[RFC] Feature Proposal – Motorway classifications

This proposal is intended to classify highways tagged as highway=motorway based on their importance. The fact that they are controlled-access highways can be accounted for by adding motorway=yes.

Please discuss this proposal on its wiki talk page.

9 Likes

Before asking people to discuss and then implement a change such as this, I’d suggest asking yourself a simple question:

How would this change help?

Helpful changes that we’ve made in the past have allowed us to map things that we previously couldn’t. As an example, long fords used to be tagged as highway=ford, but that has a big disadvantage because it doesn’t allow us to say what sort of road it actually is, in addition to being a ford. I suspect that your best route to persuading people is finding a few places around the world where people have tagged something as highway=motorway which might be a candidate for other tagging. Based on other recent topics about motorways you might find a candidate in the US, but I suspect that you’ll struggle elsewhere.

If you can’t find such an example then the current tagging scheme isn’t holding us back and this suggested change is of no benefit.

5 Likes

As I explained on the proposal page, since motorways are classified based on their physical characteristics, there is no way to indicate its role in the network. In the US anyway (AFAIK), the assumption is that they’re trunk roads, but there are some cases where that is not the case. For example, SR 613, SR 171 + Airport Connector, SR 564, US 93 Business, and US 95 Business would be classified as highway:primary if they weren’t built to freeway standards. In fact, a user commented on a changeset I made to US 93 Business, saying that it should be highway=primary + expressway=yes. Changing said road segment to highway=primary + motorway=yes would inform people and data consumers that, functionally, it is still a local road. I certainly would count this as a benefit, especially since new users might notice a freeway ending and classify that transition segment as a trunk road, when in reality, highway=trunk is used for importance.

I am certain that you don’t appreciate just how big of a can of worms you’ve opened, and how many data consumers and mapping communities this touches in complex ways. I even kind of agree with what you’re trying to do, but this is too big of a change for a global proposal. Possibly it could be implemented at a more local level to be proven out, where the data consumer impact is smaller, but even that is going to require a long series of discussions to discuss which motorways are at the highest importance level and which ones are trunk? motorways. I have to admire the naive optimism here.

It would also immediately break existing data consumers that don’t handle motorway=yes, and that makes this dead on arrival.

And thus I must quote @woodpeck:

If you want any change in OSM, you not only have to describe (and possibly implement) what you want; you also have to think about a method of transition from the status quo to whatever it is that you would like to have.
User:Frederik_Ramm, User:Frederik_Ramm/Ideas_for_API_0.7

And also @Mateusz_Konieczny:

with new tags we have luxury of making definitions. In case of tags used millions of times it is too late to introduce definitions changing how tag is used, we may only describe real usage
User:Mateusz Konieczny, Talk:Tag:landuse=forest

Good luck to you, sir.

3 Likes

Ah. My parents have often told me that I think too simplistically at times :sweat_smile: “In a perfect world, that would be the case…”

Well, do I have the solution for you! The city I live in only has either six or seven freeways and is pretty small (geographically speaking, anyway).

Edit: I forgot to do this earlier, but in the city I live in, three of the freeways would be classed as trunk roads (possibly highway=motorway since it appears to be a shortcut of highway=trunk and motorway=yes?), while the others would all be primary.

It isn’t clear to me if this proposal changes anything about the existing motorway=yes tag, which seems to already be used to indicate that a highway tagged as something other than motorway is built to motorway standards.

Is the proposal more about promoting wider use of this tag than changing its meaning? Which would in turn be more about changing guidelines for specific countries than global tag description pages?

You could argue that, but that’s not the main goal of the proposal. The primary (pun intended) goal of this proposal is to classify motorways based on their importance in the road network. This is useful because:

  1. Not all countries (USA, I’m looking at you) have a legal class of motorways. As I’ve mentioned before, the closest thing you can get is the Interstates, but exceptions exist.
  2. A new user to OSM might see a segment where a freeway is transitioning to a surface road and decide to tag it as the thing between a motorway and the surface-level road, which is likely primary. So, they retag that segment as highway=trunk… only to later notice their edit reverted after a more advanced user tells them that that is not how highway=trunk is supposed to be used.

So I originally did like the tag and I think that it’s better to have a consistent list of values of highway=* but there is a catch.

Notice that there are way more highway=* values that depend on functionality rather than importance:

  • living_street
  • service
  • pedestrian
  • track
  • bus_guideway
  • escape
  • raceway
  • busway

Not to mention that such a switch of schemes is impossible because highways are the thing that’s supposed by almost every data consumer and a sudden re-tagging of all motorways in Europe as highway=trunk.
Additionally, the majority of Europe uses highway=trunk to mean roads of higher quality, often access-controlled, meaning not depending on importance which is what I recommend fixing up first.

Instead, I think we should move away from the highway=* importance dependence proposal and move these values over to the importance=* tag.

Overall, this situation is being discussed further in this thread with my latest stance on it in my latest post.

you didn’t “explain” it, you just stated it, and it is not correct. I have commented on the wiki discussion page of the proposal about this issue.

2 Likes

In fact, it’s used for legal now. Therefore it should not be used for the physical. Key:motorway - OpenStreetMap Wiki
Discussion Proposal talk:Motorway classifications - OpenStreetMap Wiki

That is a whole can of worms that is beyond the scope of this proposal.

I have also replied to it. This issue is really more so about the US than Europe. I’ve never been to Europe, so I’m not going to force this upon a place I’ve never visited before, as I’ve received pushback in the past for doing that.

As of the time I’m writing this, I’ve been to 17 U.S. states plus DC, which is about ⅓ of them. Once again, in the US, there’s no legal status of motorways. The closest is the Interstates, but it has its exceptions (I-180 in Wyoming, I’m looking at you). There is no sign saying, “This highway is a freeway,” unless you live in California, where signs do exist designating the beginnings/ends of freeways.

If you’re trying to solve a US-specific problem, then start there, and let’s not drag the global community in a pointless debate that will go nowhere. Even then, you’ll need to start with a clearly-described problem statement that identifies who is harmed by the current tagging system.

8 Likes

I think I will stick with just the US for now. As mentioned before, some countries have legally defined a class of highway as a motorway. I won’t edit those, in part because I’ve never been to Europe before. As a matter of fact, I would have put this proposal in the US community forum if the wiki didn’t tell me where to put it instead.