The Virtual Furniture Police
This is a review, of sorts, of the book Peopleware: Productive Projects and Teams by Tom DeMarco and Timothy Lister. Then a segue to explain how typical corporate IT policies contravene some of the excellent advice in this book.
The book is ostensibly about knowledge workers of all sorts, but seems to be really intended for software developers in particular. I had heard of this book for a while, but only after seeing it enthusiastically recommended by Joel on Software, I was finally tempted to give it a read. I'm glad I did; the book has helped me to solidify some opinions that I had half-formed over the years. To that extent there is not a lot of material here which is surprising, but it is nonetheless great to see it reaffirmed in print, instead of around the proverbial water-cooler.
Our Corporate Overlords: Malevolent or Benign?
Joel describes the book as the "Anti-Dilbert Manifesto" and I think this is pretty accurate, for a three-word summary. However when I read Dilbert I am often reminded of the line from Apocalypse Now:
Kurtz: Are my methods unsound? Willard: I don't see any method at all, sir.
My own experience of corporate malfeasance is far more benign than that displayed in Dilbert. At my workplace, there does seem to be signs of humanity at work in even the most boneheaded of decisions. The random and malicious Catbert is in my experience the exception, not the rule. But maybe I'm just lucky.
Nevertheless, the idea of the well-intentioned-but-misguided management style is echoed in Peopleware. The central theme is to reject the widely-held notion that knowledge workers are — or should be — identical and interchangeable within the parameters of their role. The origins of this notion are touched on incidentally throughout the book, and frankly I would like to have seen this explored a bit more. Sometimes certain management practices are put in place for traditional reasons, sometimes for perceived efficiency, sometimes for short-term economic expediency, sometimes for corporate (or individual) vanity. The effects of all of these are described in great detail in the book.
To its great credit, the book explores many ways in which treating knowledge workers as modular and interchangeable, and ignoring the real and valuable differences in personality and background, lead to dysfunctional individuals, teams, organisations and processes. There are many variations on this theme, but I wish to focus mainly on the effect on individuals.
Empty Desk, Empty Mind
One of the many catchphrases introduced by the book is the Furniture Police. This is prettymuch a function of corporate vanity, where each employees' desk is (in the most egregious example) allowed only a single picture frame of their family members. Desks must be kept free of clutter and non-sanctioned furniture.
DeMarco and Lister give a fascinating account of how they measured (with admitted imperfection) the effect of the office environment on the productivity of a software developer. I had heard of (and seen!) the wide variation in individual productivity amongst developers, but had no idea that work environments could also be correlated with productivity variation to a similar degree: about 10-fold. In other words, the best and worst work environments represent a difference in productivity of about an order of magnitude. This is of course a very difficult claim to measure, but I think they show some justification.
In some obvious cases the work environment has a clear effect on productivity. Some days it seems that you just put the headphones on to shut out the world and try to get some work done — which according to the book is in itself a warning sign you have problems with a disruptive workplace — when the phone will ring, or someone will arrive at your desk, or email, or whatever. This seems to happen to me a lot more lately, but maybe I am just more aware of it after reading the book!
In other cases the work environment has a less pronounced, but still avoidable, adverse effect on productivity. The misplaced desire for conformity, if not uniformity, stifles creativity, represses individual styles of working, and subtly impedes the productivity (not to mention happiness) of workers. The Furniture Police are a great example of this. Dress standards are another. In yet another memorable anecdote, DeMarco and Lister cite a company that forbade the microwaving of popcorn as "unprofessional". They went on to give that particular term the scathing it deserves.
The Virtual Furniture Police
I would argue that many corporations attempt to exert the same degree of control on the virtual workspace as they do on the physical workspace. In this case the motivation is slightly different, but the effect can be just as harmful. Peopleware does not explore this topic, but I will.
In many companies there is an IT/MIS department which is responsible for making sure the technology runs smoothly, or at least limps along acceptably. To do this task effectively, they must assume that each person in the company has a workstation that is as similar as possible to all the others. Promoting a monoculture of corporate IT assets is seen as Good Practice amongst this crowd. And when you look at it a certain way, I guess it is.
Certainly if you give me the choice of supporting 10,000 people all running the same hardware and software, or 10,000 people running whatever damn hardware and software their respective nephews and cousins and friends told them was hot this week, well obviously I'd take the former. People are by nature lazy, and don't necessarily take the time to update their virus scanners regularly, or configure their firewalls, or any of a hundred other administration tasks that is required for a smooth-running network. So it's probably best that the IT department take over these essential functions for most people in the company.
But really, does IT really need to:
- restrict access to the screensaver settings?
- remove all access to the control panel so that you can't even change the screen resolution? (anecdote courtesy Julian) Too bad if your eyesight isn't great.
- take steps to actively prevent browsers other than the standard one on corporate websites?
- prevent users from installing applications or even accessing their hard drives?
- block access to CD-ROM, USB and floppy drives?
- use content- or blacklist-filtering on internet access?
Any real or imagined benefits of these measures need to take into consideration the real productivity benefits of allowing users to work in the ways that suit them, as well as the hidden costs that come with the clear indication that the organisation doesn't trust its workers.
These measures are especially pronounced for software developers, who simply require control of their development systems in order to do the job at hand.
LOL, IT Pwns j00!
It seems that we still have a long way to go in finding the right balance of user freedom and IT department convenience. While researching this post, I found a lot of discussions amongst IT staff as to how to control their users. Without a doubt, the prevailing attitude was one of restriction and control. Case in point: CIO blog having nightmares about his users accessing ... wait for it ... flickr, the photo sharing site (The Horror! The Horror!). Even the slashdot crowd was not immune. OK guys, I get it: users are not to be trusted. Bring in the BOFH.
[Often there is a good, or at least stated, justification for exercising such tight control over the users: security. Sometimes it is backed up with the need for regulatory compliance (e.g. HIPPA, Sarbanes-Oxley, etc). It is for someone else to judge, but I would guess that there are many many instances where IT departments institute policies that go far beyond what is required by legislation, or that which represents a worthwhile security trade-off. It's a tricky area, one which has not, to my knowledge, been subject to much objective scrutiny.]
Not trusting your workers may be the safe thing to do, but the Peopleware authors argue strenuously that it is certainly not conducive to building productive people and teams:
The most obvious defensive management ploys are prescriptive Methodologies ("My people are too dumb to build systems without them") and technical interference by the manager. Both are doomed to fail in the long run.
To that list, add a restrictive IT environment born out of mistrust.
I never thought I would ever quote Ronald Reagan on this blog, but he described an ideal attitude that IT departments should take towards their users: "Trust, but verify". Give people the responsibility to act autonomously, and with accountability. By all means, standardise on a default IT environment, but allow users to opt-out if they so wish. The IT department will need to monitor the users who have opted-out - along with all the other users who haven't.
In order to make a similar point, DeMarco and Lister quote (of all people) Chairman Mao:
Let a hundred flowers blossom and let a hundred schools of thought contend.
Of course Mao didn't really mean it, but we do.
All this would assume that there is a overarching concern for long-term costs and benefits, where adverse impacts on one part of the organisation would be traded off against benefits in another part. Don't count on it though, IT outsourcing is accelerating.
In other words: let exactly one corporate standard flower blossom.