Bill and Brad's Excellent Adventure
The relationship and behaviors behind and between scaling companies, internet connections, and cities
More Microsoft history and it’s role in the growth of itself, other companies around the world, and our pervasive connections to each other and the internet.
I was lucky to play a small part in this transformation. I did my best to understand the behavior of people using the software that fueled these expansions. It taught me lessons I’m now applying to understanding behavior of people interacting with place.
As interactors, you’re special individuals self-selected to be a part of an evolutionary journey. You’re also members of an attentive community. I welcome your participation.
Interplace is a place for people to interact so please leave your comments below or feel free to email me directly.
Now let’s go…
In my early days at Microsoft, the place felt more like a startup. I had come from a company of 120 people, so 11,000 people should have felt huge. But compared to the 168,000 of today, it was small. The culture was different too. People were there to make something – not to be somebody. This attitude is best ensconced in Douglas Coupland’s 1995 novel, Microserfs. Set on Microsoft’s forested campus, it gives a peek into ‘geek culture’ of the early 90s. It also exposes the heartfelt belief that we really were doing something meaningful.
“What is the search for the next great compelling application but a search for the human identity?”
“Maybe thinking you're supposed to 'have a life' is a stupid way of buying into an untenable 1950s narrative of what life *supposed* to be. How do we know that all of these people with 'no lives' aren't really on the new frontier of human sentience and preceptions?” – Douglas Coupland, Microserfs
When this book came out, Windows 95 was just starting up.
PC sales were revving. Selling at the rate of 40 Million per year it’s success pushed Intel’s revenue growth to 41% in a single year. And the promise of the World Wide Web was taking hold. Bill’s book, The Road Ahead, had come out and he had a vision. That vision was expressed in a keynote at COMDEX, a popular technology conference. The concept was called, ‘Information at your Fingertips’ and he needed a murder mystery movie to tell the story.
My team designed the PC desktop interactions in this video. We imagined what interactions would be like in 2005. At one point in the video, Bill jokingly refers to our work as a ‘rather cluttered desktop’. Soon after you’ll see Bill introducing long file names and video conferencing. But in 1995, that meant video conferencing with somebody in a public phone booth!
People around the world were buying in to Bill’s vision — including major corporations and governments. Businesses multiplied PC growth. And so did connectivity. Bill’s vision of a PC on every desk was coming true. Connecting computers to the internet really did put information at your fingertips.
Buying a PC in 1995 was the entry price to a new world of communication. Windows 95 made connecting to the internet easy, Office 95 simplified sending email, and the Internet Explorer let you, well, explore the internet. Information was not just gingerly touching the tips of fingers. Hands were cupped and people were scooping up tiny bits of addictive bytes and gorging themselves. We still are. In 1996 36 million people were on the internet. By 2000 it grew tenfold to 360 million1. A decade later it jumped to two billion. Estimates now put that number at around four billion – over half of the world’s population. A reminder of the advantage and privilege half the world holds over the other half.
GROW BABY GROW
Satisfying this growing appetite for information required more than software. Cables needed to be run, server farms needed built, and somebody had to manage all this information and technology. Enter the IT Manager. IT had been around for some time already, but installing, managing, and controlling access to Windows, Office, and all these PCs and peripherals put real strains on IT departments. So did the need to control access to the exponential growth of documents, spreadsheets, and presentations. Other comparable modes of communication, like snail mail and telephones, only grew at a rate of around 10% over a much longer period of time.
The expansion of IT departments mirrored this curve. Digital connection and communication between people inside companies fuels the same ingenuity that physical connections do in dense cities. The bigger the city, the more innovation occurs. Successful cities attract talented and conscientious people generating superlinear population growth and creativity. It’s why bigger cities and bigger companies have bigger stockpiles of patents.
“Just as bounded growth in biology follows from the sublinear scaling of metabolic rate, the superlinear scaling of wealth creation and innovation (as measured by patent production, for example) leads to unbounded, often faster-than-exponential growth consistent with open-ended economies.”
