Delivered 2018-01-30 at API Days Paris, 2018, Paris, France



Prepared Talk








Transcript (Raw)

All right. I’m very happy to be here. As Mandy said, my name is Mike Amundsen. This is how you find me. This is how you find me on LinkedIn and Twitter and all those places I would love to connect with you and learn what you’re working on and what’s interesting to you in your projects as well.

I’m very lucky to work in a group called API Academy. Several of us are here this year at API Days and we’re sponsored by CA. CA has a booth and we’ll also be spending some time at a small table. I would love for you just to come up and talk to me and tell me what you’re working on. It’s a fun group. I had the pleasure of being here a couple times over the years and talked about several things, the 50 years of computing, mechanical turks, this idea about 100 years back to the information age a century ago. Last year, I got to talk about how AI and robots are affecting work and the future of work.

I really enjoy this conference very much. It’s an opportunity for me to think about things and share and get some great ideas. I’m also very excited that the theme this year is sustainable software, sustainable society. We know lots about sustainability and IT, all the way to saving energy for the devices that we use, actually creating devices out of recycled materials and recycling those same materials. There is even a model for green software, for making sure that we reuse not just the hardware, not just the material, but also the software itself. In fact, there’s a rather nice book about sustainable software development by Kevin Tate. I like this book because it goes beyond just the software and talks about the people and the organizations and how you sustain teams and lots of other things besides just the software itself.

What I’d like to do today if we have a little bit of time, is to go back to the roots of this word "sustainability" in the modern age, why you talk about this idea of the intersection of profit, people, and planet and how that’s been applied over the last 30 years to the physical world and see if we can get us to start thinking about how we can apply it to the virtual world as well. In order to do this, I’ve asked a friend of mine, Alex Rivera [SP], he’s a fantastic, talented, U.S.-based artist, to draw some images for me. We’re gonna do this a little bit differently. We’re gonna go on a little bit of a journey.

A few years ago, I did a talk that was very similar to this. It was this story about a character that went from a simple town through a magical forest to find this place where computers and humans interacted together. It was called the Hypermedia Tale. You can find it online if you’re interested, and it was also provided to me by another artist. At the end of that story, I actually said that it wasn’t really the end. There was really more to it. It turns out that this talk is about the other pieces that I hadn’t been able to mention at that time. In fact, the cities that were involved in that story, are part of a much larger community that is known as Omniterra. All the known world, and there are lots of places in this space.

Not just the story that I told a few years ago but several other stories that I’d like to share with you today, use those as examples of how we can talk about sustainability. The first place I want to start is probably the last place in the map. It’s actually the Ruins of Monolithia, this place where there used to be this grand city but it has now been left for ruins. It’s a rather impressive place when you visit it. It’s a vast city that looks so great from afar. Actually, if you were far away, you would think the city is still alive, but as you get closer you see there are lots of things in ruins, lots of carts, and buildings left undone, the roofs caved in, but you could tell that it was once a really fantastic place.

In fact, this used to be the center of this known world. People from all over came and traded and learned and shared all sorts of other things. It was sort of the place to be in this world and it had a rather unusual aspect to it. The kings were architect kings. And as each king ascended throne, the king would create new plans for the entire city and rebuild the entire city in that king’s image. The most modern, the most fantastic, the most high-tech that it could possibly be. As the king unveiled all these plans and began to rebuild parts of this city, there was always some part of the city that wasn’t working, that was being torn down, another that was being built. This happened over and over. You can imagine as more and more kings ascended, the city became bigger and bigger and it became a bigger and bigger job to rebuild this from the ground up until finally, at one point, it would take so long to rebuild the city that the king would be dead before the project was completed. But that’s okay because you’d have another king who would start again and again and again and we would simply rebuild everything over and over until it got to the point where nothing worked. Nothing was functional because everything was being rebuilt or torn down again.

Eventually people had to move away. The city was unusable and it was left for ruins and they moved out into the countryside. That’s a key story that we need to think about. What happened was we kept rebuilding and rebuilding and rebuilding and just throwing away what we already had. It turns out that that’s not very sustainable over time. To simply start again and start again is not a good idea. Let’s go back and talk about sustainability. The modern usage of the time, this idea of sustainability in the environment and in corporations really comes from a thing called the World Commission on Environment and Development from 1987, about 30 years ago.

