At the recent Agile Singapore conference Richard Sheridan and Linda Rising presented two of the keynote talks.
Afterwards they sat down with InfoQ and discussed what it means to have an agile mindset and what it takes to design an organisation from scratch which has a culture of joy in work.
They talked about Richard's book Joy, Inc and how Menlo Innovations was built to bring joy to every aspect of the workplace.
A sample chapter from Joy, Inc can be found here.
The first part of this interview can be found here.
InfoQ: Richard, can we talk about your technical practices? You spoke about joy in the product. How do you make sure it's a good product?
Richard: I think what undergirds any culture is a shared belief system. When you share a belief system as strongly as we do, the team reinforces that belief system within themselves. Now, I will suggest this may be one of the most powerful side effects of pairing and pairing in an open and collaborative environment because a lot of people ask us, "How do you make sure this culture perpetuates this stuff? Rich, you're not there anymore. You're not the guy. You can't spend time with everybody on the team like you used to at the beginning."
I don't need to: When you join the team, within minutes, never more than single digits of minutes, you're paired with someone. You're never alone. You sit down at the keyboard, the pair that you've been assigned to passes the keyboard over and puts the keyboard under your hands. He says, "Let's get started." Of course, at that point, the new person says, "Well, what language are we doing this in?" Your partner says, "Well, it's in java." They might say, "Well, I'd never done java before” because we don't hire for that kind of skillset. We want a computer skillset, the languages we can teach.
Somebody will say, "Yes. Okay. I'll tell you what to do. Double click on that icon." At first it's like, "Oh, I can do that." Click. Then they'll say, "Okay. We're going to start programming an automated unit test because that's where we stopped." Somebody said, "Well, I've heard about those. I've read about those but I've never actually done one." "Great. Okay. We've got some template code here. We just start like this," and the person starts typing.
Of course, they're making mistakes left and right but their pair partner says, "Oh, don't do it like that. Try this. Okay?" After two or three hours of this, the new person is starting to say, "You know what, you don't need to tell me every keystroke anymore. I'm starting to get it."
We do that for five days and then we switch the pairs. So you're not just getting it from a single person. Now, you're moving around the room systematically through people.
What's fascinating is, I just saw this happen the other day, it blows me every time it happens because I can tell you there's stuff that happens at Menlo that I should be used to by now but every time I see it, it blows me away: There was somebody at Menlo who'd only been there for four weeks and they were already the mentor to someone new joining the team, after only four weeks. The reason that works isn't because they'd become an expert in everything Menlo. It's because they now have five days of relationship with four other people on the team who are within eyeshot and earshot; they’re probably sitting shoulder to shoulder with them.
If I needed to talk to Linda and she was one of my mentors, I just say, "Hey, Linda," and she'd say, "Hey, Rich!" and we're in a conversation. I can say to her, "Hey, Linda, I forgot how to do this” because it's safe to do that. She'd say, "Oh, Rich, remember we do it like this." I say, "Oh, yes. Thanks." Then we move back.
Now, the most important lesson in that is for the new person to say, "Hey, wait a minute, you just admitted to a senior member of the team that you forgot something and and she didn't seem to be bothered by that and it was something she just taught you last week and she answered in a very caring, compassionate way, 'Oh, remember we do it like this?'" Right there in that moment, you're teaching a very important cultural outcome.
Now, usually, around this time, the person who just joined will turn to their pair partner and say, "So by the way, how long have you been in here?" If they say, "About four weeks," the new person is mortified, realizing, "You mean, I could be in your chair four weeks from now?" We say, "Yes. It's okay because you'll see how it works."
What you see at Menlo is layer upon layer upon layer upon layer of those kind of reinforcing mechanisms because it isn't just about putting up something on the wall that says, "Test first design." These aren't just posters on the wall. This is our shared belief system. This is how we work.
