2009-03/Leap2A structures

Belongs to the 2009-03/Leap2A specification – see also 2010 developments

This page gives an introductory overview to portfolio information and Atom. For more detail, please see the elements page.

The structure of portfolio information
There are three fundamental types of information that can be included in portfolios: In a portfolio, any selection of these can appear together, as a set of single, self-contained portfolio items. The items themselves are the minimal units of information that make sense in their own right, and could be reusable separately from other portfolio items. Information managed by an e-portfolio system is just that which the portfolio holder wants to keep or maintain, perhaps to reflect on, and potentially to present to others. Information collected by other people for their purposes could be represented in the same format, though one would not call it portfolio information.
 * digital artefacts made or jointly made by the portfolio holder
 * reasonably short (not very long) expressions or entries, normally as text, by the portfolio holder (if longer, they might rather be considered as artefacts)
 * information about the portfolio holder (that is also valued by the holder), their abilities, achievements, experiences, activities, goals, plans and such like (the portfolio holder may not value a record of debts, for example).

Digital artefacts, including audio, video, multimedia as well as plain word processed files, may have some associated information, including metadata such as author or owner, date of creation, modification, etc., title, perhaps summary. They exist as self-contained entities, and a portfolio holder very commonly presents them by way of evidence for their abilities. The item here is the combination of the artefact itself and any metadata represented within the portfolio system.

Short expressions may have their own self-contained meaning, as for example in diary or blog entries. Alternatively they may be tied to something else. For example, a PDP process may include a learner reflecting on the strengths, weaknesses, opportunities and threats connected with one aspect of their situation. The user interface could present these as separate text boxes, so that they can be held and managed separately. Each one could have its own creation and modification dates. In this case, though the piece of text is only fully understood in its context, it nevertheless exists in its own right, as it has its own creation and modification dates, and could even have a different author. Each expression is a separate portfolio item.

Information about particular things is again of two kinds. Firstly, it can be information directly related to the thing. An obvious example is the dates associated with things that have happened, or are planned to happen. These dates have no meaning at all separated from the portfolio item they refer to. Secondly, much other information about particular things is given by the relationships between that thing and other items. A portfolio item in this case is just the information directly related to the thing, that has no separate meaning on its own. For portability of information and interoperability of systems, it is vital to be able to identify, in a common way, the type of thing which a portfolio item holds information about. Nearly all e-PDP and e-portfolio systems make distinctions of this nature.

Given the approach of regarding portfolio items as very small sets of tightly related information, representing the relationships between portfolio items is also centrally important to capturing the meaning of portfolio information. The user interface of a portfolio related tool may invite the learner to enter text under many different rubrics, and the significance of the text entered will depend on how the context in which the learner was invited to enter the text.

Selections of portfolio items are also an important aspect of portfolio practice. Portfolio information is generally thought of as collected, selected and presented, and many portfolio systems allow other people to see such selections.

Correspondence with Atom structures
The principal component structure in Atom is the atom:entry element. The main correspondence is therefore between Atom entries and portfolio items, as described above: each item corresponds to just one entry. Blog information is essentially of one kind only: authors write blog entries, which can be about anything. Thus there is no provision within Atom to distinguish types of entry, and therefore "foreign" elements in the entries are used to represent portfolio item types. Atom does, however provide the atom:category element, which is used in this specification to represent some finer distinctions of category.

The granularity of the information is highly significant. As each portfolio item is a relatively small set of information that only makes sense together, there will be a relatively large number of relationships between the portfolio items (or Atom entries). Atom provides the atom:link element to represent entry relationships, but the Atom vocabulary for links is much smaller than what is needed to represent the kinds of relationship between portfolio items. So this specification extends Atom by providing extra values for the "rel" attribute of link, corresponding to the relationship types that have been agreed practically useful in existing portfolio systems.

Atom does not have any structure particularly adapted for selections. An entry is used, with links to the items included in the selection.