> On Aug 12, 2020, at 4:49 AM, Paul Allen <[email protected]> wrote: > > On Tue, 11 Aug 2020 at 19:19, Tod Fitch <[email protected] > <mailto:[email protected]>> wrote: > > It occurred to me that the area where water flow disappears is indeterminate > [1], thus the problem mapping it. > > Ordnance Survey represents this as a sort of star of very short waterways > at the approximate point of disappearance and labels them "spreads." > > The map is representative. We can tolerate precisely specified amounts of > doubt and uncertainty. The name "spreads" indicates the indeterminacy even > if we map it as a node. Just as we render a spring as a circle on the > map, an asterisk would do for spreads. > > Perhaps a “indeterminate=yes” tag on the last node of a water way that > “peters out” [2] could be used to signal the QA tools that the end of a water > way is not a mistake. > > If we tag it as waterway=spreads the indeterminacy is implied and QA tools > can be happy there is no mistake. > > Might be useful in cases other than an ephemeral water way in the desert > though I haven’t thought of one yet. > > Useful in coastal waterways that peter out in sand above high water. Or > coastal > waterways that peter out in sand just below high water when the tide is out - > they > haven't carved a channel down to the low water mark, they just vanish into the > sand (but QA tools won't have a problem with those if they connect to the high > water mark). And yes, there are coastal waterways that carve a channel > through > a beach right down to low water and others that just peter out on the beach > close to high water. >
To clarify it for me, the a “waterway=spread” tag would be used on a node (rendered possibly as an asterisk) or on a way? Or either depending the situation? Thanks! Tod
signature.asc
Description: Message signed with OpenPGP
_______________________________________________ Tagging mailing list [email protected] https://lists.openstreetmap.org/listinfo/tagging
