[OSM-talk] Fw: calling all python coders for slippy map help

Mikel Maron mikel_maron at yahoo.com
Thu Nov 16 14:48:23 GMT 2006


This is my understanding, from Artem's explanations..

Mapnik is using PostGIS, with the schemas optimized for rendering. 
For instance, a way is fully represented in a single row .. both its geometries and tags.
In a sense, the additional joins currently employed to get renderable data are done offline, 
by the load scripts. 

For the full, editable, revisionable database, mysql is suited fine to the task.


Once planet.osm is loaded, daily (or weekly) changes will be loaded much quicker,
since it only requires production of the diff of planet.osm. 

-Mikel

----- Original Message ----
From: SteveC <steve at asklater.com>
To: Nick Whitelegg <Nick.Whitelegg at solent.ac.uk>
Cc: talk at openstreetmap.org
Sent: Thursday, November 16, 2006 2:35:44 PM
Subject: Re: [OSM-talk] Fw:  calling all python coders for slippy map help

* @ 16/11/06 11:07:52 AM Nick.Whitelegg at solent.ac.uk wrote:
> >Isn't it strange to create an xml file from a database and then reinject
> >it in another db ?
> >It seems (but i could be wrong) that most of the people who is using the
> >planet.osm always start by doing a database from it (or parse it to
> >extract a small area, which could be done easily from a db).
> 
> >Wouldn't it be more efficient to generate the planet file directly in
> >SQL format ?
> 
> I *think* this is just a temporary requirement, to quickly get Mapnik 
> tiles of the whole world.
> 
> I believe the longer term plan (Steve?) is to store OSM data in a PostGIS 

No, not at all.

have fun,

SteveC steve at asklater.com http://www.asklater.com/steve/

_______________________________________________
talk mailing list
talk at openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk







More information about the talk mailing list