Then Prime Minister of Norway, Gro Harlem Brundtland, led a committee and created this report which is now published as Our Common Future which is a great small book. I’d encourage you to read it and it lays out this notion of what sustainable development is. It meets the needs of the present without compromising the opportunities in the future. That’s a really challenging way to think about it. How can we meet the needs of today without compromising what happens in the future?

In fact, this idea of sustainability goes way, way farther back. Five, almost six hundred years ago there was this gardener by the name of John Evelyn and he created this notion that since we use wood for so many things, that we needed to take care of the forest. Take care of the trees. He wrote this book called "Sylva" about forest trees and how to maintain and sustain timber. That’s why we have forestry today. That’s why we study forestry today, this notion that timber was this precious resource that we have to constantly maintain.

There are lots of versions of this. Around the same time, around the end of the 17th century, start of the 18th century, this idea that we needed the sustainability. The challenge is we have to think about the future in some sort of way. We have to be prepared. We can’t just do something just for today if we want it to be sustained. How do we plan for the unknown? How do we make sense of something that’s not here yet?

Someone who had a great way to talk about this is Mel Conway. He wrote this paper about 50 years ago about how people invent and how the virtual world works, how people work together. And he had this great observation that as soon as you make a delegation, as soon as you make a decision about something, you’re cutting off other opportunities. The minute I decide who’s on the team, then we’re limited to the skills of who’s on that team. The minute I decide not to use a particular product or to use some other product, there are all sorts of things we can no longer do. Each step, each decision affects our future and that becomes really, really powerful.

Another pair of individuals, Tom and Mary Poppendieck who brought the idea of lean software for manufacturing into software say it a slightly different way. They encourage you to delay your commitment until the last responsible moment. How can we put off decisions until the last responsible moment so that we don’t cut off too many opportunities?

They have this nice twist to it. They have this idea that if it turns out not deciding eliminates possibilities, that is probably the responsible moment. This notion of sustainability it turns out has a lot to do with time. We need to use time to our advantage. Often time is this difficult thing for humans to grab. We don’t really estimate time very well. We don’t remember the past very well. We sort of imagine a future in sort of a quirky way. It’s hard to have a shared future. But time is really important.

This idea of time and time passing is really another opportunity for us to talk about another place on the map. That place is actually what’s called the Salt Caves of Persissus [SP]. There’s these caves up in the north where there are people who actually live and they store all of the valuable things from the city. People would come and they would collect up all of their valuables and they’d take them out to the caves to the monks who would store them in all these locations. The monks of Persissus had this ability to understand just what objects were, to label them, to characterize them, to put them all in the right place. Every time there was an event, a family event, a wedding, or a celebration of some type, people would come out and they would ask the monks to retrieve all their things so that they could enjoy them for a few days for a festival and then put them all back in the caves again.

As the city grew larger and larger, the caves had to be expanded as well. There were monks whose only job was to simply move things around inside the cave. Some never even saw the light of day. They were called the Dark Monks. But they kept everything for us and they kept everything available to us. Sometimes the people never asked for those things back, for all sorts of reasons, maybe they had forgotten they were there, maybe the family died and no one took care of the extra things. But the monks never threw those things away. They might put them in another place in the caves, sort of like cold storage, something like that, but they would never take them away.

As the cage grew more and more vast, there were some monks that just spent all their time moving things about in the end. But they always had this interface each way to hand things over to others. They kept what was valuable close at hand and they didn’t decide to throw things away just because nobody needed them now because they were thinking about the future as well.

There are lots of things we can learn from this notion about planning for the future and how we store valuable things. Often this is expressed as this idea of profit in the profit people planet model. In fact, the profit people planet model comes from a gentleman by the name of John Elkington. John Elkington was actually active on the Brundtland Commission, and he had this idea about how we could turn the Commissions' mandate about our common future into something tangible. So, he created what was called the Triple Bottom Line, TBL, sometimes 3BL. That means not just dealing with the economic value, but also the environmental value and the social value. And these can help us a lot when we think about the virtual world as well.

