I do believe that. There is no limitation on the amount of data which may be added.

Hi,
Sorry I missed your response to this three months ago somehow. I keep expecting changeset comments to appear in the OSM web notifications like direct messages do, but I don't think the website does this.

To be clear, tagging official_ref in this way was not an original interpretation of mine - I based the way I mapped these on routes around Pittsburgh where the signed and unsigned portions had different relations in a master relation for example. (Not sure if any of those have changed since I last looked, but was the situation when I made these edits originally.) I assumed that this was already a consensus as it had been done elsewhere and seemed to be part of the intent behind the US route tagging proposal which was approved on the mailing list and wiki over the summer (admittedly it's been a while since then and I don't remember exactly if tagging ref on ways was explicitly addressed in the proposal, but I took it to be part of a shift in standardizing road route tagging on relations rather than on ways).

This could be discussed further, probably on the Community Discussion forum if the current proposal about that goes through, although to be honest my priorities at the moment are more focused on mapping things other than road routes so this is not something I feel too strongly on either way at the moment. It's not that I think the lack of signage makes a route nil, just that it seems inconsistent with the distinction between the purpose of a route relation as opposed to tags on the way itself. By analogy with other I think it is more like the formerly common practice of including is_in:country=(place) on nodes. I misjudged whether or not this was already established though.

(FWIW, I am a regular 311 user - I get responses to about 1 in 100 requests and I mainly only report about issues more directly related to safety/quality of life. If I started reporting signage issues I would have no time for much else!)

Hi, sorry I missed your comment earlier. User ElliottPlack pointed out that I missed some details in tagging rail crossings before, and I used his MapRoullette task for updating crossing data to fix the mistakes I made.

When I was tagging crossings before, I was surveying on foot. There are signs and barriers which are present which are not obvious to see on foot but can be seen more easily from the road. So tagging the warning sign locations is helpful for reference to know where the crossing details are visible from.

I also am just interested in cataloguing the types of road signs (and their codes) used in the city. Maryland only keeps a complete inventory outside of the city, so in the city we can only collect this data by survey. For crossings in particular, this is actually a detail which the state makes note of the warning sign code if they know it in this table: https://data-maryland.opendata.arcgis.com/datasets/maryland::mdot-railroad-crossing-inventory-view-2/explore (so would be useful to compare)

Also note that Punjabi city names can inflect for grammatical case, that is they are expected to change in the context of the sentence. Poland already has implemented this with name:dative and name:genitive; this way more grammatically sophisticated directions can be generated. This is not really possible in English.

There is also the issue of for example, ਨ and ਣ both being represented by N in English. So the name is not really possible to represent as well when the Latin alphabet is not suited for it.

OSM uses local language names in the name key; using English because some people would prefer it to render in Carto is mapping for the renderer. Context for other tags does not work so well otherwise as well, for example, in name:etymology it is assumed etymology of the local name. Ludhiana does not have an English etymology. There is also no Punjabi wiki page documenting any English name consensus, so it is unclear how non-English speakers are supposed to find this information.

Another example, I am working on the Punjabi translation of StreetComplete, which asks users to enter the values on signs. When this is ready, I am not going to ask StreetComplete to implement a special feature asking users to make up an English translation on the spot, because this is not typical on OSM. I do not know where users got this idea that this is OK to do in India and nowhere else.

I have intentionally left the truck designations in the `ref` tag on Erdman Avenue where the signage starts as that becomes an access restricted road where the relevance is more weighted towards truck drivers.

I'm not sure about this to be honest - as I understand it, part of the intent behind `official_ref` is that on a 'typical' highway route, the highway route number(s) carry more significance than the street name and routers/renders tend to treat it as such when that number is in `ref`. To that end, North Avenue is definitely primarily North Avenue and secondarily anything else.

Information about truck routing is also duplicated elsewhere in the schema - for signage at exits, the destination sign relations and schema for motorway junctions are explicitly designed to allow for these kind of specific route guidance at those points without necessarily relying on tags on the length of the way. Then, truck route relations themselves can contain this information, and these relations serve data users better in that they can better handle route concurrencies than putting them all in an unordered semicolon list. It's generally discouraged to tag bus route details on ways themselves for this reason and those have a greater relevance to the public than truck routes. My main concern is just that the truck route designation ends up over-emphasizing information that is not relevant to the general public, and potentially confusing if/when it is favored as a name over North Avenue. A savvy truck router should have no problem accessing the necessary details from the relation(s) without needing it to be in the ref on the ways.

