OPINION: It's just as important to have the skills to know how to interpret and make use of the ever-proliferating range of data.

My brother and sister-in-law have two children, plus a dog and a cat. They can just about manage the children, but the animals are something else. Recently, at midnight, the cat dragged itself up the stairs to the adults' bedroom, and promptly started convulsing. There began a frantic visit to the emergency vet, which saved the cat's life but just about destroyed the family's bank balance. It turned out that my sister-in-law, in a rush as mothers tend to be, had used dog anti-flea powder on the cat, thinking that one would do for the other. My brother was indignant. Why, he demanded, had she not read the instructions? These, when reviewed, clearly stated that the product could be toxic to cats.

But who reads instructions? The legislative instructions we call public policy are way too complicated for most ordinary mortals. The Tax Act defies all comprehensive knowledge. And I am told that the Pay Manual for the Armed Forces makes the Tax Act seem straightforward. In these circumstances, it is hardly surprising that most of us resort to work-arounds. Or we ask someone in the office who has been there longer than we have.

Even when the instructions are clear, following them may not be straightforward. The Brisbane floods show just how convoluted these situations can become. The Wivenhoe Dam operators were supposed to follow the water-release strategies in the dam operating manual, but were apparently in the habit of seeing the strategies as labels after the event, rather than as decision-points to be observed during the management of the event. The engineers responded to the information available to them about the way the dam was behaving, in ways that accorded with their understanding of the situation they found themselves in.

Problems arose, of course, when the inquiry came along, and they needed to show that they had operated the dam in accordance with the instructions in the manual. But doing it ''by the book'' can get you into trouble rather than absolving you of it, as the military found out in the aftermath of the Skype incident. Calling off (separate) disciplinary charges against the young female cadet may not have been in the manual, but would arguably have circumvented her decision to go the media.

All this goes to show how shifting and ambiguous is the relationship between the world ''out there'' and the ways in which we try to respond to it collectively through policy in its various forms. It is impossible to imagine public policy without information. It's just that there is no such thing as ''information'' in an objective sense. There is really only what people make of information.

We hear, for example, a great deal about evidence-based policy. The problem with doing more with this splendid idea is that we are all highly selective about which bits of evidence we are prepared to take on board. None of us likes inconvenient truths, because accepting them involves confronting and possibly overturning deeply-held beliefs. The Cochrane Collaboration is an international not-for-profit organisation that is dedicated to the promotion of evidence-based healthcare.

A recent report produced by the Collaboration suggested, on the basis of a survey of the outcomes of randomised trials, that screening for breast cancer using mammography may have been oversold. Why?

Although mammography undoubtedly reduces breast cancer mortality, the screening system also over-diagnoses breast problems, causing unnecessary stress and expense for the women concerned. For every woman whose life is saved by mammography, another 10 will undergo unnecessary tests. Cochrane author Peter Gotzsche argues that, now more experience has been gained with the technology, it may be time for public health experts and for women themselves, to re-evaluate mammography programs.

However, the indications are not propitious for a rethink. So much intellectual and public capital has been invested in mammography-based screening programs that to pull back now may simply be too difficult politically.

Of course, it is not just public health programs that mix up politics and research. Indigenous programs are another example. For some years, academic research has pointed to the importance of governance (both bureaucratic and indigenous) in improving conditions for indigenous communities. But too often, the evidence of research is overturned or ignored in a rush of political impatience. The irony is that the Information Age overwhelms us with information without improving our skills in dealing with it. It's particularly hard to convince the young, who are in the habit of looking everything up on Google, that they still need the old-fashioned skills of paying attention, analysing and thinking in order to work stuff out for themselves. What else has been lost?

In the age of the email and the blog, although we probably read and write (after a fashion) more than ever, we don't generally have the time to polish what we do. We are still trying to figure out the consequences of this. I am not sure, for instance, what scholars will do when famous people no longer write letters.

Robert Menzies's letters to his daughter make an interesting collection. Today, he would send her emails, which are both ephemeral and difficult to manage. Incriminating ones can apparently be dug out of the ether. But there are always one or two that you could have sworn you sent, that have somehow evaporated. It must be a nightmare, in the modern public agency, keeping track of the information flows.

Since the Public Service itself stopped issuing standards, the Archives Act is the only centralised records management legislation governing what departments do in this respect. I don't envy them their task, when group processes involving emails lose themselves on long, incomprehensible trails. Perhaps that is why it is still necessary to hold meetings, as this may be the only way to figure out where everyone is up to.

In all this it was refreshing to obtain a letter from an ACT government department. The department, I was informed, was currently refining the work and timeframes for the background studies that would be used to inform the review of the issue I had inquired about. To ensure overall coordination, the scope of the work was in the process of being coordinated with a number of other agencies. I would be consulted following refinement of the background studies, consultant briefs and the establishment of proposed timelines.

Just when you thought old-fashioned bureaucratese had died, there is a perfect example, as dazzling and dismaying as ever.

Jenny Stewart is professor of public policy at UNSW Canberra. 

This opinion piece first appeared on The Canberra Times.