[OSM-talk] Highways tagging vs Polygon

Nic Roets nroets at gmail.com
Thu May 21 20:34:57 BST 2009


My opinion is that all defaults should be global. We should not have any
country or urban / rural specific defaults. It will mean most ways will need
a lot of extra tags. So we may need to improve the editors to make it easier
to add all those tags. For example give the editors modes like "Rural UK"
where they apply the defaults when new ways are created.

But it will make it a lot simpler for mappers. If you see a "No Cycling"
sign on a trunk road and want to compare it to the DB, then you don't need
to think about where the country border polygon ends.

None of the arguments against it (e.g. countries frequently changing the
speed limit, larger planet file etc) hold any water.

But my views may no be shared by the community.

On Thu, May 21, 2009 at 8:43 PM, Richard Bullock <rb357 at cantab.net> wrote:

> Am I missing something, or can we not just assume that e.g.
>
> each highway=residential has a speed limit consistent with urban areas in
> that country - unless explicitly tagged otherwise
>
> e.g. in the UK, highway=residential would be 30mph (48.28032km/h), unless
> tagged as something else.
>
> each highway=trunk, primary, secondary etc. has a default speed limit
> consistent with rural areas in that country - unless tagged otherwise.
>
> each highway=motorway has a default motorway speed consistent for that
> country.
>
> That way, we only need to tag major through routes in cities, and rural
> roads which do not have the usual national speed limit.
>
> The same thing applies for access restrictions. We don't need to tag that
> roads are accessible for cars - that should be assumed unless tagged
> otherwise
>
>
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20090521/2aa9817d/attachment.html>


More information about the talk mailing list