Can someone give an example? I don't understand this proposal. Anthony 04:48, 1 February 2010 (UTC)
- +1 Me neither. I think an exemple "in real life" + "as an .osm file" could help us understand the idea and how to tag it. sletuffe 14:24, 28 March 2011 (BST)
- +1 second that. please provide two or three real life examples with links to the IDs in the OSM database and/or provide some mockup screens. i think that would help non native speakers like me a lot. --Flaimo 12:19, 11 May 2011 (BST)
Just more barrier attributes ?
If there is e.g a fence or wall between two ways, they can be tagged as e.g barrier=fence on a line and height=*. If the barrier is there they can be mostly tagged as barrier=* Why should I use a relation for that? For lowered kerbs this is more useful, if they are all the same in this street or city, if not, they can be tagged at the particular highway=footway segment. --Fabi2 17:33, 18 April 2010 (UTC)
- the difference is that this proposal lets you do routing across linear connections, i.e. you can model connections between ways without really connecting them (or connecting them for routing before they really connect). Here come the barriers into play. You can define general (this means just tags, no geometry) barriers (e.g. railing, curb) and exceptions (geometry like nodes or ways) from that (lowered_kerb, "barrier=entrance", ...). Tagging them on the footway segment doesn't show where they connect to (e.g. which side in complex situations) and where they exactly are.
- You would be able to draw the barrier explicitly like you describe (but you would not have to - at least not for routing) and simply add it into the relation as barrier together with the adjacent ways. This is maybe not so interesting for barrier=fence but gets more useful for barrier=bollard or barrier=kerb (or example, could be different: barrier=road_marking)
-- Dieterdreist 01:12, 18 May 2010 (UTC)
What the status of a proposal means
The status of a proposal only indicates where in the lifecycle it is (see here). It doesn't indicate that there is an alternative tagging method available. This proposal has been in draft for eight years. It hasn't had an edit for eight months. No one has commented on it in the last seven years. This proposal is dead. If it's not ready to go out for a RFC after eight years when is it ever going to be finished?
- you insist it is dead, but it has been improved just some months ago. As I find the time, I will make some illustrations because many people are not understanding it well.
- If you set proposals to "abandoned", people will believe the tags are not good. It already happened in the past, well needed tags without alternatives and with at least sparse usage and just a proposal have been set to abandoned and it has created a lot of confusion for years (in some cases, because the user had set tens or hundreds of proposals to abandoned). --Dieterdreist (talk) 22:27, 4 February 2018 (UTC)
- I'm not sure I understand. I arrived on this page looking for a way to route over an area without drawing a way. If this proposal allows that I can and will use it, because currently I am forced to draw ways where in fact there are none. 5 march 2018 Pelderson
- You have to differentiate. This proposal is about defining data, i.e. describing the world. It defines how to describe an area that is routable, even without needing to draw the complete outline. It also allows to define barriers on this area without the need to explicitly draw them (but you could optionally do it).
- How the data you define and upload to OSM is then used by different tools and data consumers is up to these tools. While you can define an area as routable, it does not mean that the current routing engines will support this automatically. --Dieterdreist (talk) 11:13, 5 March 2018 (UTC)