We don’t do it because somebody said at the beginning we should. We do it because it has the desired effect. One of my favorite stories to tell that unnerves people every time I do it, it's not even a story. It's a demonstration. Usually two or three or four hours into a tour, somebody will ask me, "Okay. This is interesting. I'm intrigued. What evidence do you have that this actually produces better results?"
"Okay. You've been here for three hours. Pull out your phone and dial our switchboard." So I give them the phone number for Menlo and they hear the phone ring because it's out in the middle of the room with everything else.
I ask them, "How many times have you heard that in the three hours you've been here?" They realize that is the first time they’ve heard the phone ring. I say, "Exactly. Our phone does not ring for emergencies."
All of a sudden, their eyes go wide and they say, "We're taking 300 calls a day because the problems in our software." I said, "We’ve had less than 30 calls in our history. The last time our team remembers a software emergency was 2004." Our visitors can't even imagine that is possible.
It's not that we're bug-free. We're talking about emergencies. So it's -- for the younger people on the team, the people who've joined us right out of college, I think they find those stories entertaining. They don't know what I'm really talking about but they see it in the eyes of the visitors.
So they understand something about that story is really, really important to the people who are visiting. For the old timers on the team, they know exactly what I'm talking about. That's when I talk about chaos.
This is all born of the Deming philosophy when he said, "The only thing anybody ever asked for is a chance to work with pride."
All of these tours hold me accountable, too as the team hears my stories as I am leading the tours. They see and hear my passion for our shared belief system. Several of them lead tours, too as I can no longer lead all of them. All of this transparently reinforces our practices.
So there's an accountability just from the tours alone. It's kind of fascinating.
Linda: That would keep you transparent, all right. People marching through the door every day and walking around.
InfoQ: Please can you tell us more about your High-tech anthropologists?
Richard: We want to make the computers to think like humans not make the humans think like computers. That's where our High-Tech Anthropology® comes in.
It's a totally different practice. Most teams gather what are traditionally called “requirements.” Our High-Tech Anthropologists® go out and they do observation and discovery and study people in their native environments, see the invisible things that no one would ever think to ask about because quite frankly, no one knows what software needs to do. We have to discover it.
I think that's where a lot of people go wrong is they think this is the way a software system should be designed, how it should work, is knowable. It's not. It is discoverable and can be -- I think the greatest benefit of an iterative and incremental approach is as we think we've discovered things, we can test our assumptions much like an experiment. We look at our five-day cycle as an ongoing set of experiments approaching the goal.
Linda: There you go. Experiments.
InfoQ: I could use the stories and a hypothesis to prove or disprove it.
Richard: Yes.
InfoQ: What makes somebody right for that role? What do you look for in a person or what do you help them to attain?
Richard: Well, there are many qualities to a good high-tech anthropologist and we have to discover that in them too. We pair anthropologists as well. So there's no chance that we're going to bring in of somebody who is a bad fit for that and not sort that out for very quickly because what you need more than anything, at the top of the list, is empathy and compassion -- they take the “end human suffering” part of our mission quite literally.
They have to be very patient observers. They go into everything with a beginner's mind. We're not domain experts in anything and if you are a domain expert, we actually don't assign you to a project in that domain because we want you to see the invisible.
So the anthropologists have this incessant curiosity about things. They're social in the sense that they know they have to build a relationship with the person they're observing, otherwise, the observed person will behave differently. There has to be trust.
So they understand that sometimes they have to bring food, start over lunch, get to know people, get them to tell their stories, be interested in people. A big part of this is all about sense-making because we always start in the cloud of ambiguity at the beginning.
So you have to start sorting through all these discreet perspectives -- that don't seem related or correlated at all and you have to seek patterns. So they're pattern matchers as well. Then ultimately, start to take this from the cloud of ambiguity and turn it into pixel perfect screen designs. They use a lot of paper prototyping.
The anthropologists definitely embrace the make mistakes faster philosophy. Take paper designs. They literally take hand-drawn paper designs back out into the world and show people what they've done and ask them not what do they think, "Do you like this? Will this work for you?" but rather, "How would you use this to do a particular thing we've learned to do?"
