Most product development processes require "use cases" to be created very early in the process. The purpose of doing this is to try and define all the ways a customer or user will interact with the product. In other words, what will they use it for, and how do they want to go about doing that? From my experience developing new products, I've concluded that there are typically three valid categories of use cases: 1) The ones the development team and product management come up with based on their experience; 2) The additional ones actual or potential customers and users come up (using interviews focus groups, demonstrations, etc,); and 3) The ones nobody came up with, but the product ends up being used that way none-the-less. Kleenex and WD-40 are well-known, classic examples of the third category in which the unanticipated use cases led to vast, commercial success. More recent examples are SMS text messaging and Gorilla Glass.
I have a hunch some of the examples cited in the Technology Review article fall into the third category as well. In one example, Twitter was used to crowd source weapons knowledge, and an unusual type of landmine was identified in less than 40 minutes. In another example a rebel commander received critical information about how a certain type of rocket launcher works from two civilians a continent away, using Skype. This allowed his brigade to successfully attack the weapon.
I'm fairly certain Twitter never had a focus group to understand how their product might be used for weapons identification. I'm also reasonably sure that Skype doesn't have an "intelligence" division tucked between the cubicles focused on small business and grandma.
So what does this all this mean regarding the way we develop new products? I'm not ready to throw the baby out with the bath water. Today's product development methodologies work pretty well, and most companies are able to bring new product to market faster than ever. But product development professionals need to be opened-minded and acknowledge there are potential use cases that nobody can imagine... think of Donald Rumsfeld's "unknown unknowns." I once worked for somebody twenty years ago whose philosophy was to "just throw a bunch of stuff at the wall and see what sticks." Perhaps he has never been more right.