Ok, we seem to be coming from this at different angles. My point is that 
a shapefile and a database is not a "Service", so why try to attach 
metadata to it, and why try to ask it questions which it cannot answer.

imho this is the job of servers like GeoServer and MapServer to store 
and manage this inofmration, not the job of the format driver. And 
indeed, that is how i look at the DataStore class, as a format driver, 
nothing more.

Anyways, the point of view from the udig guys is clear, they think this 
is a good minimum set of metadata. Fair enough. I would like to hear 
from other people.

And again I stress that we ask ourselves to look at this in terms of a 
*concrete* use case: a need that someone has currently, not a need that 
we think someone might have down the road.


>               shapefile     database     wfs        wms
>       title       x            x          x          x
>    keywords       x            x          x          x
> description       x            x          x          x
>   publisher                               x          x
>        type       x            x          x          x     Formally 
> schema        
>      source       x            x          x          x
>        icon       x            x          x          x
> 
> Notes:
> - Schema is used to indicate the kind of service this is ... looks like 
> you would like this encoded as an instanceof check? I will stick with a 
> URI but it would be good to call it "type".
> - Icon is needed by any client api; since we have an open ended set of 
> services (ie a plug-in system) we need to ask data source providers to 
> provide icons; we cannot guess up front.
> - the remaining entries requested by the osgeo data list have to do with 
> access fees and license or somethin


> 
> For information on the contents (called "resource" to match OWS specs)...
> 
>       title       x            x          x          x
>    keywords       x            x          x          x
> description       x            x          x          x
>        name        
>      schema       x            x          x          x
>        icon       x            x          x          x
>      bounds       x            x          x          x
>         crs       
>     
> Notes:
> - name - is the machine readable name; not needed here as it is part of 
> the data model already
> - schema - is the URI for the FeatureType; I really want to track this 
> as it lets me know when a specific Style can be used, a specific editor, 
> or data taken from one place to another etc...
> - icon - even more important this time out
> - crs is not needed as it is covered by getBounds().getCRS()
> - gabriel really wanted a latLonBoundingBox, it can be considered if you 
> wish...
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Geotools-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/geotools-devel
> 
> !DSPAM:4007,47aa4eb8227691961014482!
> 


-- 
Justin Deoliveira
The Open Planning Project
[EMAIL PROTECTED]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to