Elkington said we need all three of these in order to make sense of what we’re doing. We need to understand the economic element which we’re pretty good at. We also need to understand the social element which we don’t pay a lot of attention to often but we’re not too bad. And then you can think about the environment. The trouble with thinking about the environment is that the environment is very indirect. It’s very long-term. We have to make these balances and we have to make these tradeoffs. It turns out it’s difficult to trade one for the other. You have to make lots of sort of tough decisions along the way.

It turns out being effective and being efficient are not the same thing. In efficient setup for storage, monks would throw things away after a certain amount of time, when you didn’t care about them anymore. They might store everything as the same and they might put everything in one place. But effective means saying, "It’s still valuable. It’s still useful and lasting a long time." It turns out sometimes these two things conflict with each other. Being incredibly thorough or being incredibly effective or efficient are often in conflict. It turns out Erik Hollnagel, who’s kind of a popular person in the dev ops world, talks about this a great deal, even wrote a book called the Efficiency-Thoroughness-Tradeoff. Every day we make decisions about, you know, "Should I be very thorough or should I just be quick and easy?" In fact, in software, we asked our bosses this all the time, "Do you want it right or do you want it on time?" Right? We make these questions every day.

How I get to work, how I get to work on the train or if I’m in the U.S. I drive or whatever the case may be, I’m constantly making these decisions. Every once in a while, those decisions turn out badly. I always take the train but I decide to drive today and I get in a car accident. Right? It turns out badly. That’s not my fault. I just was making the same decision I always make.

It’s important to think about this idea of trading off. Trading off often means that we have to think ahead. Sometimes, the decisions we make won’t be correct. I love this quote from George Box who is a statistician that just died recently, just a few years ago, and he has this great line, "All models are wrong but some are useful." One of the biggest challenges that we have is the models that we have about the way the world works, the physical world, the virtual world, they’re not always correct and often they have to change. Changing our models is something very, very difficult, especially when you write software because often we write software with one model in mind. Creating sustainable software often means figuring out how I can create a way that other models, models that are not mine can actually be applied and used.

There’s a line that I shared about a year ago. We were just talking about it last night, a bunch of us sitting together. That is the models in software, the model for your data is not the model for your objects, right? The model for your objects is not the model for your HR [SP] resources. And the model for your resources is not the message that you pass back and forth. Each one of these are independent elements. If you can think about them independently, you can do the same thing that Poppendieck was talking about. You can put off decisions until later. You can change the message model without affecting the data model. You can change the object model without worrying about the resource model. You can create these elements where I can decide later. Deciding later [inaudible 00:15:39] sustain the designs we have a little bit longer.

What we’re focusing on is not is just the short-term ability to create something that’s efficient but also the long-term effectiveness, the actual value built into our network, the value built into our software. That’s really a tough challenge. This notion of finding the value and our models being different leads us to the next place on our map. This is another place called we’re gonna visit the Scholars of Aggregato [SP]. This is a community where people are the engine of knowledge, people are the engine of learning. The Scholars of Aggregato are rather peculiar. They’re sort of savants. They each know something very special and they’re very good at it and they can recall information very quickly. The challenge is the information that they know is not always useful by itself. It needs to be added to others.

There’s this community where everyone has this special knowledge but the real skill they have is this ability to work together as a team to solve a problem. There’s this marketplace where people can come and go and mix and ask questions and they gather up little groups of people and talk amongst themselves and pass information along until finally they get to the answer they need. It’s this collaborative cooperative experience. It’s not just a marketplace like this, but it’s a marketplace of knowing. Their ritual for putting all this together is known there in the Scholars of Aggregato as the "dance of knowing." All the processes, all the things that we put together, all adds up to understanding something.

