This section focus on the framework learning context, presenting the proposed supporting process, its concepts, phases and tool requirements.

For the sake of clarity, the meaning of learner, artifact and knowledge-base is explained to prevent misinterpretations when reading this page.

• Learner. Any framework user, developer or evolver that needs to acquire knowledge about the framework.
• Artifact. For the context at hand, this means any documentation artifact available to the learner and present in the proposed collaborative environment. Is is assumed that there is a framework documentation artifacts repository (FDAR) present for consultation by the learner.
• Knowledge-base. This regards the storing facility of the collaboratively generated learning knowledge. Despite referencing the FDAR, it is a different data-source.

The idea behind the proposed collaborative approach evolved from what is commonly known as “pave the cowpath”.

The expression has its origin in a poem written by american Sam Walter Foss (1858- 1911) called “The Calf-Path”[Fos]. The poem tells the story of a strained calf, lost in the woods, who, when returning home, “made a trail all bent askew, a crooked trail, as all calves do”. That trail kept being followed by beasts and humans until today, currently being the main streets of a metropolis. The moral is that “[..] men are prone to go it blind, along the calf-paths of the mind, and work away from sun to sun, to do what other men have done.” This poem serves as criticism to the lazy, narrow-minded men that mindlessly follow pre-defined paths without questioning its effectiveness or usefulness. Another similar allegory is the Cage of Monkeys¹⁾. This poem is popularly attributed to the streets of Boston, given their peculiar layout.

Of course, this connotation has issues. Its common knowledge that cattle are actually pretty good at finding the path of least resistance, which is, often the best route for a road. But let’s transpose this concept to the context at hand.

Cows walk with their heads down, are beasts of habit and usually move in herds. But a solitary cow is, usually, smarter than the bunch as it can’t rely on the group to reach her goal, whether reaching a pasture or returning home. In fact, they have a good sense of direction and can quickly retrace their steps back to the herd or to the point of origin. The herd factor is simply a matter of blind trust. Smarter cows keep their heads up and introduce independence to the herd, making it more wise, and question the effectiveness of the trail they take.

Transposing to the collaborative framework learning context, all cows (framework learners) are smart. Therefore, the cowpath becomes the steps the learners took to reach a solution. The problem is that there is no stepping on the grass, that is, those steps aren’t being recorded. Most probably, the next learner that undertakes the same steps will not be aware of a pathway forming. By paving that pathway, it becomes easier for future learners to quickly reach the same solution. This pathway is called learning path.

Providing framework learners (smart cows) with learning paths (paved cowpaths), improves their learning experience by focusing of the relevant knowledge (steps) other learners (collective herd) already have. This allows for a quicker, more effective knowledge transmission, in the sense that it provides the learner with directions on which artifacts to look at and in what order.

If a road is not used and maintained, its pavement breaches, erodes and decays, making it harder to use. It might be because there is a better road than this one. This is also true with learning paths. The quality of the learning paths is maintained by the community of learners. The most useful and effective learning paths are prone to evaluation and rated accordingly. This rating indexes the learning paths, so that the most used and approved by the community of learners are presented first. It follows the If it is useful, it is used, if not, it is abandoned rule.

Putting it simply, providing a learner with the steps others (learners) took to solve their problems, can improve the learning experience and produce better and quicker outcomes. The motto is: Show me how you learnt it. This section details the four-step learning knowledge cycle (Figure 1) the proposed collaborative approach defined as a means to support the previously stated. The goal is to non-intrusively capture the learning steps a framework user takes, store it in a shareable knowledge-base, where other users can access it. This knowledge relies on the community’s potential to maintain its relevance and quality, by rating it and allowing the system to recommend possible next steps that aid on the learning task. The four steps are detailed next.

This is the first step of the learning cycle. Here the learner begins her learning quest to find knowledge that might solve her problem. The trail of steps is captured as she browses through the artifacts, trying to find the relevant knowledge that might help her. This step ends when she is satisfied with her findings.

On the second step, the learner looks at her captured learning path and clears the weeds, that is, improves it. This is done by trimming off those steps that, despite taken, didn’t lead to the required knowledge. Seldom a novice learner takes a straight route to the knowledge she needs, unless in cases where she is already strongly familiar with the artifacts and needs little or no assistance to reach her answer (she would, by then, be considered an expert). This step allows for the improvement of the captured learning path, as to prevent other learners from running in circles or hitting dead-ends. Afterwards, the pruned and grafted learning path is stored in a knowledge-base.

The third step regards the sharing and rating of the learning paths stored in the knowledge- base. The learners access the knowledge-base, searching for learning paths that might help them. They evaluate its usefulness (taking the steps, a.k.a., walking through or just inspecting the visited artifacts) and rate them according to its effectiveness. There are no standard quality metrics here, the learner simply gives her opinion on how satisfying a specific learning path was for the current context.

This step enables the recommendation of possible next steps (on a learning path that is being currently captured), based on previous learning paths other learners have took. As such, this step occurs during the first one (Capture). Of course, this recommendation has an heuristic that relies on the amount of learning paths already captured. The more learning paths get captured (and rated), the better results the recommendation step provides. Usually, this is intrinsically sensed by the community, so this step is a motivating feature that spurs the participation of the community.

Knowledge is useless if you can’t get to it. In order to be able to access the information, we have to give it meaning. As with the notion of Definition [Lon10], humans need some form of classifying information so that they can, semantically, store it and access it easily. The web communities have, consciously or not, developed a light form of providing this categorisation, through what is called tagging.

Tagging is the labelling of an entity (usually a web page or something with a URI²⁾) with words or phrases so one can remember them later and group them with related finds. This was a shift from a common, rigid and hierarchical form of categorisation (e.g. folders) into a more flexible, grouping and descriptive-like form. It provides more means to obtain information as it enriches the identification of objects. On a folder-based categorisation, one needs to remember the exact name of the folder, whether, in a tag-based categorisation, one only needs to remember an aspect of the object in question (that hopefully was tagged that way). Of course neither approach is perfect, therefore they can be combined, complementing each other³⁾.

This tagging phenomenon has quickly spread across the web, and led to the notion of folksonomy.

The term Folksonomy is credited to Thomas Vander Wal, for combining the words “folk” and “taxonomy” to create this neologism from what he calls “bottom-up social classification” [Wal]. Shirky defines it as “socially created, typically flat name-spaces”. An essential feature of these terms is their public nature, that allows users to instantly determine how others have used the same terms in categorizing their own content, and view terms others have added. This cycle of use and observation enables the community to shape the folksonomy, encouraging useful applications and eliminating useless ones [Shi]. As such, a good definition for folksonomy is given by Sturtz [D.N04]:

In practical terms, a folksonomy is the complete set of tags - one or two keywords - that users of a shared content management system apply to individual pieces of content in order to group or classify those pieces for retrieval. Users are able to instantly add terms to the folksonomy as they become necessary for a single unit of content. The value in this social tagging is derived from people using their own vocabulary and adding explicit meaning, which may come from inferred understanding of the informa- tion/object. People are not so much categorising, as providing a means to connect items (placing hooks) to provide their meaning in their own understanding.

Popular examples of systems that use such categorisation are Flickr⁴⁾, where users can tag their digital photographs while uploading them to the system and YouTube⁵⁾, where the same is allowed for uploaded videos.

Consequently, the proposed learning knowledge cycle uses these notions to categorise its elements, that is, its learning paths. Tags are used by the learners to both store and search for the information they need. Therefore, there is no forced categorisation or taxonomy imposed to the community. It is the community itself that shapes the way it wants the information to be categorised.

< back to start page