[OSM-dev] Blue sea for tiles at home

Frederik Ramm frederik at remote.org
Tue Apr 24 07:14:04 BST 2007

Hi Lars,

> When it works, it looks great.  But there are many places where it 
> doesn't quite work.  Perhaps this is because of bad data, e.g. 
> line segments pointing in the wrong direction, or coast lines 
> being broken.

Yes, but I hope we'll fix that quickly.

> Would it be a terribly bad idea to inject these datapoints into 
> the OSM database, i.e. nodes with water=yes or water=no, or 
> something like that? 

I was thinking of doing this not for oceans (because we'd have to spread 
nodes over 70% of the world which would, to a superficial analysis, make 
the ocean tiles "non-empty" because they have a node on them) but, in a 
later version of the software, for larger inland areas like very large 
lakes or forests. But that would first need some changes in 
close-areas.pl to work with these as well, currently it does only 

> If you do this, then anybody can correct 
> errors by editing the attribute of that node.  We could even 
> clarify by adding more such points at the middle of islands and 
> lakes.  Instead of a special lookup in that table, the information 
> would come as part of the normal API request.

I too think it is important that everbody is able to make changes but 
until now thought it sufficient that everybody can edit the coastline, 
and hoped that Martijn would complete the index file to reliably cover 
the whole world. Do you have issues with the index file in Europe?


Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00.09' E008°23.33'

More information about the dev mailing list