That really leads us to the second element in Elkington’s model: the people. People are incredibly important. In Elkington’s model and the sustainability model that comes from the Brundtland Commission and others, this is the social variable. This is education and equity and access to resources and well-being and quality of life and social capital, all these other things that are really difficult to quantify. Who really has access to information? Who can get access to a computer? Who has access to a job? Who has access to health care? All of these things kind of fall into place and we have these same challenges in the virtual world as well. Who has the tools? Do we all have open source tools? Do some people have better tools than others? These are all things that add up.

There’s a great story that former President Barack Obama was visiting Silicon Valley a few years ago and kind of being lectured to about, "How we would do it in Silicon Valley. This is not a tough problem. We know how to solve problems there." He had this great observation. He said, "You know, government’s never gonna run like a startup or like Silicon Valley because democracy itself, putting people into the mix is messy. It’s not predictable. It’s not deterministic." He said part of government’s job is dealing with the things that nobody else wants to deal with. He said, "If my only job was to create one app and I didn’t have to worry about the people who couldn’t afford it, my job would be easy. But," he said, "that’s not my job. The app is the country and my users are everyone. That’s a much better job. I can’t eliminate someone."

One of the challenges in software is there are lots and lots of people who aren’t included in the equation. We don’t care about them, that they can’t afford it, that they don’t have access to it. That happens even inside the organization as well. We don’t often think about people as individuals, empowered and enabled individuals. We think of them as resources, as elements that we can use in some sort of way. That can also be very challenging. This idea in the community where we have this "dance of knowing," this is what we have with every team. We have this notion where we can collaborate and we can share and we can think. But if you design an organization where you’re mostly competing against one another for a job, mostly competing against other teams, collaboration isn’t rewarded very much at all.

It turns out we’re connected in fundamental ways, fundamental ways that we forget, that are easy to forget because they’re not so tangible. Computers and networks teach us about that connection. Social media teaches us about that. The World Wide Web teaches us about that but it’s difficult to live that life every single day. I wanna quote one of my API Academy colleagues on this. "The structure of our company, the way we actually communicate, affects the things that we build. If we’re collaborative, it’s easiest to build collaborative products and collaborative tools. If we’re competitive it’s harder to see the advantage of building something that others can use without asking, where others can participate without coming to us first, where we aren’t the gateway. We’re not the only source. Where we empower people that we’ve never actually met before." But that becomes incredibly powerful.

I’m gonna go back to Mel Conway. The key thesis of Mel’s understanding about the way committees and groups worked 50 years ago is this notion that organizations that design systems, design systems that are a copy of the way they communicate. The way we treat each other, the way we create teams, the way we interact with the community is all going to affect what we create. What we create is going to affect that community. It turns out the group as a whole is more important than any one individual. Groups are incredibly powerful. Groups can embody a great deal of knowledge. But that group gets its characteristics from each person. It turns out if I change the team, if I have a team that works on the user product and I change the team, I put someone else on the team, that’s a different team. That’s not just another resource plugged in. That’s a different team and that’s gonna affect the product itself. That reminds me of this idea of the Theseus Paradox. Theseus was this ancient mariner who was traveling the Mediterranean, traveling the seas, and he was gone for so long that he had to keep repairing his ship and would keep adding new wood to the ship and replace a plank or replace a board or replace the masts and replace the keel. By the time he got home from this long voyage, he had replaced every single part of his ship with another piece of wood. So, the question is, is that the same ship that left or is that an entirely different ship? That’s what people are like inside an organization. As you create things that sustain and enable and empower people, you have to be ready for the notion that they change the complexion of the organization in fundamental ways.

If you want the organization to be incredibly predictable, people are not very helpful because people add their own. Change is inevitable. When we build systems and interact with systems, they’re actually not static. They’re dynamic. They change and change is really, really important. There’s another place in the known world where change is actually the way things work and this is called the Valley of the Metamorphs.

The Valley of the Metamorphs is a very, very peculiar place. It’s populated with lots of abstract sculptures. Nobody knows how they got there or how they were created. They looked like they’re chiseled out of stone but actually they’re smooth to the touch. They feel almost plasticky but they’re cold. Nobody knows how they got there or what they’re made of, but the story is they could change themselves over time. They actually morph into other shapes, into other objects. Nobody knows why. Some people think it has to do with the environment or the surroundings. Maybe people interact with them in some kind of way and for that reason they end up changing shapes and they’re changing aspects of one another, but nobody’s sure.

