Talk:Michigan
October Meeting
- some folks are redoing the hwy classification system (again). active discussion between two folks in slack, see also
- https://wiki.openstreetmap.org/wiki/Michigan/highway_classification
- https://wiki.openstreetmap.org/wiki/United_States/2021_Highway_Classification_Guidance#State-specific_classification_standards
- "Wisconsin is a rigorous example of the process most states have gone through: identify the most important cities in and just outside the state and then figure out the best routes between them. NY happened to have a DOT-provided classification code that was able to directly map to motorway, trunk, and primary, but most states only have Functional Class which isn't a perfect mapping to OSM highway=* as it's purpose isn't to provide a routing graph. At a cursory glance most of the most important communities in Michigan look to be connected by highway=trunk in the UP and motorways in the LP, with the glaring exception of Traverse City. ... In the past there was much use of highway=trunk to mean enhanced expressway-type construction. The emerging consensus is to use expressway=yes to tag that improved physical infrastructure and use highway=trunk for the most important long-distance connections that aren't built to motorway/interstate standards as happens when low population densities don't demand such big roads, such as crossing the UP. ... The text on the Michigan page the OP linked to still uses highway=trunk to mean enhanced physical infrastructure "nearly a freeway" rather than the new connectivity-based usage of the tag." (quote from Adam Franco in #osmus Slack) --ChristopherGS (talk) 18:51, 11 September 2022 (UTC)
August Meeting
- Zoning in Greater Lansing. most of the city is not zoned with landuse=residential and such. what should be the sources for that? --Tuttiton
- Also there are named residential areas (e.g. https://www.openstreetmap.org/way/51775284) should the name belong to neighbourhoods instead?--Tuttiton
July Meeting
-Slack is talking about Michigan's approach to https://wiki.openstreetmap.org/wiki/United_States/2021_Highway_Classification_Guidance Can we organize around this? I know treestryder has done a lot of work in this area already--ChristopherGS (talk) 16:04, 29 June 2022 (UTC)
-Bike amenity classifications based on this video: https://www.youtube.com/watch?v=nsaiHhQvNSY&list=PLqjPa29lMiE0kc0pNa1px1RritclPgmvW&index=4 --ChristopherGS (talk) 16:04, 29 June 2022 (UTC)
South East Michigan Council of Governments
--Rye (talk) 17:37, 27 December 2016 (UTC) Has anybody interacted with SEMCOG - South East Michigan Council of Governments? They have high quality building footprints with address information with a license that only requires attribution:
Michigan DNR Open Data
Could the DNR's Open Data be used for public land boundaries, park locations, forest campground locations, trails and the like? http://gis-midnr.opendata.arcgis.com/
--TreeStryder (talk) 04:46, 26 June 2017 (UTC)
Act 51 Maps
I'm not sure where to add this, but I found a list of city/county maps that the state maintains to show detailed ownership of roadways: https://mdotjboss.state.mi.us/SpecProv/act51.htm
--Davenport651 (talk) 15:48, 28 January 2022 (UTC)
Mapping Rivers
Should riverbanks be mapped using the technique shown here waterway=riverbank. If so, should the riverbank be a part of the river's relation? Should the rivers Name only exist on the Relation? Or, on the riverbank and every segment of river way? See the Grand River 10716343
10716343 and the Huron River 12366431 12366431 as examples of the different riverbank mapping schemes.
The consensus formed during a monthly meeting was yes, riverbanks should be mapped using the technique shown here waterway=riverbank and they should be part of the river's relation.
Note that use of waterway=riverbank is now discouraged, see note on that page
--Emersonveenstra (talk) 14:34, 10 August 2022 (UTC)
Thank you Emerson, this is a note copied over from a meeting (I think in 2016?) it probably should be updated. To my knowledge there are no occurrences of waterway=riverbank in Michigan anymore, see https://taginfo.openstreetmap.org/tags/waterway=riverbank#map --ChristopherGS (talk) 14:48, 10 August 2022 (UTC)
The Centers of Michigan
How should three points that should be in the same location (two monuments points and the corner of the townships) be reconciled in the Meridian-Baseline State Park.
The consensus formed during a monthly meeting was to merge the mysterious and inaccurate survey marker with the historical marker and average the locations from GPS traces and Mapillary imagery.
Michigan History Center markers
Michigan has its own set of historical markers, provided by the Michigan History Center, that can be added to the map along with the ones provided by the NRHP.
