[talk-au] National Park unclosed poly, want advice

[Nathanael Coyne]

Thanks Paul, I'll see if that's an option, especially as it's not a named
forest.

Now I've also got this situation where the much larger but very low
fidelity forest poly is conflicting with my new high-fidelity poly for the
actual Murramarang National Park

https://www.openstreetmap.org/#map=14/-35.6942/150.2560

I don't know whether to clean up the edge of the larger forest poly to use
the same coastline nodes as my national park poly because there's also the
issue that the national park probably can't be considered a subset of the
forest poly as there are non-forest regions within the national park!



Nathanael Coyne (Boehm)

www.purecaffeine.com

Canberra, Australia
0431 698 580

On 22 April 2015 at 05:41, Paul Norman <penorman at mac.com> wrote:

> On 4/18/2015 11:35 PM, Nathanael Coyne wrote:
>
>> I'd appreciate if someone could have a look at it and tell me if it is
>> valid or invalid and if invalid then where I should start with fixing it
>> up, given it covers such a large area.
>>
>> http://www.openstreetmap.org/way/204064026#map=10/-35.8657/150.1611
>>
> The way itself isn't a polygon, but is part of the multipolygon
> http://www.openstreetmap.org/relation/1388357
>
> It is valid, but an area this large is a problem to deal with for a number
> of reasons
> - You're more likely to conflict with people editing it
> - Checking validity is hard, as you need the entire  polygon
> - It poses problems for rendering toolchains
>
> Coming from an area with many areas of trees, I have a few techniques. One
> is to end up splitting on an arbitrary line. While this fixes the problems,
> it's not ideal. A better route is to find an appropriate break in the trees.
>
> Typically along roads, rivers, power lines, and oil and gas lines there is
> a break in the trees. I try to find one and trace this break. This also
> tends to find clearings (
> https://www.openstreetmap.org/#map=15/-35.8707/149.9854), unmapped roads
> branching off, and other features of interest.
>
> Don't feel you need to be perfect, particularly in a case like this where
> the original tracing is from landsat.
>
>
> _______________________________________________
> Talk-au mailing list
> Talk-au at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-au
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20150422/bb26e048/attachment.html>