[Auscope-geosciml] GeologicSpecimen

[stephen richard]

OK, I found the rest of this discussion (Which O&M version for 
GeoSciML). I think using the ISO19156 O&M model makes more sense at this 
point.

I don't think we need to specialize SFN_Specimen to restrict 
materialClass. That can be done at the application profile level.

I also don't think we need a specialization to account for parent child 
sample associations. The role property on the association is important 
to distinguish different kind of parent child relations (split, mineral 
separate, insoluble residue....). Admittedly these relations are 
convolved with processing method, but I think they are different kinds 
of part-whole relations that are useful to distinguish. Keeping the 
generic SamplingFeatureRelation allows representation of these using 
controlledVocabulary. Goes back to the philosophical question of 
restricting semantics in the xml schema or using document content (hard 
vs. soft type). Guess I'm a soft typer here.

SFN_Specimen  inherits  parameter:NamedValue from SF_SamplingBase. The 
parameter property describes 'an arbitrary parameter associated with the 
SF_SamplingFeature. This might be a parameter that qualifies the 
interaction with the sampled feature, or an environmental parameter 
associated with the sampling process. The semantics of the parameter 
shall be provided as part of its value.'  NamedValue is essentially a 
Key:value pair. This could be used to capture geologicSpecimenType.  The 
specimenType remains the only compelling rationale to me for a 
GeologicSpecimen specialization.  As far as adding a specimen type on 
SFN_Specimen in ISO19156, its a soft type vs. hard type question. Lets 
propose to ISO (someone who has authority to submit request...) and let 
them decide if they like soft typing.

Relation from specimen to sampled feature (provenance) "provenance 
examples (windmill, water bore, monitoring bore, RC drill hole, outcrop) 
"  These can be represented using the *SF_SamplingFeature-> 
sampledFeature-> GFI_Feature* association to any feature, to capture one 
or more domain features that are sampled.  No problem here.

one other issue--the current GeologicSpecimen has a *currentLocation 
*property, which I assume is a curation location. This doesn't exist in 
SFN_Specimen.

So, if we're OK using SFN_Speciment.parameter.NamedValue key-value pair 
for specimen type, we don't need geologicSpecimen. Query filters are 
somewhat simpler if there's an explicit SpecimenType attribute, so my 
balance tilts slightly towards specializing a GeologicSpecimen from 
SFN_Specimen. If we need a currentLocation, then theres a better case 
for a GeologicSpecimen.

steve


Use of

Oliver.Raymond at ga.gov.au wrote:
>
> Hi Bruce,
>
>  
>
> The previous email between me and Gilly is attached.  I haven't 
> received any comment from anyone else.  I'll put the email 
> correspondence on to the Twiki.
>
>  
>
> 1. The attached email outlines Gilly's reason for creating 
> GeologicSpecimen/materialClass - basically just constraining it more 
> from GenericName (in O&M) to ScopedName (in GeoSciML).  I am a bit 
> ambivalent as to the benefit of doing this - I'll let you and Gilly 
> discuss if you want to.
>
>  
>
> 2. The scope notes for GeologicSpecimenType are:
>
>  
>
> "GeologicSpecimenType describes the broad range of specialised 
> geologic specimens. 
>
> eg:
>
> outcrop specimen,
>
> float specimen,
>
> drill core,
>
> rock chips,
>
> drilling mud,
>
> dredge sample,
>
> thin section,
>
> powder,
>
> mineral separate,
>
> mineral grain,
>
> mineral grain mount,
>
> probe burn spot .
>
> Use the samplingMethod property to provide more details of how the 
> sampleType was obtained."
>
>  
>
> The scope notes for GeologicSamplingProcess are:
>
>  
>
> "Use GeologicSamplingProcess to indicate the process used to obtain or 
> create the GeologicSpecimen. eg:
>
> diamond drilling
>
> percussion drilling
>
> piston core drilling
>
> vibro core drilling
>
> channel sampling
>
> sea floor dredging
>
> crushing
>
> mineral separation
>
> melting."
>
>  
>
> I (and the GA databases) think they are two separate concepts that 
> need description.
>
>  
>
> 3. On a related issue, I think we need to relace 
> geotime:GeochronSpecimen with gsml:GeologicSpecimen.
>
>  
>
> Cheers,
>
> Ollie
>
>  
>
> /------------------------------------------------------------------------------------------------/
>
> /Ollie Raymond
> National Advice,  Maps and Standards Project/
>
> */Geoscience /**/Australia/*
>
> */ /*
>
> *Address:* GPO Box 378, Canberra, ACT, 2601, Australia *|* *ABN:* 80 
> 091 799 039
>
> *Ph:* (02) 62499575 *|* *Fax:* (02) 62499992 *|* *Email: 
> *Oliver.Raymond at ga.gov.au
>
> *Web:*  
> http://www.ga.gov.au/minerals/research/national/nat_maps/nat_geol_maps.jsp
>
> *Google Map 
> <http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1> 
> *
>
>  
>
> -- This message was created with 100% recycled electrons --
>
>  
>
>  
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengeospatial.org/pipermail/geosciml/attachments/20091020/3001c562/attachment.htm>