At the point where they're starting to show off their designs, the early designs in particular, but this goes throughout the entire process, they learn number one, to bite their tongues and be quiet and watch. They must come prepared for needing a great amount of humility because their brilliant design is failing right in front of their very eyes.
Rather than say, "Oh, it's right here. It's so obvious."
Linda: "Why can't you see?"
Richard: “Why can't you see that button right there?” They've learned to ask questions like, "What were you expecting?" Then "What do you think would happen next? What would have helped?" There's going to be more understanding.
Oh, I remember when we were doing a diesel motor diagnostic tool, a button to begin the testing process was called "start." The anthropologist showed this motor mechanic, the screen, and watched and said, "So how would you do an initial diagnostic?" There was this big button there called Start. The guy was holding it in his hands and he's staring at the screen and looking around and looking around.
Finally, the anthropologist said, "Well, what were you looking for?" He says, well, he's looking for a button called "Test." They said, "Oh, we call it Start." He shakes his head and he says, "You can't call it Start." They ask, "Why not?" He says, "Because --"
Linda: Start might start the engine.
Richard: He says, "That's the last button that I would ever push because I don't know what tools I left behind. I don't know if my colleague's got his head banging about by the fan or something." He says, "I would never push the Start button," at that point, they pull it back. They scratch out the word Start. They put on a little post-it note. They write the word, "Test," they hand it back.
"So what would you do?" He says, "Yup. I'd press Test."
We watched were surprised to see he pressed it with his thumb. Our customer thought that they use their index finger like a stylus. We didn't tell him how to use it. We just watched and observed. Then we noticed, these guys have huge thumbs.
So we took them him to the photocopier, took pictures of their hands, and cut them out and made really big thumb-sized buttons. All of these things are part of High-Tech Anthropology®. Quite frankly, it is every bit as creative an activity as coding and probably more so and some of this, because there are so many different approaches they have to figure out as to, "How are we going to gather the data?"
Linda: Initially not paired?
Richard: Correct. Yes. We didn't pair in the earliest days of Menlo and we didn't necessarily story-card the way we do now. When we started pairing, they all bristled and said, "Oh, you can't do this in pairs. That isn't the type of work we do. We need our think tank."
Linda: So why did you change your mind about carrying that role?
Richard: I think this is typical of Menlo where we said, "Yes. We know. Let's run the experiment."
Linda: But something must have turned. I mean like --
Richard: I'm sure -- typically, what would trigger an experiment like that, like what we think has been long enough, I have to really think hard as to what caused the first pairing experiment. But it was probably on-boarding of new people where we said, "Hey, we've got this relatively easy construct for onboarding new programmers. We don't have the same thing for onboarding high-tech anthropologists. Let's try the pairing thing."
Linda: Are there any other roles that are not paired?
Richard: So the project manager role is not strongly paired but they are running a series of experiments now on this. They do sit right next to one another. Our project management system is so routine. It's so standardized that they can easily pick up another person's project when one of the project managers goes on vacation.
We proved that beyond a shadow of a doubt when Lisa took a seven-week vacation, went to New Zealand, and Australia and Hong Kong. Emily took over for her on a project and it was pretty seamless. Of course, one of the reasons we know it was seamless is because one of our rules is you cannot check email when you're on vacation.
So Lisa was off the grid for seven weeks. They paired ahead of time to get ready for the transition. I know I've heard them just recently talking about this. "We need to do more of this pairing thing for project managers." So it is the least strongly paired role on the team.
Linda: Can you think of any others that are not paired?
Richard: Well, I think she's looking at me.
Linda: Yes, I am.
Richard: And saying, "What about your COO?" Yes, okay. I will say -- I would say I'm here alone but often people do go on trips with me. It's a little more expensive to take another person to Singapore --
Linda: I would say that you're on the road, who's back at the farm?
