Page tree

Versions Compared


  • This line was added.
  • This line was removed.
  • Formatting was changed.

Session 1- What is a sample?
(OBI) "Material Sample" == "Specimen" - to support a scientific investigation 
(SOSA) Feature which is intended to be representative of a FeatureOfInterest on which Observations may be made. 
Note intent - indended intended to be representative - there is an intention to make observations. 
Why do we use he term Feature? Comes from physical sciences side. We can conflate with "Thing".
The statistics community will attack our example about our definition of sample (for them they use 'sample from a population/distribution').
Sample vs specimen - specimen is a subset of sample? spatial samples? 
Should adapt to 'one or more' Features as __representative__ of a FeatureOfInterest. Can defend against statistical definition. Sample can be assigned to the bigger thing.
Transects/boreholes/faces/etc - Are these samples? These subsample spaces. Ideas from fluid earth guys - characterise the atmosphere/ocean along transects.
Ideas from establishing base stations, with the intest intent to measure around base stations
Sample - act of restricting some larger space, and then the data are artifacts that come from the sample (photographs, etc)
So _not_ just material samples.
A sample fills a role.
Issues over 'transects' - also used as sampling protocol in life sciences. 
Example of flight lines in geophysics, _observations_ are magnetics etc constrained along flight line.
Sample relationships are not a closed set - the ways samples can relate to other samples are legion.
`[Sample] -- isSampleOf --> [Thing]` Part of intention of sampling is the 'thing'. So resampling the same outcrop, we sample the same thing but get a new sample. 
Every sample is a product of an activity (that is bounded in time, and maybe in space).
In IGSN world we start with pieces of rock, now what do we do with boreholes?
Repositories have scope - e.g. ORCID = people, Something else for cruises, etc etc. What is the scope of IGSN?
Action items:
- General agreement with definition
- We need a simple document that has the definition that can be understood by 'non-ontologists'
- We need different versions for different communities (translations)