Some people even think it’s sort of an imagination, that if you watch long enough, there’s nothing that happens. But it turns out they change ever so imperceptibly over a long period of time until eventually if you sort of check over history, you realize, "I didn’t understand but it actually is changing. There actually are differences." Nobody knows how this process works. It seems like in the valley change is constant. Not only is it inevitable but it’s constant. That leads us to the third and final pillar of Elkington’s story, the planet.

Again, to quote Elkington, he talks about this notion where companies endeavor to benefit the natural order as much as possible to minimize their effect on the environment, especially their negative effect on the environment. That reminds us, again, about Gro Harlem Brundtland and what her commission talked about. The environment and development are the same thing. They’re inseparable. There’s an old joke that there was a spill in the ocean and it was polluting things. Somebody said, "Well, what are you doing about it?" Said, "Well, we’re gonna tow it out of the environment and tow it somewhere else." Right, so then it won’t be harming things. Well, of course you can’t do that, right? The environment is everywhere. We are the environment, and we are responsible for everything that we build and we do over time.

In fact, the great Law of the Iroquois, the American Iroquois had this notion of seventh generation foretelling [SP]. I have to look seven generations into the future as to how what I’m gonna do today will affect that future. I’m focused on the future even just as much as I am today and this takes us back to the original notion about how we might do that.

That’s a real challenge in the software world, in the IT world. It turns out that change is not just a discrete jump from one thing to another. It’s often very, very incremental. These incremental changes are why software breaks, why things don’t work the way we thought of. We call them bugs, right? Because I didn’t expect that to happen when these things interacted. As we build larger and larger software, this becomes more and more common because there are more and more things that we can’t predict.

Just recently, within the last few years, we’ve talked a lot about how we can deal with change and how we can deal with the way things change. Sam Gibson from Thought Works wrote this great piece called "Monoliths Are Bad," right, because now we have a name for them. That thing we used to do is bad, but this is gonna be good, right? And he talked about this idea of making change happen faster. I actually kind of prefer the notion of making change happen more safely [SP], all right, because change happens all the time. We are our own Theseus Paradox, right? All the cells in our body change over time. Are we the same person we were five years ago? Last year, when I was here, I had a beard. That’s all gone. Am I the same person?

The planet changes and everything else changes as well. Inside the corporate ecosystem things change. One of the more challenging aspects of Elkington and this notion of sustainability is thinking about this cooperation between suppliers and customers, stakeholders and even competitors. Amazon talks about this notion of actually hosting the competitors on their platform as a way to create a closer connection to all things.

This idea of stakeholders and connections leads us to the last place that I want to visit on this map today and that’s a place called Mount Roy, this peak that sort of sticks out of the valley. This is an amazing place. This is an incredibly difficult craggy peak that everyone wants to make a pilgrimage to. There’s supposed to be this giant Maori man sitting at the top of this mountain, this all-knowing, all-seeing man that has the answers to all questions. People sort of revere the notion that this man exists, so they constantly make this trek up to the mountains. But it turns out it’s very arduous, it’s very difficult. In fact, several people never make it to the top. They actually just sort of stop along the way, disgruntled and disappointed and set up shop and sort of wailing passersby and say, "No, no. there’s no one at the top. There’s 'actually just me." And they sort of create their own little communities. It turns out getting up there is really difficult.

There are stories that he’s a smart man, a brilliant man sometimes, an angry man, sometimes even a lonely man working alone at the top. What he’s working on is this giant longboat where everyone can ride inside and have all sorts of symbols, all of the symbols from all of the cities in the known world as well as other cryptic elements. Then someday he’s gonna come down from the top of this mountain and tell everyone about what this longboat means and what this means for the rest of the community. He has this idea where he can see all things. But, again, some people think he doesn’t exist at all.

