First, more importantly, if anyone has a comment on research regarding the effectiveness of informal learning, please attach it to Peter Isackson's comment below.
For me, the question is still, what is informal learning? For the sake of effectiveness and clarity, let's have those two conversations in parallel.
For a bit of context, and also to keep this from getting personal, I would also like to quote what I wrote in "Learning By Doing," about early, pre-hype stages of any new technology, or re-launches of "re-discovered" approaches (just to make sure we steer clear of these traps):
Theory:Wouldn’t It Be Great?
New ideas first tend to bubble up in academic papers, magazine columns, blogs, or conference panels. Those who introduce the emergent idea are very detailed in critique of the last generation, very vague in their descriptions of what the actual solution would look like, and very enthusiastic about the promises of what the technology will accomplish.
A lot of history analogies are brought in (we all get to learn about Dutch shipping patterns or the advent of the abacus), as well as out-of context quotes. There is often some graph of how large something is (dot-com companies loved showing how fast Internet access was growing as a justification for their own business model), no matter how indirectly related. The technology, if there is any, tends to be described in a pure environment, one without legacy systems.
As far as the theory goes, it sounds good. And almost all who hear this new theory nod and think, “This makes a lot of sense. This could be big.” But there are no examples of it working the way it is described. At best there are precursors that are “sort of” similar. Or a wild success in a different industry is held up as a model.
So, with that "dead elephant" out of the way, what is informal learning, anyway? I grant you that 95% (or some other, huge, made-up number) of learning happens outside of formal learning. The "largeness" of that number (like the dot-coms referring to the largeness of the Internet) does not a case make, as plenty of knowledge management vendors can attest.
And much more specifically, unless you work at Google or an IM vendor, what is your "value add" to it?