[Aust-NZ] GeoNetwork Annual Workshop [SEC=UNCLASSIFIED]

[Simon Pigot]

Bruce.Bannerman at dpi.vic.gov.au wrote:
> Now having said that, just how different is the Australian Profile 
> from ISO 19115 19139?
>
>
Hi Bruce and John,

If you're asking just simply about the differences between the 
Australian profile and ISO19115/19139 then I'm sure John Hockaday and no 
doubt others have presentations that will give all the details (I think 
John's presentation to the metadata open forum in Sydney would be a good 
place to look).

If the question is wondering why a profile which isn't vastly different 
to ISO19139/19115 took a while to implement in GeoNetwork then maybe you 
need to think about some of the surrounding issues. First and foremost, 
there is the question of schema parsing in GeoNetwork to build an 
editable representation of any 19139/19119 based metadata record - this 
was working in the BlueNet MEST 1.0 release in October/November 2007 
(after the OSDM sponsored GeoNetwork workshop) and in trunk 2.1 - but 
some of the more arcane stuff wasn't sorted out until Dec 07 (it was 
incorporated in the trunk for the 2.2 RC0 release in the first week of 
Jan 08).

At the same time, there was the problem of a general method of 
supporting more than one profile of ISO19139 (a variation on existing 
support for different metadata standards) in GeoNetwork and the extra 
problems of adding support to GeoNetwork to find and use the ancillaries 
associated with the different profiles, including:

- schematrons for content checking: CSIRO developed them for ANZLIC, we 
included them in the validation report and added support for error flags 
in the editor
- xslts to convert to/from the new profiles: developing (we developed 
one for the Marine Community Profile, and there was the ANZLIC supplied 
on), testing and including in the tools
- menus to support profile groupings of elements eg. ANZLIC Minimum, 
ANZLIC Core for data sets etc

and probably others that I've (happily!) forgotten.

Ideas for implementing profiles were also not very accessible - we 
followed discussions on SEEGRID and used the WMO profile as an 
additional example - and it follows that there were very few (if any) 
accessible metadata tools that handled profiles and ancillary info at 
that time.

Of course, like most things, this will be looked upon as 'straight 
forward' with the benefit of hindsight and as it moves into the main 
stream :-).

Hopefully this might provide some of the answers to your question 
(assuming I've understood it correctly)!

Cheers,
Simon