search

BiodiversityOntologies / bco

Biological Collections Ontology
Creative Commons Zero v1.0 Universal
22 stars 3 forks source link

Resolve nature of Observations #54

Open tucotuco opened 9 years ago

tucotuco commented 9 years ago

Use case: How do we model a photograph taken by a naturalist? Is that a result of a MachineObservation or a HumanObservation or both?

pbuttigieg commented 9 years ago

I suppose it comes down to what you choose to be the differentiae between the two. If the main differentia between human vs machine observation process is the agent executing (and perhaps guiding) the entire observation process, I would say its a human observation process which has machines (or other technology) as participants and machine-processes (e.g. the recording of the image by the camera) as parts. This would differentiate if from more autonomous machine agents.

tucotuco commented 9 years ago

My original thinking on this was to differentiate between processes whose outputs were created by a machine that might have characteristics of interest and those whose characteristics were embodied in the human.

I think these two statements distinguish between the two different views about what constitutes a machine observsation versus a human observation. If we can agree that this is the distinction, we can decide (perhaps) which is the one that is more relavent for the questions we might want to ask in our domain.

Stimulus that results in the observation is detected by a machine.

The information content as output of the observation is created by a machine.

On Fri, Mar 20, 2015 at 1:58 PM, pbuttigieg notifications@github.com wrote:

I suppose it comes down to what you choose to be the differentiae between the two. If the main differentia between human vs machine observation process is the agent executing (and perhaps guiding) the entire observation process, I would say its a human observation process which has machines (or other technology) as participants and machine-processes (e.g. the recording of the image by the camera) as parts. This would differentiate if from more autonomous machine agents.

— Reply to this email directly or view it on GitHub https://github.com/tucotuco/bco/issues/54#issuecomment-84069646.

ansell commented 8 years ago

Personally, I am assigning dwc:basisOfRecord => HumanObservation to images taken by a photographer who is physically present and MachineObservation to those where the photographer has setup an autonomous camera.

The reasoning for me is that I don't want to create confusing distinctions between photography and other cases such as microscopy or manual temperature measurements where a human uses a tool directly for an observation where we naturally refer to them as human observations, not machine observations.

ramonawalls commented 8 years ago

Maybe the solution is to have several classes that divide observations among different axes: -What makes the decision to record the observation (human or machine) -What records the observation (human or machine) -What form the output data takes (digital or analog) (not sure we need classes for this one)

ramonawalls commented 6 years ago

Part of the BoR discussion for issue #89.