What does this mean? What does this tell us about sustainability? How can we turn our notions of sustainability into something meaningful for us today? How can we take these values from the physical world and bring them into the virtual world? There’s this other quote that I really love from Dr. Donella Meadows. She and a team wrote a book called The Limits to Growth in the 1970s. This was 15 years before the Brundtland Commission where she explained this notion of, there’s only so much we can keep growing to. I love her line, "Running the same system harder or faster will not change the pattern as long as you don’t revise the structure." We have to change ourselves. We have to change our point of view. We have to change the way we build software, we have to think about connecting to the communities. This is what we have to do. Let’s go through these three steps again and think about them from our own standpoint, from the IT standpoint. If you’re a company, an IT company, you need to focus on value, not just profit. Profit is short term, value is long-term. That can be incredibly challenging.

If you’re an NGO or a nonprofit company, you have to make sure that you align the IT with your own company values. Are you using software that actually doesn’t promote your values? Are you using software just because? Is it really making a difference in your community? And as users, we have to question the value of what we give away, all our personal data, all our behavioral information, for what we get in return. Are we actually getting something important?

In the case of people, in the second leg, companies need to empower people and enable communities, enable groups, learn how to make their own decisions, provide them principles. NGOs need to make sure that IT serves and not demands. Are we doing it just because that’s the way the computer does it or do we have software that actually really allows us to meet our goals in our community?

For consumers, we need to be aware of how our labor-saving devices actually work. Who’s building these devices? What conditions are they building them under? When I have this cool little feature where it automatically does something for me, is there some human, some mechanical turk behind the scenes actually solving this problem for almost nothing just for my convenience? We have a responsibility there as well.

Planetary level means that we have to leverage change and not fight it. Change is a feature, not a bug. We need to build change into what we do. We need to take responsibility for the environment we create and we have to own it. Finally, we have to take note about how the tech that we use affects people around us. Are there people right next to me that have no access to this information? How much privilege do I actually exercise? How much does what I’m doing actually impact the environment around them? Not just physically but also virtually in terms of sharing ideas and allowing people access and clear and open debate and the opportunity to speak their mind. Am I enabling that or am I actually making it more difficult? Meeting the needs without compromising the future, that’s the kind of software that we can sustain over time.

I just want to mention a couple more things and then we’ll be done. Chamath Palihapitiya has this idea. He used to work at Facebook, and he has this notion he said recently that, "Social media’s eroding the core foundations of how we behave by and between people." Social media, it turns out may not be sustainable. It may be profitable. Maybe people are making a great deal of money on what we’re sharing, but it may not sustain healthy communities.

In fact, this ability to sustain healthy communities is starting to drive a wedge into Silicon Valley. There’s a sort of a divide between what’s healthy and what’s not. Larry Fink, who he runs the BlackRock Investment Group, $6.3 trillion worth of wealth, I’m not quite sure, some crazy number, and he was just at Davos recently and he made this statement: "Society is demanding that companies serve a social purpose." This takes us back to the way the Brundtland Commission was talking 30 years ago today. Serving a social purpose becomes really, really important. A software we build, a software that we share, is really a social purpose. When investors start talking about this, this is probably important. We’ve already seen today there are great things we can do right here that are right here, in the building. Program obsolescence, giving people an opportunity to express their views safely, connecting people with technology, reducing the amount of waste and the whole notion of sharing all together, these are all examples right here in France that you can get involved with today.

There are lots and lots of other examples around the globe as well. We’re incredibly lucky. We have this opportunity to effect so much change. It can be a tough journey. It can be tough to try to make that climb. It seems like it’s so much easier to take the short-term profit, the short-term benefit, ignore the people that don’t have access, not taking on Barack Obama’s example and think that everybody always takes everything to heart.

There are lots and lots of challenges but there are also lots and lots of opportunities to improve things. This can be a great start in the software community to think about sustainability the way we think about it in the physical world, this notion of profit, people, and planet. Hopefully, as we learn about all the places around us, all the opportunity we have, all the places we can go, this gives us more and more opportunity to create sustainable software, to me, the challenge of those seven generations that the Iroquois talk about. And that’s what I have. I hope I get a chance to talk with you more this week. Thank you very much. Thanks.