Richard: Well, James Goebel and I are interchangeable partners.
Linda: So you are paired?
Richard: You might as well call us co-CEOs the way we run the business, there's no question there. It extends beyond that though because one our themes around pairing is to solve the tower of knowledge problems. I had become a tower of knowledge when leading tours.
They said, "We should have the team participate. How would we do that?" Someone says, "I know. How about if we pair people with you on the tours and you learn over time to shut up just a little," they probably said it nicer than that but I can hear it in their words. "So give them a chance to talk." Quite frankly, I was delighted because there were two things I heard when others were with me on the tour.
One is, they would tell my stories in their voice, in their recollection, and in their perspective, which was fun to listen to because some of my stories are actually their stories.
So they would tell their version of the story and I'd be listening like, "Oh, this is fascinating." Then, of course, far more fascinating than this, they would start to tell their stories, stories that I have no knowledge of whatsoever because I'm not in the team. I'm not amongst them every minute or every day. I'm not right there seeing how things are happening.
I would start to learn more and more about Menlo by being on these tours with them. Now, it's funny, it's a little unnerving for me to watch that part of Menlo be fairly autonomous. I mean, I have to be gone a week, there's ten tours of Menlo this week alone. I'm not there for any of them and nobody's calling me up saying, "Hey, Rich. What are we supposed to say in the tour?" It's all embedded in the lore of the company.
Linda: So what if you hired somebody who was allergic to dogs?
Richard: That’s an interesting question. The basic rule around that is you are responsible for your dog. If your dog is causing problems in the space then your dog loses and the dog has to go home. There are some people who are afraid of dogs, or they don't know how to respond to dogs and, of course, with all the tour groups, they have to be careful around this too. We have -- it's gotten a little weird in Menlo at times beyond even maybe what I'm totally comfortable with but in the spirit of running the experiment, I don't step in right away but one young man had pet rats and he brought them in and they would sit on his shoulder while he was coding and it was a little unnerving for a few members of the team but that was just his thing and he's trying to be a little eccentric. Well, one day, as visitor came in and he's walking through the space and he starts kind of scratching and itching and he's kind of getting a little stuffy in the nose. He's looking at his -- he's looking around and he says, "Look, I don't mean to offend you but you have a rat problem here at Menlo." We asked, "What do you mean?"
He says, "Well, look," he says, "I'm allergic to rats."
He says, "I worked in a lab for years where there are lab rats and I just developed an allergic reaction to them," and he says, "Quite frankly," he says, "I'm having an allergic reaction as if you have rats in this space and I don't mean to offend you because I'm sure you'd not want rats in your workplace."
"Well, funny you should mention that…” So needless to say the rats had to go. So yes, people always ask us, "Have you ever had any failed experiments?" That might have been one of them right there.
Linda: I'd like to talk about animals in the workplace and there's a lot of evidence that it does help but there are people, even if they love dogs, who are allergic to them and then there are other people who don't like dogs or who are afraid of dogs --
Richard: Right. Dogs sense that and can respond inappropriately -- when they sense the fear and that sort of thing. So well, I'll just make it very clear. When tour groups come, we tell them, listen, number one, if there are dogs in, and they're usually is at least one dog in every day, we tell our guests ahead of time. If there's anybody in the tour group that has any fear of dogs or any reaction to dogs, we can just lock them away in the back office and it's okay. Don't feel like you're putting us out for asking that.
So every once in a while, we have to put the dog away. We haven't had anybody allergic to babies yet.
Linda: I love the dogs. Actually, when I talk about the dogs, I have a quote from a guy who went through this too. So I do a half-day class on this. When I talk about dogs in the workplace, he's, "Oh, that would be so great. I have dogs." Then he started working for Google and they do have dogs in the workplace.
He sent me a little email and I put that on slide. He said, "I finally get to have a dog and it's so great to have all the dogs.
Richard: What I probably love most about Menlo is this sense that we honour the whole person. I think the baby part of the story is one of the most stunning examples but there's so many others. We'll just say, "Look, you don't have two lives."
