A number of years ago, the company I worked for had the admirable idea of introducing a knowledge base, very similar to the one managed by Microsoft Support.
The idea was for the organisation to house its information in the one repository, thereby fulfilling the role of the “single source of truth”.
All staff – especially those with direct customer contact – could then mine the repository on a JIT basis to retrieve the information they needed when they needed it.
As an L&D professional, a light bulb switched on in my head…
I wonder if we can use this tool to support learning?
In particular, I was keen to add a navigational aid to the database: a simple tree structure that outlined its contents. By following the tree, an employee could explore the repository’s assets while gaining a sense of the relationships among them.
Unfortunately my wish was not granted. As a strict performance support tool, the knowledge base relied solely on a search box to facilitate immediate targeted access.
Fast forward to today and a former colleague of mine who still works in the company tells me that the tool is widely regarded as difficult to use. Despite its intention to be the single source of truth, employees tend to exhaust other possible sources of information before consulting it as a last resort.
The pitfall of search
I was reminded of this disappointing affair as I watched Dr Mitchell Whitelaw, Associate Professor in the Faculty of Arts and Design at the University of Canberra, present his excellent talk at TEDxCanberra.
Essentially, Dr Whitelaw advocates the “show everything” model of information discovery, instead of the presumptuous search model.
I love his use of Sidney Nolan’s classic painting Ned Kelly to illustrate his point…
Think of the world that surrounds Ned – the ground, the clouds, the sky, everything – as the contents of a repository, whether it be a database, a corporate intranet, or the World Wide Web.
A search box presumes you know what you’re looking for. So if you’re looking for the tax rate applicable to superannuation contributions, for example, you might key in something like “contribution tax rate superannuation” then click “Search”.
That’s great for information retrieval. Google has made millions out of it.
However, it’s not very good for learning. Why? Because you don’t know what you don’t know, especially when you’re a novice in the domain. Referring back to the example above, you might not know what superannuation is, or what contributions are, or the fact they are taxed.
Take a closer look at Sidney Nolan’s painting. If you rely on searching only the key terms that you are aware of, it’s like focusing on what you can see through Ned’s helmet.
A world of information exists outside your field of vision, but because you don’t know it’s there, you’re unlikely to find it.
Dr Whitelaw’s proposal to remedy this problem is to show all the contents of the repository in one hit.
Yes it sounds crazy, but before pre-judging, check out his commonsExplorer app which lets you browse the public photography collections of institutions around the world. It’s impressive!
If we stick with the painting metaphor then, the “show everything” visualisation replaces Sidney Nolan’s Ned Kelly with Susan Sherrah’s Kaleidoscope – Land and Sea…
Can you believe I saw this painting in a tourist brochure at the hotel after returning from TedxCanberra? Talk about serendipity!
From the sublime to the ridiculous
While the “show everything” approach to knowledge management revolutionises browsing, I’m not sure how effective it is for deep learning.
In fact I find it a little bit overwhelming. From not enough information under the search model, to too much information under the “show everything” model, the learner is lost between two extremes.
We need balance, and the cognitivist in me believes we can achieve it via scaffolding. Practically speaking, this means categories and tags.
To illustrate, let’s return to my knowledge base example…
IMHO, an L&D specialist should have consulted the various SMEs across the business to identify the super categories that represent the sum of the knowledge contained therein. These would manifest as icons on the homepage.
Upon clicking an icon, the learner is presented with a listing of the assets tagged with the corresponding hashtag. In addition, sub-categories can be explored further to refine the listing, and so on and so on.
This may remind you of the Yahoo! Directory. This is indeed the concept, but in a workplace environment it is managed much more tightly for the purposes of the staff in the organisation.
The road well travelled
It’s old fashioned these days, but I still believe in the value of an expert guiding a novice.
I maintain the expert has an obligation to contribute their expertise to the knowledge repository, and to tag it appropriately. The L&D specialist ensures this happens efficiently through support and governance.
The repository becomes an open environment for the learner to explore at their discretion, yet is structured enough to guide their learning and form a mental model of the domain.
Learning theory geeks like me call this combining constructivism with cognitivism, and I think it’s really powerful.
The best of all worlds
Does my proposal preclude a search box? No.
Does it preclude a “show everything” visualisation? No.
In fact, I suggest including both on the homepage along with the category icons.
If the learner needs to find something immediately, they can search for it; if they want to browse the content, they can play with the “show everything” visualisation; if they want to learn deeply, they can dive into the categories and sub-categories.
It’s all the same knowledge, but with smarter design, it serves everyone.