Monthly Archives: March 2013

The Open Office and How to Work Better

The open office plan is older than you think:

In the spring of 1962, a fourth-year British architectural student … stumbled across a small article in a trade magazine about a new workplace design that had taken hold in Germany…as “fundamentally a reaction against Nazism”.

It’s hard to set it up correctly:

The layout was based upon an intensive study of patterns of communication – between different parts of the organisation, different individuals.

And there’s a delicate balance to be struck:

Loud background noise … affect[s] workers’ ability to concentrate, and therefore their productivity, [but] in some open-plan offices, particularly the “classroom” type, there can be too little noise.

There may be a technological solution to the noise problem:

A certain amount of noise seems to be desirable – like the hum of a busy restaurant that allows a table of two to enjoy a private conversation.

Some companies are looking to technology to help get this balance right – broadcasting “pink noise” from speakers (a sound similar to white noise, which makes human speech less discernible).

But in any case, Europe has moved on from the open plan office:

Northern European office buildings today are “highly cellular”, he says, with everyone having “the right to a window they can open, a door they can shut and a wall they can beat upon”.

This sounds nice; why don’t I have it?

In the UK and North America, by contrast, design is mostly driven by cost rather than worker satisfaction, and open-plan layouts remain the norm.

The future may be unrecognizable to our parents:

[Architect Alexi Marmot] describes a building she visited in Switzerland which offered workers a choice of sofas, coffee table areas, libraries, pool-style recliner chairs and even “a botanical garden with a few work tables among the plants”.

As always, it boils down to a last-mile problem of designing context to shape human behavior:

But to give employees the freedom to wander about with their laptops, hiding from colleagues or seeking them out as they wish, may mean some organisations have to rethink the way they work and communicate.

“The building’s easy, the architecture’s easy,” says [architect Frank] Duffy. “It’s thinking about how to use the buildings that really is challenging.”

HT @timharford

Lamb’s Blood and Regression to the Mean

Q: Is smearing lamb’s blood over my doorway unnecessary?

A: Yes. You make sacrifices because they’ve always worked for your ancestors. However, your ancestors were not well schooled in statistics and experiment design. They offered sacrifices only when times were dark and all other means had been exhausted. Since (for example) the amount of rainfall is random, regression to the mean suggests that it will tend to get better after it has done really badly. Therefore, most sacrifices seem to work, and you are violating zoning laws and pissing off your neighbors for no reason.

Happy Passover!

(Inspired by Karen Armstrong’s Buddha.)

Portfolio Theory and the Laboratories of Democracy

The common thinking is that the USA benefits by having many states because they will have different policies and we can thereby learn which policies are most effective.1 This idea is intuitive but actually I think the opposite is the case: as government gets larger it becomes more (theoretically) capable of doing the kind of experimentation that leads to better policy.

Think of a government as holding a portfolio of policies. Each one has an expected return and a variance.2 Some policies are less likely to succeed than others, but they make up for it by having higher potential upside or being cheaper up front. However you wouldn’t want to construct a portfolio out of only risky policies because you don’t want your whole government program to fail.

So it’s elementary that one governing body making 50 policies will have more experimentation–and learn more, and have better policy next year–than 50 governing bodies each making 1 policy. In fact in the latter case they would probably all take the safest possible policy.

To be sure, states may may differ on what they think is the safest possible policy (or what a good outcome looks like). For example, several states are moving toward legalizing marijuana and the federal government has not done so. But state “experimentation” likely crowds out federal experimentation. If we had more real, by-design experiments on the federal level, that would be the weak point that advocacy groups3 would attack. State level policy is a hole in the dike; the pressure would come through somewhere else if this channel were blocked off.

Moreover, even though states differ, they are still all doing what they think best–they are not, by and large, consciously experimenting. And experimentation has to be conscious because experiments can be designed to maximize what you learn from them.

Think for a minute about classroom size. States (or districts) have different maximum classroom size rules, so we can use differences in outcome to see the effects of classroom size on learning. But then we learn only about how classroom size matters between 20 and 30 students in a class under standard teaching methods.  On the other hand, a conscious experiment would introduce much more variation and we might find something new entirely. This is another way of saying that state “experimentation” can only tell us where the local maximum is, but we might be stuck on a very short hill next to a high mountain.

