[OSGeo-Standards] Inline GML? Good idea poorly executed

Jody Garnett jgarnett at refractions.net
Wed Oct 24 21:13:59 EDT 2007


Jody Garnett wrote:
> I find it consistent with inline GML for SLD; but I also hated it 
> there as well :-) For a while SLD was being hacked up really badly, 
> when they added inline GML and additional WFS connections becomes a 
> really bad "context" document. Basically trying to slave an external 
> WMS to draw content and messing up several clear standards in the 
> process.
>
> My thought was always to:
> - have a good context document; purpose is to describe a Map
> - have a good style document; similar to FeatureTypeStyle wrapped up 
> as a document
> ...
>
> At the end of the day I would like to have
A few people commented that I left them hanging; sorry for the cut and 
paste problem (composing email while waiting for a build - like a good 
developer). The point I wanted to make
 is earlier in the email; a clear separation of concerns is important. 
The SLD specification should be split into a context document and a 
style document; not made to serve both masters.

The split has started with SLD 1.1; getting OWS Context working is the 
next step. Deciding where the appropriate place for "markup" to live is 
part of the journey.
Jody


More information about the Standards mailing list