[OSM-talk] Tiny island coastline errors

[Cartinus]

On Tuesday 25 November 2008 22:05:31 Martijn van Oosterhout wrote:
> Not sure if it was exactly this problem, but it is due to the size. I
> don't know if the word "island" is appropriate, more likely a large
> rock :) Did someone actually get on to the thing to measure it? It's
> certainly way smaller that you can accurately measure with a GPS
> device.
>
> I'm looking for the cause. The bit of code I thought it would be
> doesn't appear to be it after all.
>
> Does it actually make sense to tag such things as coastline anyway?
>
> Have a nice day,

I finally had time to look at this again.

I first thought the error in lake Erie was probably an artefact of the Landsat 
pixel size / PGS import process. However it is clearly visible on the Yahoo 
Imagery. There it is however slightly bigger than the PGS data suggests. I 
stretched the island to match the aerial image and it now has a diameter of 
10m in stead of 6m. Lets see if the error is gone tomorrow.

For the error at Cyprus I looked at the OSM data, the Yahoo Imagery and the 
next picture on Wikipedia: <http://en.wikipedia.org/wiki/Image:Rock.jpg>
My conclusion is that it is probably the white spot in the waves in the middle 
of the picture. So I changed the tagging.

I left the error in Estonia alone. As it looks like someone reimported the 
Estonian coastline from some source other than PGS.

-- 
m.v.g.,
Cartinus