– Geoffrey West. Scale: The Universal Laws of Life, Growth, and Death in Organisms, Cities, and Companies.
To accommodate this growth, the infrastructure must scale accordingly. But whether it’s an ethernet cable plugged in to a PC, a copper pipe connected to a toilet, or the capillaries feeding those information seeking fingertips the endpoints all start small. Universally small. Be it Boston, Bamako, or Bangladesh, the clear plastic fitting at the end of that network cable is the same. The diameter of piping connected to the toilet is also roughly the same. So are the smallest blood vessels in every human’s freakishly flexible fingers all they way down to their tiny typing tips. All of which connect to progressively larger conduits that accommodate increasing amounts of data, water, or blood. All the way to massive data gateways, waterworks plants, and plasma pumping hearts.
“The pipe that connects your house to the water line in the street and the electrical line that connects it to the main cable are analogs of capillaries, while your house can be thought of as an analog to cells. Similarly, all employees of a company, viewed as terminal units, have to be supplied by resources (wages, for example) and information through multiple networks connecting them with the CEO and the management.”
– Geoffrey West. Scale: The Universal Laws of Life, Growth, and Death in Organisms, Cities, and Companies.
It’s the work of fractals and the power of sublinear scaling. It’s what makes an elephant live longer than mouse, and companies like Microsoft to live longer than most startups. But an elephant is also big and slow, and so are big companies.
As the year 2000 approached, other growing companies suffered the same fate of growth. The burden of managing these layers of modern administration increased the total cost of ownership of assets. The cost of managing equipment, software, and data, exceeded the cost of the software used to create it. Office 97 was deemed ‘good enough’ by the industry. This meant every new feature we added only diminished the return on their investment. Microsoft Office faced the same fate natural organisms do – a growth plateau.
POWER TO THE PEOPLE
This marked a notable shift in our approach to user research. Instead of focusing entirely on the tasks and activities of individual end users, we had new problems to consider: reducing the total cost of ownership and increasing the return on existing investments. Our infamous attempt at expanding menus in Office was one such example. Instead of a bottoms-up ‘geek culture’ approach to software development that was tied to individual needs and desires, we were looking through the lens of IT managers and business decision makers.
Meanwhile, the kind of things people were doing in Office was expanding and diversifying. We discovered people used Excel for everything from poetry to programming. Word users made love letters and legal documents. We were constantly surprised by how people used Office. I recall a site visit where we observed an office manager at a car dealership using Word. She was making a flyer to post in the break room. We watched as she created a new document, typed and formatted some text, spent a few minutes laying it out, and then hit ‘Print’. What happened next surprised us. She closed Word, and was dutifully prompted to save her document. To our amazement, she declined. We politely interjected, “You just lost all your work!” She responded, “No I didn’t, it’s sitting right there on the printer.”
As we observed more and more behavior we found ourselves looking across individual use cases in search of commonalities. We would cluster and clump collections of behaviors into buckets with names like, “Create”, “Communicate”, “Synthesize“, and “Collaborate”. Before long we distanced ourselves from those individual behaviors. The language became obtuse and abstract. The knowledge of how people were using the product on the ground became mediated through layers of corporate administration.
There were decisions being made among the ‘corporate elite’ in corporation that had measurable impacts on how people were using Office. Decisions that impacted their satisfaction of our product and of their livelihood. But much of that became insulated from us. IT didn’t much like Microsoft researching their employees. Knowledge of how real people did real things in real ways was information just beyond our fingertips.
