[Tagging] Feature Proposal - Voting - Rivers Classification

[Kevin Kenny]

On Oct 17, 2017 4:53 AM, "Christoph Hormann" <osm at imagico.de> wrote:

On Tuesday 17 October 2017, Kevin Kenny wrote:
> It's impossible to base a rendering decision on something that isn't
> represented by any tag.

That is not true, you can produce a lot of information through analysis
of the data and by connecting it to data outside of OSM (which is
usually outside of OSM because it is outside the scope of OSM).


All good points. I'm afraid I get a bit prickly because I have several
times been told that entering field-observable attributes of actual
geographic features is 'tagging for the renderer'  - I think simply because
whoever was spouting off was not interested in those features.

In any case, when I said 'something', I meant 'some thing' - an actual
observable object.

I concede that 'relative road importance' strains that definition. But I
fail to see where any conceivable renderer would be able to get the
information if we don't tag it. 'Relative importance' is not needed for
symbology - that's determined by physical attributes (carriageways,  lanes,
shoulder width). It is, however, what would guide a rendering decision
about the appropriate zoom level at which to display a way. Some ways that
are pretty awful, physically, nevertheless should be shown on relatively
small scale maps because they're the only road connections among
significant communities.

For what it's worth, except for 'motorway' and the problematic 'trunk', the
Wiki definitions all are based on relative importance, not physical
attributes. Arguably they're wrong, but a lot of data have been entered
following them. (And a lot of bad data have been imported from TIGER or
foisted upon us by NE2. Don't get me started.)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20171017/8a911044/attachment.html>