[OSM-talk] Coastline Update

[David Groom]

Paul

A big thank you for providing all of the data.

It has helped to greatly reduce the number of error points over the last 
week.

Regards

David

----- Original Message ----- 
From: "Paul Norman" <penorman at mac.com>
To: <talk at openstreetmap.org>
Sent: Saturday, March 31, 2012 4:26 AM
Subject: [OSM-talk] Coastline Update


>I have completed another coastline generation and it has uploaded. This
> version respects odbl=clean.
>
> The shapefiles are in their normal place at
> http://pnorman.dev.openstreetmap.org/coastlines/
>
> Included is a .osm file with all the error points.
>
> An overview can be found at http://maps.paulnorman.ca/coastlines2.png
> Detailed views:
> Great Lakes: http://maps.paulnorman.ca/coastlines-lakes2.png
> Europe: http://maps.paulnorman.ca/coastlines-europe2.png
> US West Coast: http://maps.paulnorman.ca/coastlines-west2.png
> Australia: http://maps.paulnorman.ca/coastlines-au2.png
>
> There are no significant multi-square flooded or dry areas. The following
> areas have significant number of error points:
>
> Pudget Sound in Washington State
> The mouth of the Columbia river in Washington
> The Eastern Australia coast
>
>
> The points indicated by the maps and by processedc_p files are where
> coastcheck encountered an error and had to guess where the coastline
> continues. These should generally represent transitions between ODbL clean
> and ODbL dirty sections of the coastline. Islands with no ODbL clean
> sections will not generate any error points.
>
> http://www.wightpaths.co.uk/coast/CT-only.php and
> http://suncobalt.homeip.net:82/coastline.php are two visualizations of
> errors but neither has yet updated to the new data.
>
> I hope to complete one more run of the ODbL-clean coastlines before the
> downtime.
>
> During the downtime I will be running a set of ODbL-clean and conventional
> coastlines (and a planet file).
>
> If diffs are available during the rebuild process I will be generating 
> them
> then and reloading my database when the ODbL planet is published.
>
> Technical details:
> This new run takes into account odbl=clean. It may not correctly handle
> 1. Objects that are dirty via a changeset override
> 2. Objects that WTFE reported clean but are now dirty
> 3. Certain sequences of edits and tag additions that are not likely to 
> occur
> frequently with coastlines and which require access to a full history
> database to evaluate
>
> The data is from 7 AM PST and the ODbL status is slightly more recent.
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>