[Auscope-geosciml] RE : Testbed proposition [SEC=UNCLASSIFIED]

Boisvert, Eric Eric.Boisvert at RNCan-NRCan.gc.ca
Tue Mar 2 15:15:14 EST 2010

I though this was called a 'profile', when you decide that part of the model you want to implement and what should be supported and how.  I see this as an architecture level decision - Am I right ?

-----Message d'origine-----
De : Ben Caradoc-Davies [mailto:Ben.Caradoc-Davies at csiro.au] 
Envoyé : 2 mars 2010 00:33
À : auscope-geosciml at lists.arcs.org.au
Cc : Boisvert, Eric
Objet : Re: [Auscope-geosciml] RE : Testbed proposition [SEC=UNCLASSIFIED]

On 01/03/10 18:12, Boisvert, Eric wrote:
> Fine with me, but this is more testing available mapping technologies than testing the model itself.

Not quite. Implementations can choose to support a subset of use cases. 
The reference data set, queries, and instance documents need not be constrained to those that can currently be mapped. Implementers can report the use cases they support.

By providing the reference data set in an RDBMS-compatible format (e.g. 
SQL), you increase the likelihood that an implementation will be able to build a test suite. If the reference data set is in a persistence-technology-neutral form, it might still work, but we run the risk of adding more layers of complexity. Can you imagine test data in XML, or JSON, and the things that could go wrong getting it into an RDBMS? I suppose we could have the test data in the form of simple features.

Kind regards,

Ben Caradoc-Davies <Ben.Caradoc-Davies at csiro.au> Software Engineer, CSIRO Earth Science and Resource Engineering Australian Resources Research Centre
26 Dick Perry Ave, Kensington WA 6151, Australia

More information about the GeoSciML mailing list