I'm not sure if this is what you mean by schema, but here's the Geo
draft as it stands on opensearch.org -
http://www.opensearch.org/Specifications/OpenSearch/Extensions/Geo/1.0/Draft_1
I worked on a discussion draft to be submitted to the Open Geospatial
Consortium, which has some small changes from the original:
- a "relation" element added, which provides for simple geometric
queries (within, overlaps, disjoint)
- the "geometry" element contains Well Known Text, which allows one to
specify complex polygons
The draft is here, aiming to become officially an OGC discussion draft
in the summer, and then I'll send Andrew patches to the version on the
opensearch site.
http://docs.google.com/Doc?id=dctb97jj_0jnxzwd4s&hl=en
As for the time extension, we have used it but I don't think it has
the same amount of acceptance, and I would be interested to hear about
others' usage or about alternatives.
http://www.opensearch.org/Specifications/OpenSearch/Extensions/Time/1.0/Draft_1
Can you tell us more about your project? :)
cheers,
jo
--
The library of congress has a nice extension to ISO 8601. It supports
ambiguity and pre-Christian Era dates. They have a good writeup on the
problems they were trying to solve with it as well as their proposed
solution.
http://www.loc.gov/standards/datetime/
Ralph
Okay that's great, thankyou.
It occurred to Andrew that perhaps Tie meant an XML schema, like the
one found here:
http://lastfmapi.svn.sourceforge.net/viewvc/lastfmapi/lastfmlib/trunk/lastfmlib/src/jaxme/OpenSearch.xsd?view=markup
I'm not sure how this would be useful to OpenSearch client developers,
but it would probably reassure the OGC to see one, so I can look at
making one in the nearish future.
thanks,
jo
--
The changes are aimed at providing more power to spatial queries while taking
advantage of database capability that anyone answering these spatial
queries is likely to have on the backend (PostGIS, etc).
- a "relation" element added, which provides for simple geometric
queries (within, overlaps, disjoint)
- the "geometry" element contains Well Known Text, which allows one to
specify complex polygons
The latter replaces a tuple string format from the earilier version,
but one could feel free to support it for backwards compatibility reasons
and we should mention this. This element is for specialists anyway,
most queries will use the simpler geo:box search area, This is
basically my request after prototype implementation.
The former is a subset of the OGC Simple Features Interface. It has
been requested by several people. It would be extremely useful when
you have data sets at different scales for an area and you can't infer
the level of detail that is being asked for.
That is about it, most of that proto OGC draft is boilerplate and layout.
cheers,
jo
--