After the establishment of object oriented design in the software realm, the concept of the learning object was inevitable. The idea of assembling learning objects to produce educational programmes in the same way that computer programmes are assembled from software objects was irresistable.
However, there are three pathological flaws in the concept of the learning object that pre-ordained its ultimate failure.
The first is that, unlike software objects, which can call methods on other objects in a standard manner, there is no corresponding architectural assembly mechanism with which to build complex learning objects from simpler ones.
The second arises from the fact that a learning object is static at the highest definitional level (although it might contain inner processes), making it impossible to use to model processes directly. Teaching and learning are processes and have to be modelled using processes.
The third is that the use to which an object can be put is not an intrinsic property of the object itself. It depends on context.
The concept of paradata has been introduced to solve this third problem. If one takes the view that this can save the patient then the flaw might be non-pathological. However, if the cure is to introduce a whole new descriptive data layer, then the patient might live, but with what quality of life? (This grants that the patient might be able to die with the other two other conditions rather than die of them.)
Kungkhies are based on intended learning outcomes and activities. In this formulation, paradata becomes redundant. The functionality of paradata is assumed by the intended learning outcomes of each activity.
Of course, in formal education intended learning outcomes already occupy a central position. Despite this, there has been little theoretical development of the concept. The theory behind the kungkhie platform offers several developments in this area: a canonical formulation of intended learning outcomes, a propositional logic of intended learning outcomes, and the idea of pedagogically complete sets of intended learning outcomes. These developments form the basis on which machine learning techniques can be used to analyse and organise intended learning outcomes.
Ah, you say, an intended learning outcome is heavily dependent on context and thus suffers from the same pathological flaw as the learning object.
Ah, I say, the intended learning outcome is presented within its own process-centred standard wrapper context - the kungkhie.
The logic and canonicalisation of intended learning outcomes will be presented later.