Since "ref:mhc" has already been taken, I suggest "ref:mihc". I added the first marker, which shares a source with the HMDB, under this tag today (2023-04-19), but won't add any more until this tag (and the addition of these markers) has been approved.--Mecheye (talk) 18:15, 19 April 2023 (UTC)
Tagging key:foot for roads
I'm wondering how roads (e.g. primary, secondary, tertiary highways) should be tagged for foot=* given that MCL 257.655(1) states that "Where sidewalks are provided, a pedestrian shall not walk upon the main traveled portion of the highway. Where sidewalks are not provided, pedestrians shall, when practicable, walk on the left side of the highway facing traffic which passes nearest." (According to MCL 257.20 "highway" refers to "the entire width between the boundary lines of every way publicly maintained when any part thereof is open to the use of the public for purposes of vehicular travel," which seems quite broad to me, and would seemingly go as far as to cover residential streets.)
Is there some way that this could be determined by routers if the sidewalks are separate features instead of using the sidewalk tag on the road? I'm not sure whether it would be better to find all highways with sidewalks and mark with foot=no or to find highways without sidewalks and mark with foot=yes. Of course, this depends on the default value of all of these roads, and going off the default US value at OSM tags for routing/Access restrictions (which I could not find overriding access restriction defaults for Michigan), it would be having foot=yes as default and marking foot=no when sidewalks are present (except maybe if the road is tagged with sidewalk=yes/both/right/left, as this could be interpreted by default by a router as foot=no on the road and foot=yes on the sidewalk portion).
The biggest issue I see is that tagging each individual road with foot=no when it has sidewalks would require ever individual way to be modified if the law were changed. Maybe we could use some other value for foot=* that would directly suggest a reference to the law, or maybe roads just need to be explicitly tagged with sidewalk:*=separate for aid of the routers. Then all that would need to be done is state on the wiki that those roads are only default foot=yes if they don't also have a positive sidewalk tag (e.g. "yes", "right"/"left"/"both", "separate", etc., just not "no" or null), and that if they do have a positive sidewalk tag they are default foot=no.--Aenet (talk) 07:05, 1 December 2024 (UTC)
Personally, I think tagging sidewalk=yes/both/right/left/separate should be enough. That is if, like at the bottom of the very long Default speed limits page, a section were added to the OSM_tags_for_routing/Access_restrictions page for defining custom restrictions and their filters. And such a section were actually used by consumers.
Since our recent discussion on changeset 159287397 (achavi, OSMLab, FacilMap), I remembered StreetComplete has quests that attempt to answer this need. Have a look at:
- Does this street have a sidewalk?
- Are pedestrians forbidden to walk on this road without sidewalk here?
--TreeStryder (talk) 17:00, 1 December 2024 (UTC)
Is there a way to find roads with nearby separately mapped sidewalks/footpaths to speed up the process of adding sidewalk=separate to roads that don't have sidewalk=* explicitly marked?
Also, what is the process for getting a section added to OSM_tags_for_routing/Access_restrictions to define custom restrictions? Should there be a new subsection under OSM_tags_for_routing/Access_restrictions#United_States_of_America with details about Michigan custom access restrictions (just like the table for default US access restrictions) or should it be something else? I don't quite understand how Default_speed_limits#Road_types_to_tag_filters says anything about custom restrictions unless you mean how the table above describes what the default cases for speed limits are. If that is the case, would that mean that the current format of OSM_tags_for_routing/Access_restrictions should be consolidated into a large table like what is seen on the Default speed limits page? I guess I'm just not sure on the process of actually officially stating the default access restrictions in a usable way for data consumers.--Aenet (talk) 17:32, 1 December 2024 (UTC)
An Overpass_turbo query could be used to find what you seek.
The idea that data consumers could determine access by a road's sidewalk tag, referencing a Wiki table, could use further discussion by a broader audience.
I suggest using StreetComplete for now.
Just curious, are you seeking to solve a particular issue you have seen or experienced?
--TreeStryder (talk) 18:29, 1 December 2024 (UTC)
I’ll check out Overpass_Turbo for that.
I don’t have an Android device so I can’t use StreetComplete. GoMap!! has a similar feature with Quests though, and you can make custom quests as well so I should be able to replicate any StreetComplete quest in GoMap!!.