Ah OK I see this may be signed somewhere in the county. Pittsburgh splits up the relation for unsigned and signed portions and I see you did that for one in MD, we could do that for more relations in Baltimore / the county. I did this for one so far

Hello,
Have you been on Druid Hill Ave and McCulloh St recently? The route signage has largely been removed. Swan Dr sign has also been removed to make it part of McCulloh. Also, official ref is intended for routes where the route number is more obscure than the road names. Nobody uses MD 129 as a ref in Baltimore over street names so this way routers & renderers can tell it apart from highways like 83 where the number is recognizable

Yes - back to city now, and Dundalk is city now too

Sorry about that, I didn't even think that's where the boundary for that neighborhood was. I'll make sure to be more diligent about checking going forward, this is what happens when you have 200+ neighborhoods in a small city

fixed, thank you

fixed, thanks. fixed Madison Park (adjacent) as well

meant to say ft worthington updates

I've been afraid to touch some of these out of concern someone might still be using them for geocoding, but it had to happen at some point. Careful if you're doing bulk edits though, there's a handful I've added manually for specific reasons. They look like this for reference (have a demolition date and some additional related tags): osm.org/node/9678978311

Part of what the ref:plot on those is helpful for is conflating them with data about community-operated urban farms and parks, and/or city-operated water wells leased through the Adopt-A-Lot program, a number of which have mappable names and addresses. Would help make the map less bare in some areas.

Additionally, something I've been on the fence about that likely warrants discussion at some point is using address interpolation ways on city blocks which no longer have addresses. For example, there is a clear utility in being able to distinguish the 4700 block of Park Heights Avenue from surrounding blocks through geocoding even though it currently has no buildings currently. Future construction will take on that block number - which is pretty likely along Park Heights in general given recent projects like Renaissance Row, and the TOD focus on West Cold Spring station - and the city does use OSM for internal service requests, which can occur on demolished blocks.

I've reverted these US-40 edits just so you know. (Not personal, but because I know I am not the only person who disagrees with this and these kinds of changes require community buy in.)

One of these changes you made is for a way which is not actually a ramp in the sense the other ways connected to I-83 are, but rather a loop for buses and maintenance vehicles. For example, during light rail closures, the buses that go into North Avenue station take you up into that loop to drop you off. As far as I know, situations like this, and service roads connected to highways are not necessarily meant to be treated the same way as regular ramps, but it seems like there's a lack of clear guidance on this as I've realized the wiki contradicts itself in places about these topics.

Also, please see this conversation - osm.org/changeset/121794056

I am not sure if I am interpreting ElliottPlack's comment correctly, but if links are meant to reflect their access restrictions, then it's likely many of these are supposed to be downgraded rather than upgraded.

There is no consensus about the local highway classification guidance yet. Please take a look at the recent discussion about this in #local-maryland, I strongly disagree that there is any reason for there to be "trunk" roads in Mount Vernon.

Sorry, I didn't finish a sentence. I wish changeset comments could be edited. About the Cold Spring ramp - the northbound one to I 83 which has a sidewalk in it school zone signage in it that I had to walk across and even in to get to school. I don't know where data consumers would be finding out about that stuff, that's still a motorway link because it's a "to" ramp

I did not intend for these edits to be disruptive and sorry to have caused any confusion. I am not sure where I have misunderstood something or what the "correct" way to map these now; this is not something I came up with but just was doing on the basis of what I read on the wiki and what is done elsewhere on the map in Maryland (various parts of Prince George's County are mapped like this). There is wording on the wiki like "They take the highest classification they go to," which I thought specifically was saying that they belong to the road they connect to rather than come from. I didn't notice this last time I looked, but the two picture examples actually show two different methods of classifying them which seems like a contradiction worth clarifying somehow.
The one on top shows links classed based on the highest road they come from, the one on the bottom shows them classed based on the highest they go to osm.wiki/Highway_link

I was also under the impression that while motorways themselves have access implications, links do not necessarily. The wiki is not clear on this and just says that situations like this are "rare" (rare in Europe I'm guessing) so as far as that goes I would think data consumers would just have to be conservative until there is a clear guideline. These particular ramps do not share the same access restrictions as the motorway, as is evidenced by the cross walks going through them, both on "to" and "from" ramps. If it really is the case that links also imply the same access restrictions, then don't these (and most of Baltimore's) "to" ramps need to be downgraded as well? I'm not actually sure how data consumers would be able to tell for example, that the northbound "to" ramp from Cold Spring (I actually had a conversation with someone at Baltimore DOT a couple months ago about school zone signage in highway ramps - there's a small possibility there could be more of that in there near future. And a different conversation with someone from at Bowie city govt last year about the sidewalks they're planning on building along some stretches of motorway link once the apartments planned for Melford Blvd are built.)