[Tiling] FOSS4G BOF?

[Oliver Tonnhofer]

Hi tilers,

there are lots of things to discuss in the tiling BOF and I want to attend the Python BOF as well. Maybe we can find some time at the code sprint?

On 20.08.2011, at 19:19, Gabriel Roldan wrote:
> But the problem we face is that (to the extent of my knowledge) there
> are no pre-defined "grid sets" for EPSG coordinate reference systems
> so it's hard to configure "tiling schemes" or "gridsets" for CRS's
> other than, say, 4326 and 900913. Not to talk about making them
> "inter-operable" like in making tiled layers coming from different
> tile servers match.
> Now, the EPSG database contains each (or almost each) CRS's area of
> validity with unit of measurement information and all. But also I
> guess most servers don't need/want to rely on the EPSG database as
> it's just easier for them to work on pure cartesian space with some
> independently of most CRS details but perhaps bounds/tile origin
> coordinate.
> 
> So If this makes sense may be we could try to define a common/shared
> database of pre-defined "tiling schemes"/"grid sets" for most EPSG
> CRS's based on the EPSG database information?


I like the idea and the EPSG database would be a good starting point to create these grid definitions, but I see one issue with it: How do you define the resolutions of the levels? For some people the resolution doesn't matter and they can live with factor 2 between each level, but most of our customers have GIS clients that show maps at fixed scales (at least often enough to cache at these scales). I think it is nearly impossible to come up with a common set of scales and even if, these scales will result in varying resolutions with different software (72/96/OGC dpi).

Regards,
Oliver

-- 
Oliver Tonnhofer    | Omniscale GmbH & Co KG    | http://omniscale.de
http://mapproxy.org | https://bitbucket.org/olt | @oltonn