think this road isn't appropriate to be 'trunk'? Do you disagree with tagging important inter-regional highways as 'trunk'? If so, why?

Bradley

Hi Adamant1,
The proposal to move main interregional long-haul routes to 'trunk' is here: osm.wiki/United_States/Highway_classification. Specific guideance proposed for CA here: osm.wiki/California/Draft_Highway_Classification_Guidelines. Most of the discussion around this has been on the #highway-classification and #local-california channels of the OSM Slack, which in my experience has been much more productive for discussion than changeset comments or the talk-us mailing list. I generally don't like having these discussions over private message since I think they ought to be public. If you are opposed to using this, the talk page of the wiki article has been useful as well. I apologize that I haven't specifically reached out to you but these changes have been discussed with a handful of other CA mappers as well as other US mappers working on this proposal without any objection. If you are concerned about the 'trunk' reclassing proposal in its entirety Slack is where the majority of this discussion has been happening and I would encourage you to voice your objections there.
Bradley