[OSM-talk] Why place matters, slides from Vanessa Lawrence talk

[Artem Pavlenko]

On 11 Jan 2008, at 13:09, David Earl wrote:

> On 11/01/2008 13:00, Artem Pavlenko wrote:
>
>>> (a) Mapnik works on planet, yes?
>> No, it works on postgis db which is populated with osm2pgsql from  
>> planet.
>
> Yes, I know that. I meant that it is coming from planet, not  
> directly derived from the main database like osmarender.
>
>>> So perhaps use the planet diffs to determine areas which have  
>>> changed and proactively mark all such areas dirty.
>> The problem is how to merge planet diff into postgis , I think  
>> Jonb has done some work/research in this area.
>> Are there existing tools (osmosis?) that given a planet diff would  
>> return 'dirty' areas?
>
> Everything listed in a planet diff is by definition dirty, yes. I  
> don't think you;d need to change the rendering process at all -  
> keep on converting the full planet to database; just have a new  
> means for marking dirty areas - derived from the lat/lons of all  
> nodes in the planet diff corresponding to the current planet, plus  
> the lat/lons of all nodes of all ways listed in the diff.

Marking 'dirty' areas by processing planet diff sounds reasonable but  
we only generate diffs weekly, afaik.
My understanding is that generating planet and planet diffs takes  
very long time at the moment which brings us back to the eternal
quest of improving main DB. I'm not even suggesting to use postgresql :D

> David