[RFC] Feature Proposal - Make cycleway:both the default to indicate both sides

[roelant]

@tordans

In case we don’t find common ground on the current definition, how could I present this proposal in a way that avoids naming the current definition?

Possibly by literally not doing that and/or avoiding the discussion alltogether, or at least trying to present it in a more neutral way. Personally I thought westnordost did a great job of that here.

To elaborate on that: Right now your “opening statement” is that cycleway= is unclear, which is cause for debate. It feels like opinion more then fact, because on the one hand it’s unproven (and perhaps unprovable), while on the other hand not everyone is actually in agreement with it (already disproving that it is fact to everyone). But yet you state it and treat it as a hard fact afterwards. So softening it to something like:

@westnordost

It might be enough to mention cycleway is not explicit about for which side the cycleway is meant, but usually it can be assumed that it is meant for both sides except in the case I mentioned.

… probably represents the reality a bit more closely/objectively.

And then adjusting the proposal to what that should mean in terms of the relation between those to tags, editor support (e.g.: support both, but don’t enforce a preference/default, that’s up to the creator of the editor), adjustments in terms documentation (also allowing and reflecting that consensus-reality).

You might even distinct in what values this is at play for. In other words, in which cases that assumption has proven problematic.

For instance, I see little added value in cycleway:both=no compared to cycleway=no, because “it’s not there” really means the same “it’s not there on both sides” to my mind. Where is I do understand based on the discussion that it is more problematic for other values, such as cycleway=track - in which case a recommendation (and explanation why this is the better way of doing it) to could in order.

@tordans

On Citation: Thanks for explaining your point @roelant. I hope this thread and the proposal will be a good enough basis for a decision for most. I will not have time to add more references, citations, and sources beyond what we already have. I am happy to add suggestions from others, though.

Fair enough. And at some point, also, all (or at least: enough) is said and done really.

On a separate note, I don’t think there are many – or any – proposals that conform to the quality level you describe.

Probably. Though proven or “fact based” discussion (which is what it represents to my mind, and I believe is the “spirit of the law” behind “citation needed”), is that it suggests a certain level of objectivity and thus neutrality.

I’ve voiced concerns on that note before and hope to have clarified above how I believe it influences the proposal.

Thank you again for the hard work you’re putting into this. Sincerely.

[Mateusz_Konieczny]

@roelant

Right now your “opening statement” is that cycleway= is unclear, which is cause for debate. It feels like opinion more then fact, because on the one hand it’s unproven (and perhaps unprovable), while on the other hand not everyone is actually in agreement with it (already disproving that it is fact to everyone). But yet you state it and treat it as a hard fact afterwards.

has anyone tried looking at existing uses of cycleway=track or cycleway=lane to check how it is used in practice? This may indicate either lack of consistent use of use being actually consistent (I guess that best form would have a list with random samples and checking what it seems to map based on aerial…)

This is not a foolproof method, but if situation is strongly in one direction it may be quite strong indicator…

[balchen]

To me, tagging highway=* + cycleway:both=* is redundant to tagging highway=* + cycleway=*. But this may be specific to my common tagging scenarios, and I’m open to others having a different experience. As others have already requested in this thread, it would be helpful to see some real-world cases where highway=* + cycleway=* causes unresolvable issues for data consumers.

As a data consumer, I treat cycleway=* and cycleway:both=* the same. That may be too simplified, but it hasn’t caused noticable issues so far.

I do see a lot of issues that stem from people insisting on tagging everything on a single line instead of drawing a separate line for the cycling thingy. If you’re tagging a painted cycle lane, it probably belongs on a common line. A physically separated cycle lane probably benefits from having its own line. A separate cycleway definitely needs its own line.

I’m curious to see how many of the presented real-world cases would be better resolved by drawing a separate line for the cycling thingy.

[westnordost]

@balchen

A separate cycleway definitely needs its own line.

It’s mostly a matter of level-of-detail.

The opinions have been turning in favor of rather mapping bike paths (and sidewalks) rather on separate lines, especially on bigger roads, however, mapping bike paths and sidewalks on the road-line has always been and continues to be correct.

I think it would be difficult to seek consensus on clearly recommending one approach over the other, thankfully this is not part of this proposal, so let’s not get off-topic.

@balchen

it would be helpful to see some real-world cases where highway=* + cycleway=* causes unresolvable issues for data consumers.

It (at least) adds some ambiguities in edge cases (~> oneway roads).

@tordans added some info here Proposal:Make cycleway:both the default to indicate both sides - OpenStreetMap Wiki in his last edition.

You can “resolve” these ambiguities as a data consumer by interpreting them one way or another, … but another data consumer may interpret them yet another way.

[balchen]

@westnordost

@tordans added some info here Proposal:Make cycleway:both the default to indicate both sides - OpenStreetMap Wiki in his last edition.

The examples are hypothetical, so the primary question is how many of these translate into real-world cases. If (mostly) everyone already agrees on the same interpretation, the issue remains hypothetical.

@westnordost

The opinions have been turning in favor of rather mapping bike paths (and sidewalks) rather on separate lines, especially on bigger roads, however, mapping bike paths and sidewalks on the road-line has always been and continues to be correct.

It was not my intention to agree on using one instead of the other, but rather to highlight that if tagging everything on a single line causes issues, perhaps the solution in these situations is to not tag everything on a single line.

Using highway=* + cycleway=* to mean that there is a completely separate road meant for cycling that runs somewhat adjacent to the road represented by the line seems to be an endeavour that’s essentially dead in the water in terms of real-world practicality. You can do it, but you will face issues at some point, and when you do, the best solution is probably to draw a separate line for the cycleway. Many of the hypothetical issues raised seem to arise from this usage, but that would/will be clearer if real-world cases are presented.