[obs] Joining DwC, OBOE, PO and PATO

Simon.Cox at csiro.au Simon.Cox at csiro.au
Thu Oct 28 18:14:17 PDT 2010


In O&M we would first recognize that there is a feature-type 'occurrence' and then have observations of each of the properties of an instance.
The observations would have observedProperty 'taxon', 'location', 'time' etc, and could be packaged

(a)    as n separate observations, one for each property, but all observations having the same featureOfInterest (which would actually be just a URI)

(b)   as 1 observation with a complex-property, and a tuple as the result.
Observation is itself also a feature-type, and the 'relatedObservation' association can provide context.
Simon

From: obs-bounces at ecoinformatics.org [mailto:obs-bounces at ecoinformatics.org] On Behalf Of Matt Jones
Sent: Friday, 29 October 2010 12:47 AM
To: Cam Webb
Cc: obs at ecoinformatics.org; Chris Mungall
Subject: Re: [obs] Joining DwC, OBOE, PO and PATO


On Thu, Oct 28, 2010 at 2:46 AM, Cam Webb <cwebb at oeb.harvard.edu<mailto:cwebb at oeb.harvard.edu>> wrote:


Very useful.  And should DwC eventually accept an Observation class itself, the oboe:hasContext of the fruit observation could point straight to that. Or even to the dwc:Occurrence itself of we can consider an observation to be a type of Occurrence.

Fascinating.  In high-level conceptual terms, I have always considered an occurrence to be a form of observation, not the other way around.  This is because an occurrence records the presence or absence of an individual of a species at a particular place and time.  An observation is a record of any type of measurement within any context, and so might include, for example, the measurement of the size of an organism at a location and time.  This observation can then be used to infer that the organism was present at that place and time, and therefore that we can say it occurred there as well.

Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/obs/attachments/20101029/0b2d658c/attachment-0001.html>


More information about the obs mailing list