Core review

We have recently completed a review of all the proposed OBI Core terms. For the first part of the review process we used this wiki page to coordinate. For the final stages of the review process we used several Google Docs.

Documents
A number of documents were used in to coordinate review during the Philly2013 workshop:


 * Core review spreadsheet
 * OBI Core: Data
 * OBI Core: Inferences
 * OBI Core: Specimens
 * OBI Core: Plans
 * Suggested review criteria

We have also developed a detailed prototype for modelling data for measurements, predictions, and settings. See the latest versions of the files in our Subversion repository: PDF file, OWL file, source text file.

Completed Terms
Review of all terms has been completed. For terms reviewed using this wiki, the change notes are here. For all other terms the change notes can be found in the appropriate Google Doc. The "Core review spreadsheet" contains the master list of reviewed terms.


 * acquisition
 * note: material acquisition and information acquisition will be added to OBI Inner Core, since they are expected to be more widely used than their parent.
 * note: We have discussed the potential overlap with "service" terms in OBI. There does not seem to be a problem as terms currently stand, but we can revisit the question in the future as modelling of services matures.
 * major: add "SubClass Of: has_specified_output some ('information content entity' or 'material entity')"
 * minor: remove alternative label "material acquisition"
 * minor: change textual definition to refer to "planned process" rather than "process"
 * assay
 * major: update textual definition to further distinguish an assay from a prediction: "A planned process with the objective to produce information about the material entity that is the evaluant, by physically examining it or its proxies."
 * controlled variable specification
 * added to inner core
 * major: change "SubClass Of: 'is part of' some 'plan specification'" to "SubClass Of: 'is part of' some 'study design'"
 * major: added example of usage, shared bewteen controlled variable specification, dependent variable specification, and independent variable specification.
 * minor: update curation status to "ready for release"
 * data transformation
 * major: make the logical definition more specific than a generic planned process by adding an axiom referring to "algorithm" (a child of "plan specification"): "SubClass of realizes some (concretizes some ('algorithm' and ('has part' some 'objective specification')))"
 * minor: update curation status to "ready for release"
 * dependent variable specification
 * major: change "SubClass Of: 'is part of' some 'plan specification'" to "SubClass Of: 'is part of' some 'study design'"
 * major: added example of usage, shared bewteen controlled variable specification, dependent variable specification, and independent variable specification.
 * minor: update curation status to "ready for release"
 * device
 * note: Over the past year we have tried to apply several differentia to distinguish devices from reagents, without much success: a macroscopic/microscopic distinction; requiring that reagents undergo chemical reactions; requiring that devices are not consumed. On the 2012-12-17 developer's call we decided to take a different approach and define "device" as disjoint from "reagent".
 * major: modify textual definition "A material entity that is designed to perform a function in a scientific investigation, but is not a reagent."
 * major: add "Disjoint With: 'material entity' and ('has role' some 'reagent role')" (this is the logical definition of "reagent" in REO)
 * minor: add example of usage for an autoclave
 * minor: add editor note explaining the current approach
 * minor: remove some of the old editor notes
 * documenting
 * major: Revised definition to improve the grammar: "a planned process in which input information is used to create or add to a document"
 * minor: Update curation status.
 * minor: add example: "Copying the readout from an instrument into a spreadsheet."
 * note: no objections to definition change from IAO (email thread)
 * information acquisition
 * major: add "Disjoint With: 'data transformation'"
 * independent variable specification
 * major: change "SubClass Of: 'is part of' some 'plan specification'" to "SubClass Of: 'is part of' some 'study design'"
 * major: added example of usage, shared bewteen controlled variable specification, dependent variable specification, and independent variable specification.
 * minor: update curation status to "ready for release"
 * interpreting data (renamed to "drawing a conclusion based on data")
 * major: Add class obi:drawing a conclusion (subClassOf obi:planned process, subClassOf has_specified_output some iao:conclusion_textual_entity)
 * major: Add label 'drawing a conclusion based on data' as primary label to 'interpreting data'
 * major: Add axiom 'drawing a conclusion based on data' subClassOf has_specified_input some iao:data_item. As a result this class should be inferred as subclass of the new class 'drawing a conclusion'
 * investigation
 * major: Add axiom obi:investigation subClassOf has_part some obi:drawing a conclusion
 * material acquisition
 * major: add "Disjoint With: 'material processing'"
 * minor: change example from "investigators animal facility" to "bred in the animal facility of the institute as a service to investigators".
 * planning
 * minor: Add example of usage that has a robot scientist making the plan
 * minor: Update the metadata to make clear what "Source PPPB Branch" means.
 * study design execution
 * major: remove "has_specified_output some 'data item'" from equivalent classes. (Keep minimal restrictions that need for defined classes)
 * study subject role
 * major: rename to "participant under investigation role"
 * major: update textual definition: "A role that is realized through the execution of a study design in which the bearer of the role participates and the results of the study design execution are about that bearer"
 * minor: Add an editor note to say that this role is distinct from "specimen role", but it is possible for a participant to realize both "specimen role" and "participant under investigation role" at the same time.
 * minor: Revise examples for style and add another: A lake could realize this role in an investigation that assays pollution levels.
 * note: it was decided on the 2012-11-26 call that we should not move the children of this term at this time, but a tracker item has been added flagging the children for later review.

