When your grandmother – or any other adult demonstrating their infinitely superior wisdom for a moment – told you with an air of conspiring, “Doesn’t ask doesn’t get”, they had a good point. Apart from making a positive change from “Mustn’t grumble”, four well-chosen words communicated more than many a longer screed. Or, more accurately, a nebulous, windy cloud of a question.

There’s a fascinating post at Mark Gould’s Enlightened Tradition blog, Asking better questions, getting better insight, that ponders knowledge as something subject to push and pull. We’ve got quite good at push, albeit in an unfocused sort of way. If we live in an attention economy, it’s least partly because the need to pay attention and to pick your way through tidal waves of ‘information’ is becoming a modern survival technique. And export knowledge abounds, fizzing between the ears of the knowledge workers around us and the whirring on the hard-drives and the cloud stores of our latest gizmos.

But somehow this abundance of know-how manages to co-exist with equally cloudy stores of ignorance. As Mark Gould puts it:

Frequently, however, I see people asking quite open-ended questions in the hope that something useful will pop up. I suspect that what actually happens is that those with the knowledge to assist don’t answer precisely because the question is too vague.”

Years ago, I sat in a meeting of Aslib, the national body for specialist librarians and information bureaux. We were gathered at the cusp of the IT revolution: mobile phones were making their clunking first appearance, and ‘portable computers’ were triggering lumbago in middle-aged men in selected pockets of the South-East. One of the verdicts of the assembled gathering was that, in the future, the advantage would lie with the people who knew where to find the information rather than the people who knew it. Faced with the information equivalent of gold doubloons buried on an immense sandy beach, the coin hunters with the metal detectors would be an odds-on bet compared with the pebble turners and the sieve-sifters.

But is that how our questioning skills have actually developed? Have we all become expert data miners, casually parsing complex Algebraic formulations to retrieve exactly the right article on deciphering the papyrus scrolls of the Etruscans? Have we buffalo. Our discrimination between the options for the look and feel of the handsets that we merrily thumb away at may be acute, but the same can’t always be said of what we’re skimming through on them. As Mark comments:

But what about those areas where great thought is required. How do we build people’s capability to get to the insight and expertise that will help them solve the trickier problems that clients bring?
We can throw technology at the problem again — search engines will allow people to draw on the vast pool of work that has already been done. Sometimes that will disclose a really useful document that contains just the right information to help the lawyer arrive at a suitable answer. More often, though, it will produce nothing at all or many documents none of which actually help directly.”

Gould is writing in the context of Knowledge Management in a large legal practice (an environment heavy on the ‘push’ of policies, procedures, practices), but he’s not the only interesting recent commentator on the impact of different tools aimed at supporting discovery. In his book, Together: The Rituals, Pleasures and Politics of Co-operation (a follow-on from his earlier The Craftsman, which we much admired), Richard Sennett observes the unexpected side-effects of Google Wave.

Implemented to support a discussion group, they noticed that ‘side topics’ – the little branching enquiries that occur in debate – are sidelined by the software, which tries to steer them, satnav like, to the mainstream of the conversation. We can use the machines to converse, but do we have the same conversation we’d be having without them? Sennett notes that Jaron Lanier (author of You Are Not A Gadget, also reviewed here earlier) is also voicing some ‘side topic’ concerns on this issue:

Lanier’s caution is that in ordinary usage the technology is more likely to bend human will than to bend in response to it; put another way, you have to struggle with or deform an engineering social programme to practise complex social exchange.”

At 182 characters, Lanier’s insight would require him to reword it if he wished, for example, to share his insights via Twitter or SMS. I’m not usually in the habit of quoting Ayn Rand, but she was alert to the danger of hastily embracing the soundbite version:

The uncontested absurdities of today are the accepted slogans of tomorrow.”

The dialogue that we have with our machines is not immune from favouring the obvious over the complex. Technological breakthroughs that promise unparalleled access to unknowable quantities of knowledge have instead largely driven us to skim read the first half-dozen and give up. Google provides an infinite ocean of information, yet most of us look no further than the first 20 results. Hardly anyone uses advanced search, a problem Stephen Turbek explores at Boxes and Arrows. Everyone knows there’s too much information, and that a better version of the fairground grabber is needed to unearth the nuggets, but we’ve not given great attention to improving the offering:

Despite its name, advanced search has not advanced very far. There is great power to conquer the overwhelming number of search results, but the current standard presents barriers to users. Specifically,

  • The link is often small, vague, and does not describe benefits to the user
  • Advanced search pages typically have confusing page design for the few who make it there.
  • There is generally poor search revision functionality: Once a search is performed, the “advancedness” is lost. For example, the Google advanced search delivers the standard search results page. You have to get the query right the first time; there is no opportunity to adjust your query.”

We’ve spared ourselves the long trek to the library, the training in using Boolean algebra and the trundle home with the weighty tome that we have access to only for a strictly time-delimited period, so you’d think we’d have time to click ‘Advanced’ and type into three boxes rather than one, but no. Most of us used to be lazy enough to get librarians to find stuff we could have followed signs to; we got rid of librarians because computers were cheaper, not because we didn’t (mis-)use the librarians.

It’s like learning the Swedish for “Can you direct me to the nearest chiropodist?” but not knowing enough Swedish to understand if anyone answers. It’s not a failure to ask questions that’s the problem; it’s the failure to know how to formulate questions – and the failure to explore questions with those that ask them. (That being slow, old-fashioned, and requiring the interplay of something truly ‘thinking’.)

Question-seeking and answer-providing – teasing out what someone really wants to know so you can give them exactly the answer they’re looking for – are collaborative activities. It only takes one to Google, but it takes two to tango. We’ve discussed previously that feedback is a conversation, not a monologue: the person providing the feedback shares responsibility with the recipient to ensure that the feedback is either actually understood or modified where it’s been based on incomplete perceptions.

Yet the two-way element is one that so often needs more careful attention than it is given. The interrogation-style interview is a staple of broadcasting and media practice, but it’s rarely well done. Even in the best Today programme scenarios, there isn’t often a willingness to ‘go off script’ and explore an interesting comment when it arises. Without which the listener can come away with the sense that the questionee enjoyed putting their feet up for a few minutes, all the better to allow the interviewer to blow smoke into what we should probably leave best described as “a great orifice for radio”.

So how do we countermand the rush to the quick, simple answer? How do we encourage the asking of detailed questions, and the willingness to explore other people’s questions so we can enrich them better with our answers? Are there elements here also of reward and of culture? Is it as simple as the fact that we don’t pay people to help each other, so they don’t? Would assessing our ability to facilitate and grow and develop each other give us a more meaningful way of rewarding contribution?

If so, what’s the risk in doing so that earns this currently absent reward? I guess it depends on whether we see knowledge as the golden egg, or as the goose that lays it. If we view it as the egg, each of us will be encouraged more to sit on our assets and guard our bullion: golden egg as bargaining chip. But if we view knowledge, wisdom and insight as the goose, the productive, egg-laying goose is worth more than her grumpy, defensive egg-bound siblings. And the organisation gets more eggs. If knowledge sharing in your organisation is currently something of a pantomime, might it be worth swapping the characters?

Add to: Facebook | Digg | Del.icio.us | Stumbleupon | Reddit | Blinklist | Twitter | Technorati | Furl | Newsvine

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Furl