STRIKING THE BALANCE
The same effects of scaling occur in cities as well. Small towns are like startups. They have their own version of ‘geek culture’; transcendental motivators that rally a group around something bigger than themselves. As the city grows, it too attracts the best and brightest. And as the population grows, so does the infrastructure needed to sustain it. The close interactions of smart talented people yields new innovations that generate revenue and attract more talent. City growth seems to be impervious to scaling laws. Luis Bettencourt, a physicist and complexity scientist who studies cities, calls them ‘social reactors’. By his estimation, there’s only one thing in nature that behaves like cities. The sun. In a Santa Fe Institute interview by host Michael Garfield , Bettencourt says,
“…the only system that has sort of these properties that I know of exist in nature that concentrates things, increases interaction rates and admits products at a rate that's higher, per unit of mass is the sun, is a star and that's a reactor, right?”
Researchers have been observing and understanding the behavior of people and place for centuries. And they too have struggled with the same things I did trying to understand the behavior of people using complex systems like Office. Studying people on the ground indeed yields insights at a micro-level. But after collecting and amassing mounds of data patterns emerge, generalizations are made, and soon they’re looking at humans in the aggregate.
Researching at the macro level-begins with agglomerated aggregate. But patterns also emerge. Moreover, it allows for the study of behavior of decision makers, policy writers, and power brokers. These decision makers work together to convive, concoct and sometimes conspire ways of influencing the outcomes of organizations. It’s important we study how planning and long term execution of coordinated strategies impact the world. Only then will hidden influences of our economic and political systems emerge. In synthesizing decades of human geography research by social scientists, researchers Reginald Golledge and Robert Stimson came to this conclusion:
“It was found that the long-term plans and objectives of Western capitalist societies often benefited small, elite groups of capitalists more than they benefited people with low incomes and that planning for personal welfare diverged widely among capitalist and socialist and social-welfare economies. Decision making at the macro level was thus defined as a constrained process, undertaken and implemented by a powerful elite, often encompassing goals that differentially considered segments of society or economy.”
– Reginald G. Golledge and Robert J. Stimson. Spatial Behavior: A Geographic Perspective
Understanding human behavior is a challenging and dynamical undertaking. After all, in the words of Douglas Coupland,
“What is human behavior, except trying to prove that we're not animals?”
I’m not convinced we’re smart enough to fully understand or describe human behavior. It’s like looking through a telescope with one eye and a microscope with the other and then coherently explaining what you see. But it does involve a little bit of both of these techniques. It requires us to sit with people in their context to fully grasp the how their physical environment impacts their work, lives, attitudes, and beliefs. When you see someone, as I did, straight out of college sitting in a broom closet with a desk and a laptop using Excel for 10 hours a day, it’s hard to un-see.
Somebody in ‘corporate’ decided it was a good idea to put a human in a closet to work. Like an animal. It’s essential we understand why people do these, and much, much worse things to other people. Every organization has a group, or two, of just a few people with a privileged view. They reckon and wrangle, poke and prod, and wager which message to tangle and skew. I’ve been there, I know, they impact me and they impact you.
The same is true in the cities we live. And the regions and counties and countries, too. By trying to know all we can know and showing the world what we’re we willing to show, a bigger vessel of understanding can flow. In the words of Reg Golledge and Robert Stimson,
“A paradigm for examining human-environment settings needs to encompass a complex set of relevant variables and their functional relationships. It includes the physical and the built aspects of environment; it allows for roles of culture and its related social and political systems and institutions; it identifies the evolution of culture over time through technology; and it recognizes intervening psychological processes as filtering mechanisms in how humanity perceives the environment and acts within it.”
Our collective behavior has fueled a rate and pace of growth that we cannot survive. Microsoft, as a company, has survived longer than most. Looking at over 22,000 U.S. companies, from 1950 to 2009, only half made it past 10 years.2 I’m half way to 60 and many scientists would agree that I’m as old as the Anthropocene – a geological epoch marked by the negative impact our behavior is having on the planet. The majority of my life has been spent molding and understanding the behavior of people behind a computer screen – billions of finger tips making, taking, and shaping information through software. My one wish is that they, we, look at our reflection in that screen and ask, “Will my current behavior allow the next two generations to survive?”