Discussion Notes
For terms reviewed using this wiki, the discussion notes are here. For all other terms the discussion notes can be found in the appropriate Google Doc. The "Core review spreadsheet" contains the master list of reviewed terms.


 * study design execution
 * note: Alan asks whether there should be the axiom "has_specified_output some 'data item'" at all. Bjoern insists that there should always be a data item produced. Alan asks about studies that are inconclusive. There is always data collected. The question is whether it is the "specified output" of the study design execution.
 * investigation
 * major: Add an objective to produce a conclusion. This has not been implemented. On the developer's call 2012-11-16 the consensus was that the objective is implied by "has_part drawing a conclusion". We may decide to specify an objective at a later date.
 * major: We discussed making the axiom "has_part some documenting" more specific.
 * Suggestion 1: follow the model of "(planning and (has_specified_output some 'study design'))" Update 2012-09-25 This suggestion was dismissed on the developer call 2012/09/17.
 * Suggestion 2: We can say things like: investigation that has specified output conclusion equivalentTo: investigation that has part document that has part conclusion textual entity. Update 2012-09-25 The current logical definitions of obi:investigation (subClassOf: has_part some iao:documenting) and iao:documenting (subClassOf has_specified_output some iao:document) already entail that one or more iao:document(s) exist as specified outputs of the iao:documenting processes that are part of an obi:investigation. With regard to the development notes for the general model of an investigation these iao:document(s) might be iao:report(s) (iao:report subClassOf iao:document) but they don't have to be and this is an appropriate view at this level of generality as documenting processes are processes that as a minimal requirement bring results of observations into a tangible, lasting form while reports as outputs of reporting processes put related information content into a much more comprehensive form (to inform a recipient of certain aspects of an investigation). The former can safely be assumed to be part of any investigation while this may be not so clear for the latter. In conclusion, what is suggested here is already in place.
 * Suggestion 3: has_part some 'interpreting data' (or 'drawing conclusion') Update 2012-09-25 As a result of our discussion of obi:interpreting_data (see below) we agreed that the axiom obi:investigation subClassOf has_part some obi:drawing a conclusion should be added. James pointed out that based on the current definition an obi:investigation always generates some iao:data_item as specified outputs of the obi:study design execution that is part of the obi:investigation and therefore, as data items are understood to pertain to empirical data, the new axiom could be narrowed down to obi:investigation subClassOf has_part some obi:drawing a conclusion based on data.
 * Action items:
 * add axiom obi:investigation subClassOf has_part some obi:drawing a conclusion -DONE-
 * interpreting data
 * major: Rename to 'drawing conclusion' or add 'drawing conclusion' with subclass 'drawing conclusion based on data' and make 'interpreting data' as a subclass of 'drawing conclusion based on data' Note: Based on current logical definition, label 'drawing conclusion' is more suitable and it is not limited the inputs as data item. However, 'drawing conclusion' cannot cover 'or to identify what additional data is necessary to draw conclusions' which was addressed in the textual definition. Update 2012-09-25 The motivation for reviewing this class is that the current label implies that data items (taken as empirical) are inputs to any such process while the logical definition entails a conclusion as outcome of any such process (which is not in the label). An additional concern is that there may be reasoning processes which do not involve empirical data (e.g. mathematical proofs) As a result of our discussion we concluded that for any process of "interpreting data" the result of that interpretation (whatever it is) amounts to one or more conclusions based on the data being interpreted. Vice versa, any process of drawing a conclusion based on some set of data may be regarded as a process of interpreting that set of data, which makes these classes basically equivalent, though the logical definition would need to be changed. Deprecation or relabelling 'interpreting data' was discussed, currently (2012-09-25) opting for relabelling
 * Action items:
 * add class obi:drawing a conclusion (subClassOf obi:planned process, subClassOf has_specified_output some iao:conclusion_textual_entity)
 * add label 'drawing a conclusion based on data' as primary label to 'interpreting data'
 * add axiom 'drawing a conclusion based on data' subClassOf has_specified_input some iao:data_item Note: As a result this class should be inferred as subclass of the new class 'drawing a conclusion'
 * Modelling "possession" or "contol" over resources:
 * note: It would be useful to be able to model "possession" and "control", which are important parts of the textual definition of "acquisition". But in the 2013-01-07 call we decided that this is not an urgent need, and have left it for later.
 * acquisition
 * note: We want to be able to model "possession" or "control" of material entities (and probably information content entities), which will probably require new roles. These roles were proposed in the context of "material acquisition" but we should consider generalizing them to "resources" that could be material or information:
 * major: add 'material possessor role' = borne by person or organization, realized in choices about what to do with material (this would be for other people to refine to determine kinds of ownership)
 * major: add 'material acquisition source role' = a role borne by a person or organization that is realized in an material acquisition process in which a material possessor role is transferred from the material acquisition source to the material acquisition destination (it gets difficult to write this nicely with multiple role and bearer relationships...)