On 07/09/2018 10:59, David Woolley wrote:
On 07/09/18 10:47, Martin Wynne wrote:
The great advantage of this definition for mapping is that it is an
undisputed fact, on the ground.
You put lots of caveats into this, which leads lots of grounds for
disputes.
One thing to remember is that OSM is international and the
town/village/city concepts don't map cleanly to other cultures. Even US
English has a rather different concept of city.
I think it has to be mapped according to local conventions. Using US
terminology for British placenames, and vice versa, would result in
labels that are counter to the expectation of most users. There isn't a
global, one size fits all solution.
What I've seen, in the context of other countries, is population being
favoured as the determiner. Obviously you can get over-pedantic about
borderline cases.
I think population is the most useful fallback in the absence of more
relevant local information. Fortunately, in England at least (Wales and
Scotland are different!), we do have an accessible and usable source of
legally definitive information for most cases.
The thing that makes the UK difficult is that the tag values look like
the common language terms and match well enough to be right, a lot of
the time.
I don't see that as a problem. It makes it more likely that the right
tag value will be chosen by someone who is simply mapping from personal
knowledge rather than canonical sources. Which, most of the time, is
fine. We only need to revert to an agreed definition in cases where the
chosen value is incorrect.
Mark
_______________________________________________
Talk-GB mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk-gb