One other concern is that some policies (for example, voting rules) are harder to experiment with on a centralized level because the unit of analysis is the government. That’s a fair point–but most policy is not like this.

For most policies, looking at portfolio management theory suggests that a centralized government is likely to take more risks, do more experimentation and learn more than decentralized states.

  1. This idea is known as “laboratories of democracy” and was brought to my attention by the Charles Pierce series of the same name. []
  2. Whether you think about this on the level of “the government wanting to be effective” or “politicians wanting to seem effective to get re-elected” the analysis is the same though some of the results may be different. []
  3. Or lobbies, or special interests. []

Words, Feelings, Memory, The Thing Itself

A whole canefield of words has grown up between La Maga and me, we have only been separated by a few hours and my sorrow is already called sorrow, and my love is called love. . . I shall keep on feeling less and less and remembering more and more, but what is memory if not the language of feeling, a dictionary of faces and days and smells which repeat themselves like the verbs and adjectives in a speech, sneaking in behind the thing itself, into the pure present. . .

What Are Other People Like?

How do we know about the internal lives of other people?

First, most importantly, we have to assume that they are similar to our own inner lives. This means that we see the world as much more homogeneous than it actually is because our own experience can only ever be one data point. On the other hand I talk about my inner life pretty often and other people are often surprised–am I calibrating my expectations correctly? Or are the reactions systematically tilted one way or the other?

Second, we get some sense of other people’s inner lives from reading–especially novels–and from conversation. There is a large selection effect–we only know about the inner lives of introspective people. This is troubling, since many people are not introspective. Though it’s quite hard to distinguish between people who are not introspective and people who just don’t talk about this stuff. This gets to the larger point of this post–what would it even feel like to not be introspective? Our only idea is from those times in our own lives when we don’t feel introspective.

Should Robots Have Emotions?

Yes.

My thinking on this issue was sparked by Emotional Design by Donald Norman (recommended). DN talks more generally about the role of emotion in designing technology and concludes that we should try to program robots to have emotion for at least two reasons:

Emotions are helpful to the robot. Imagine you have three robots in your house: a general household robot (Tony), a coffee-making robot (Manu) and a dishwashing robot (Tim). First thing in the morning Tony’s highest-priority job is to get you coffee, so he goes to Manu to pick up some coffee. Manu in turn requests a clean cup from Tim. But Tim is waiting for Tony to collect dirty cups from all over the house. So if these robots are playing “by the book”, they will get stuck and you will never get coffee. And you will never be able to program all of them for all possible situations like this one. On the other hand, if you program Tony with a feature that looks like frustration, he’ll eventually move on to his other tasks, pick up the coffee cups, and then come back and get you coffee.

Emotion–or even the appearance of emotion–is helpful for the humans using the robots. This is much simpler: we are intuitive and emotional beings and on some level we interact with everything–trees, programming interfaces, multinational institutions–as though they were also emotional people. Designers will be more successful when they plan for this tendency and design machines that cater to it. Machines that use or show emotion are richer and more engaging to interact with. Similarly, as machines become more important in my life, the emotions I feel when I deal with machines become more important: my life will be improved if I can avoid just being angry at stupid machines. Boring example: If the cable box is displaying regular definition instead of HD, my first inclination is that the machine is stupid or stubborn–and I feel angry. But it would not be hard to make me to think instead that the machine is confused but trying to solve it–and I might sympathize with it, which would be better for me.

This is a limited summary of a small part of the book, and I’d recommend that you buy the book (or at least borrow my copy). Future posts are likely to link back to this, so get excited.

Data-Driven Decision Making

I spent the weekend at a conference about the use of statistical analysis in sports decision-making. Among other things, we talked about player evaluation, the role of randomness, in-game coaching decisions, and optimal risk taking based on game situation. All of these through the lens of using data to improve how we act.

Just now I went to the grocery store wearing flip-flops: “oh, I don’t think it’s that cold outside.” Turns out it’s 34 degrees.

You can lead a horse to water, but sometimes he forgets to drink.