"These are not the facts you're looking for"

"Open-plan offices are bad for productivity and employee health", says an Australian study.

Now, I've been a technical journalist. I know how the news reports "studies". And this is one of those (countless) cases where I'd really like to see the parameters, as well as the results, of the original study. Because "open-plan office" can mean pretty much anything.

"The research found that the traditional design was better - small, private closed offices." There's that whore of a word, "traditional", again. What does it mean this time? How many industries have "traditionally" given people private offices?

Not journalism, for sure. Every newsroom in the world, I believe, is and always has been open-plan. It's fitting that the news should be written in a noisy, distracting environment, because that's just how most people will be reading it.

Education. How many schools give each teacher their own private office? Not many, where I come from. Heck, even at university, my tutor had to share an office.

Police officers. Hospital doctors and nurses. Travel agents. Bankers. Air traffic controllers. The 25% or so of the population that works in manufacturing or construction industries. The much higher number that works in retail. What percentage of these spend most of their working hours in their own private offices? And yet the expectation exists -- as if we all, somehow, thought we were destined to be treated like middle-managers in some 60s sitcom. In some sense, it feels almost as if we should expect that. A private office is a sign of status, of privilege, something we're somehow supposed to aspire to.

And that, of course, is how the issue has become politicised, and that's why I'm going to treat any study on the subject with grave suspicion. I now want to know, not only the complete employment history of these researchers, but also who funded them.

"The problem is that employers are always looking for ways to cut costs, and using open-plan designs can save 20% on construction."

And what about rent? Even allowing for the extra meeting rooms and social spaces you need to provide, you can still fit at least twice as many generous-sized desks in a given floor space if you don't have to lay out a whole room around each one.

I've spent a lot of my life working in at least four very different open-plan offices. (Well, six really, but for just four different companies.) In publishing, there is no doubt that open-plan is the only sane layout to have; if anyone in that industry claims that their colleagues are distracting them from their work, they haven't understood what their work is.

In software, the picture is more mixed. There's a persuasive argument for letting team members communicate with minimal barriers. In this view, if you put someone in a position where they have to stand up to communicate with team-mates, you cut down the flow of information by at least two-thirds; if you force them to get up and walk somewhere, it's more than 90%. But (the counter-argument goes), in practice these informal, personal communication channels don't scale well; they're fine when there's just three or four people in the room, but at any size bigger than that, someone will be left out of the loop. You need more formal communication, less reliance on informal.

On the other hand, there's an argument (put forcefully by, among others, Joel Spolsky) that says -- backed up by a whole pile of alleged facts -- that programmers in private offices are "more productive". Microsoft, apparently, is famous for putting all its programmers in private offices. Which just goes to show that productivity is no guarantee of quality.

I've enjoyed working in an open-plan office. It was quiet, because there was rigid self-discipline on all sides about talking; and the lack of privacy didn't bother me, because we all had a degree of trust in our management. You could spend a week doing nothing but surf the web, without feeling the need to conceal it; people would assume either you were working, or you were legitimately trying to work but simply blocked for the moment. Motivation was high, productivity was high.

But I've also seen the opposite. Partitions, stifling that easy, collegial non-verbal communication, while doing nothing to cut sound or provide privacy. Three or four conversations going on at once, sometimes across several intervening desks. Managers who make no pretence that they even want to know what we're doing, as long as it's work. Morale low, productivity suffers.

How many of these factors has the study taken into account? I don't know, and I doubt if I could find out, even if I tried. It's a politicised issue, and it's not hard to frame the parameters of any study to make sure it comes up with the right answer.