[auscope-geosciml] RE : reference to ISO19139 in GeoSciML 2.1 [SEC=UNCLASSIFIED]

Oliver.Raymond at ga.gov.au Oliver.Raymond at ga.gov.au
Fri Oct 8 00:44:54 EDT 2010

Hi Ben,

Apologies.  These are the days that I have dark thoughts about giving this game away.... :-P

All along, I have been looking at the xsd files on the subversion tag called "2.0_final".  I see now that these v2.0 schemas are not the same as those on www.geosciml.org which are v2.0.2.  The v2.0.2 schemas have all the correct 'byReference' tags and have no imports of gmd schemas (that change was made from v2.0 to v2.0.1).  So my email this morning about validation and vocabulary.xsd and earthmaterial.xsd having erroneous gmd references is bunk.

It appears that the v2.1.1 schemas have been edited from copies of the "2.0_final" schemas, not the v2.0.2 schemas.  Hence the gmd imports wrongly appear in the v2.1.1 schemas.  That in itself is easily fixed.

But... I have done a lot more research into v2.1.1 backward compatibility and the OGC versioning rules.  I'm pretty sure now that fixing the problems around v2.1 ShearDisplacementStructure cannot be properly done as a v2.1.1 patch release under OGC versioning rules. Our desired fix (ie, editing the v2.0 geologicStructure schema), while it is a simple fix, is not backwardly compatible with the existing v2.1 services delivered by OneGeology-Europe and is therefore illegal under OGC rules as even a minor version release (eg, v2.2).

In conclusion, I think that the v2.1.1 schemas should be removed from www.geosciml.org , and that the fixing of the ShearDisplacementStructure bug has to wait until v3.


-----Original Message-----
From: Ben Caradoc-Davies [mailto:Ben.Caradoc-Davies at csiro.au]
Sent: Friday, 8 October 2010 2:05 PM
To: auscope-geosciml at lists.arcs.org.au
Cc: Raymond Oliver; Francois.Letourneau at RNCan-NRCan.gc.ca
Subject: Re: [auscope-geosciml] RE : reference to ISO19139 in GeoSciML 2.1 [SEC=UNCLASSIFIED]

On 08/10/10 06:37, Oliver.Raymond at ga.gov.au wrote:
> 1.  The ISO19139 import in GeoSciML v2 is not an unused import.  The import is required for the GeoSciML v2 schemas to validate because they refer to ISO19139 MD_Metadata elements, even though those elements are "byReference" and can't be encoded inline.

Are you sure that the import is required if the gmd elements are used
only in appinfo? I do not think appinfo is validated. I thought it only
had to be well formed. The gmd reference is in the text content of the
element as well.

This is valid:

Kind regards,

Ben Caradoc-Davies <Ben.Caradoc-Davies at csiro.au>
Software Engineering Team Leader
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre

More information about the GeoSciML mailing list