I guess I can’t remember any cases where a router has sent a pedestrian route down a road with sidewalks, so I’m not sure why I thought of it in the first place, I might have just seen the law and started tagging it, can’t really remember. Aenet (talk) 00:00, 2 December 2024 (UTC)
CATA Route Relations and Overall Organization
Looking over the CATA route relations, I see some inconsistencies and improvements that I plan to work on. I plan to go through the route relations and match their names with the ones provided at the CATA master system map (eff. Aug. 2024) to provide a source of consistency. Current inconsistencies primarily include variations in style of name, (compare "Bus 1 - Downtown Lansing - Meridian Mall - Westbound", "CAMPUS CRUISR SOUTH & WEST CIRCLE-SOUTHBOUND"). I suggest the use of the name included on the master map, matching as closely as possible to the name on the map (not including reference/route number or direction, as that information should be encoded with separate tags) (e.g. "Downtown Lansing - Meridian Mall", "North & South Neighborhoods"). Other names (such as those listed on the schedules page, if different) can be added with alt_name=* on the relation, once again, without route number or direction in the name. (e.g. ref=33, name=North & South Neighborhoods, alt_name=Campus Cruiser South & West Circle;Union - S. Neighborhood). See here for reasoning behind proposed naming convention. If descriptive names are desired, I would suggest using description=*. Discussion about naming is highly encouraged.
More importantly than the aesthetic changes are the functional changes that should be made. I propose that we begin migrating the CATA network (including stops, routes, super relations) to current Public Transportation standards as described at Public transport and Buses. This should involve including bus stop nodes in the route relations in the order described at Buses and giving them the role of platform. It should also involve cleaning up the bus stop nodes already present by putting the stop number in the ref tag (rather than the name tag) and the route number(s) in the route_ref tag (optional if included in route relations). Stops can then be given names according to the name shown when entering the stop number on the [and departures] page with stop number in the ref tag (e.g. ref=1720, name=WBD Gr. River past Maplewood D, route_ref=1).
This isn't really complete, but I will leave it here for now so it can be discussed. --Aenet (talk) 09:26, 3 December 2024 (UTC)
I'm planning to begin working on the bus stops to start adding them to the route relations shortly. For any stops that have their stop number as a name=*, I will be moving the stop number to ref=*. I may start setting the stop names to those shown on CATA's Stops & Departures/Schedules pages for the stops. I will wait to change the name for the route relations until I have at least some agreement/discussion from fellow mappers, as my suggest does not match the value format shown on Tag:route=bus or Public transport. --Aenet (talk) 16:30, 3 December 2024 (UTC)
For routes that are not loops (have sections labeled as outbound and inbound by CATA) that currently have only one relation, I am going to create separate relations for the inbound and outbound section. I will reuse the old relation as the outbound section (assuming that to=* and from=* match that of the outbound section), creating new relations for the inbound section and a super relation for the entire route. I will keep the names for these as they are with a suffix of either " - Inbound" or " - Outbound" and the same name will be used on the super relation. --Aenet (talk) 16:46, 3 December 2024 (UTC)
So far I have cleaned up the roads for the Route 1 and Route 16 relations. It would be good for someone to check my work, since I don't have experience editing relations. I might wait a bit before adding bus stops to the relations because many stops still need to be added and most don't have stop numbers yet which makes it hard to quickly find the appropriate stops. --Aenet (talk) 15:49, 4 December 2024 (UTC)
Does anyone know of a way that I can automate getting the stop number for each stop along a route using the dropdown on the CATA schedules page then putting that stop number into the Stops & Departures page and getting the location, amenities, and routes on stop? --Aenet (talk) 04:48, 4 December 2024 (UTC)
Though we all seem to be deviating from this practice, these talk pages are intended for discussing the Wiki page itself. Not so much subjects pertaining to the readers of the page. The email lists are the defacto OpenStreetMap communication channel and there is a list dedicated to transit.
Lansing topics would be best discussed on its talk page, which I see you found. And don't get me started on CATA's inbound/outbound routes which are actually loops.
Here I will try to focus on the broader topic, which I would consider to be global.
Personally, I do not believe it is wise to map things that the community cannot keep up to date. Somethings are better unmapped, or mapped in a vague way, than for the data to quickly become misleading/wrong.
Maybe a better approach would be to convince the transit agency to maintain their own routes using a tool like Go-Sync.
--TreeStryder (talk) 04:50, 7 December 2024 (UTC)