We talk about work-life balance. You only have one life. We have to honour the whole thing. If you don't feel like coming in and pretend you're this professional that has no other activities or responsibilities or connections to the rest of the world while you're at work and then when you're outside of work, you can go deal with all those things. No, they're all real. They have to be dealt with all the time.
Because a lot of people look and say, "Oh, babies in the workplace. Doesn't that undermine productivity?" I ask, "Compared to what?"
When there's a snow day at Menlo, kids just come in to the office because --
Linda: But what you do mention is nursing mothers. I know nursing mothers and that was really a hassle because you have to go periodically pump out their breast and that's a hassle. Whereas, if the baby is there and – yes. It's easy to do that.
Richard: Over time, we got better and better at supporting that. Who knew that there are even awards for this stuff but the first couple of years we did this, we got a bronze medal for supporting breastfeeding in the workplace.
Linda: Wow.
Richard: So a couple of years later, we got a silver medal. Then -- because we asked them, "So what are you looking for?" They started describing to us. We did the best job we could. We got this silver medal. Finally, when we were building out our latest space, we said, "Hey, look, we're doing a build out, what does it take to win the gold?"
They gave us all the specifications for a gold medal standard breastfeeding support workplace. We did every single one of them. Just this year, we were given the gold medal for breastfeeding support in the workplace.
InfoQ: Congratulations.
Richard: We didn't do it to get the award. We respected the agency that was coaching companies on how to do this better and said, "You guys have thought about this. We only think about it when we have to. We probably don't think about it completely enough. What do you all think?"
“You need to have a sink and refrigerator and a glider and lights that dimmed and a soundproof room for the sleeping baby," and then we put monitors in that room so if the baby is sleeping in the room, you could still hear the cries or anything like that. It wasn't that expensive. This isn't like some huge cost or expense.
InfoQ: Linda, any final thoughts from you?
Linda: I'm just so happy to be here. I mean, this is a conference where I not only got to have an engaged or, I've lost the word, hungry. Yes. I like that. Hungry participants, some of whom I know I'm going to be hearing from, some email. But now here on the last night to have dinner with you two guys and to hear lots more Menlo stories, which I'm going to tell and Sri Lanka and Mumbai as my next ports of call.
So I'm just one happy camper. Thank you, Shane.
InfoQ: Thank you. And Rich, thank you very much.
Richard: Thanks for having us.
About the Interviewees
Linda Rising is an independent consultant who lives near Nashville, Tennessee. Linda has a Ph.D. from Arizona State University in the area of object-based design metrics. Her background includes university teaching as well as work in industry in telecommunications, avionics, and strategic weapons systems. She is an internationally known presenter on topics related to agile development, patterns, retrospectives, the change process, and the connection between the latest neuroscience and software development. Linda is the author of numerous articles and has published several books: Design Patterns in Communications, The Pattern Almanac 2000, A Patterns Handbook, with co-author Mary Lynn Manns, Fearless Change: Patterns for introducing new ideas and soon to be released in March 2015 More Fearless Change. Her web site is: www.lindarising.org.
From kid programmer in 1971 to Forbes cover story in 2003, Joy, Inc. author Richard Sheridan (U-M BS Computer Science '80, MS Computer Engineering '82) has never shied from challenges, opportunities, or the limelight. While his focus has always been on technology, his passion is process, teamwork, and organizational design, with one inordinately popular goal: the Business Value of Joy! Sheridan is an avid reader and historian, and his software design and development team at Menlo Innovations didn't invent a new culture, but copied an old one ... Edison's Menlo Park New Jersey lab. Some call it agile, some call it lean … Sheridan calls it joyful. And it produces results, business and otherwise. Six Inc. magazine revenue growth awards, invitations to the White House, speaking engagements around the nation, numerous articles and culture awards, and so much interest have led to a tour a day of the Menlo Software Factory™.