Above, I was envisaging face-to-face questions and answers, some of which may be explicit (as in the informal interview process I described and could easily imagine as realistic), while others are implicit, as in many mainstream job application processes (up to interview, where the questions become more explicit again).

I suppose the application form is one area where questions are explicit and could be presented and answered electronically. Generally, you know my interest in electronic questions and answers, but this post is really not aimed directly at that, but rather at the question of representation, which is relatively unconcerned with the medium of the questions. Of course, generally the face-to-face context allows much more richness and nuance in answering.

I like your idea of automatic interrogation, though in this context I haven’t thought that through. I think everyone has an interest in reusing questions and answers, and if both were formulated in as reusable a format as possible, people would save a lot of time in recruitment, as well as facilitating automatic interrogation. On the other hand, there genuinely are uncommon questions for most jobs — a kind of “long tail” if you like — and we have to keep those well in mind for practical purposes, and not pretend that every question can be answered in prearranged ways.

Thanks for your post and especially for framing the framework in terms of such key practical applications.

Speaking of which, and to clarify, are the Q&A sessions you describe between job-seekers and prospective employers something you imagine or intend to occur electronically? For example, would job vacancy definitions using the framework be the ‘instructions’ for employer or recruiter systems to automatically interrogate the portfolio or competency-based CV of an applicant?

I agree time is definitely significant in terms of claims to possess or requirements for the possession of competence. The issue to me is whether we need to explicitly have time represented in learning outcome or competence definitions themselves.

In terms of the question types I am positing, naturally time-related questions are vital in many applications. But again, maybe the time may be factored out neatly from the definition itself?

It’s one of these really interesting and vital questions – if we can create an effective solution while placing considerations of time somewhere else, convincingly, maybe that will be the better thing to do, as it will be simpler. If we find we need to represent time, now we need to do so will depend on the examples we can find. The fact that there are important time related questions to ask about individual competence does not of itself imply that there are important time-related factors about the definitions of that competence. As you indicated, variations in what counts as competence, or what learning outcomes are current, should be dealt with by updating the definitions themselves, keeping the old dated versions for reference.

The first can be managed by including some value of time associated with any claims of competence or assessment. I would suggest a fifth type of question concerned with timeframes (rather than shoe-horning into 2 or 4).
“During what period were you programming in Java?”, “When did you last play the piano in public? ” or “When were you awarded ECDL?”.

The second could come under the scope of InLOC, though I don’t think we have considered it much yet. The requirement is not so obvious, though is fairly simple to include for any property or relationship.

To some extent “time” is dealt with by versioning of LOCs or frameworks, creating new items and deprecating old ones, building relationships between items (e.g. X replaces Y).

Maybe a more comprehensive model should include optional time properties for all items and relationships not just the whole framework. A single date time, or start – end date time should suffice. Perhaps duration as well.

Mike