On Thu, 26 Sep 2019 at 19:17, Martin Koppenhoefer <dieterdre...@gmail.com> wrote:
> > > > One question is should they be rendered, and most > > people seem to agree that they should. Should the buildings be tagged > as disused? So > > the wiki implies. If they should be tagged (in some way) as disused, > then how? > > you could add building:use=no > or building=* disused=yes > I've used the second of those two. disused:building doesn’t make a lot of sense I don't think there is any meaningful semantic difference between that and adding disused=yes. The building is in the disused part of its lifecycle, both express the same thing. But one way (currently, on standard carto) causes the building to render and the other does not. and will make the buildings disappear from many applications. And we're back full circle. I agree with you on that one. Kevin Kenny put it eloquently that he's not lying, he's just telling the truth in a way the renderer understands. What he didn't say is he was talking of how standard carto currently does things and he, like I, hope it will continue to do so. If standard carto suddenly stops rendering physical objects with disused=yes then I will stop using that tag. If standard carto suddenly starts rendering disused:amenity=pub with a pub icon then I will stop using that tag. But others have chimed in saying the renderer is perfectly entitled to render a disused building (whichever way we tag it) in any way it wants. Which is true. But it would be nice if we had a degree of coherence across OSM such that standard carto (at least) could agree to support certain expectations. Standard carto COULD choose to render motorways the same colour as rivers. If they ever did, some mappers would tag motorways as primary highways with appropriate lane counts and some mappers would see it as no longer being worthwhile mapping motorways. We don't have any promises from standard carto that they will never render motorways the same way they render rivers, we assume they wouldn't choose to do so. In the case of disused objects, two ways of tagging produce two different results, each of which is desirable in specific circumstances, and it would be nice to have an assurance that we could rely upon that behaviour (or something like it achieved with different tagging) in future. BTW, I found a different problem with a recent change to standard carto. I won't say what it is or this thread will derail further. I can't fix it by choosing alternative tags (valid although perhaps discouraged). I can't fix it by tagging for the renderer and lying about what's really there (not that I'd do that even if it were possible). The only possible way to get a satisfactory rendering is lying about the position of a very significant feature, and I refuse to do that. So I no longer map features of that type - there are plenty of other things to map and I won't waste my time on features that render misleadingly badly. The data, prized above all else by some, won't be corrupted by me but it will be less complete than it could have been. -- Paul
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
