[OSGeodata] geodata schema

[Jo Walsh]

dear Norm,
On Tue, Aug 01, 2006 at 04:03:31PM +0100, Norman Barker wrote:
> Thanks for the pointer on Geo-Wxnkers to this group.

Thanks for coming to join in here and expanding the power of OSGeo's
formidable borglike group mind... :)

> https://geodata.osgeo.org/source/browse/geodata/trunk/metadata/ and
> (the svn checkout for 'guest' seems to be broken - is there no guest
> access?)

I didn't realise there was not :/ I couldn't see how to enable it in
the overview CollabNet documentation ... I've asked their elusive 'support
people' how I can do this. In the short term, I'd advise to register for an
account on the website at http://www.osgeo.org/ - request a 'project 
contributor' role - and checkout with your username... I deeply regret
that this whole process is so unnecessarily complex.

> Although I know python quite well, I would prefer to develop in Java if
> possible.
 
It would be great to have multiple language implementions of the same
idea. (And Chris is always keen to 'embrace and extend' everything into
GeoServer ;) ) The few pieces of work i've done are more focused on
indexing - the RESTful part is just a "insert someone else's better
idea here" stub. Chris Schmidt is interested in working on editing
frontend and an OpenLayers-driven browsing interface. My own plan for
the next 10 days with the python implementation is to 

- get this working with Sean Gillies' OWSLib and use that to start
  indexing GetCapabilities
- revise the schema a bit based on use cases that have come up here
- get some basic GeoRSS-out going on the RESTful side
- plug into whatever Chris can spare the time / energy to produce
- have something to demo by next Thursday (another irc meeting...)

> My motivation is to avoid the CSW-ebRim profile, and to encourage the
> uptake of RESTful catalogs, and further explore this rapid catalog
> development and see if enough metadata can be captured. My motivations
> are also from a business perspective - more catalogs mean more chance to
> utilise meaningful data and a better use of clients.

This sounds great! and very much on the track we're going down here[0].
> 
> Whilst I think 1 - 3 are quite easy, have there been any plans for
> develop the API for the catalog front end?

I need to catch up on the discussion which i missed last week - i'd
like to find a starting point of common practise to work from - rather
than worrying over design too much - and extract Stefan's thoughts
and others' into a page on the OSGeo wiki.
http://wiki.osgeo.org/index.php/Simple_Catalog_Interface could be a
place to braindump / filter into.

> Have you chosen a license for the database schema and python scripts you
> have already implemented? 

I am bad at thinking about that sort of thing. There's not much there
to license yet, to be honest. ;) BSD or GPL would work fine for me.

> I would be looking to hack + and then document (1) - (4) in the next 5 -
> 10 days.

I really look forward to seeing how you get on and goodness knows i
need the encouragement-by-example to get into gear on this... :)

cheers,


jo

[0] i submitted a talk to FOSS4G 2006 on this whole topic - basically
gleaned from the collective thinking of the group here over a series
of IRC chats about geometadata (i knew nothing about the domain to
start with... i am a semantic-web-head who drifted into geospatial
more or less by accident...) - in the hope if they accepted the talk
i'd *have* to shut up and write some code :) - this was back before
the RESTful catalog conversation bloomed again. 


Have A Nice Metadata
-------------------- 

Managing metadata well is the key to making a repository of open
geographic data really useful and re-usable. Creating metadata can 
be a dull chore. Part of the problem is over-focus on production, 
rather than consumption; on standards, rather than on usage. The 
Geodata Committee at OSGeo has been working on a "simplest
useful thing" approach to geographic metadata. This talk discusses
ways in which metadata creation can be made easier (by automatic 
extraction from standards based formats), more fun (through 
'conversational' irc and email interfaces rather than
filling in forms) and more useful (via GeoRSS syndication and
machine-assisted web-based search).