Las Vegas 2018

Applying Dickens to DevOps

https://www.linkedin.com/pulse/from-dickens-devops-carmen-deardo/ "It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us…" Charles Dickens introduction to A Tale of Two Cities.


Transformation journeys are not a linear path to success. At times I have used the above phrase to describe the spectrum of feelings present during a DevOps journey. There are incredible things that are happening by some of the leading, innovative teams. Things like reducing the time to run a test suite from hours to minutes. Or going from deploying once every 2-3 months to deploying 2-3 times a week or daily. Or being able to reduce the lead time of a story from creation to deployment from months to days.


But at the same time, there remain challenges such as over governance of change control requiring more manual escalations, teams with too much WIP being interrupted by countless project meetings, and local optimizations being implemented which don’t improve our ability to deliver business value quickly.


This talk will focus on the ebbs and flows and highs/lows of that journey using Dickens' Tale of Two Cities as a backdrop. It will provide a realistic message that is also hopeful and positive, of what one may expect to encounter and how leaders of these transformations can persevere.


DevOps Transformation leader with expertise in the implementation of DevOps, Lean and Agile practices to improve software delivery across all Technologies. Recently worked on the implementation of DevOps practices to accelerate delivery at a Fortune 100 company. This includes providing a continuous delivery pipeline to give Business areas the ability to deliver as fast as they need to based on their own determination of cost, risk and value. My goal is to consult with other companies to help them determine how they can drive accelerated results across their delivery value stream.


Named as one of the top 100 DevOps Leaders ( http://techbeacon.com/100-devops-leaders-enthusiasts-experts-you-should-follow-today) and one of the 25 must follow enterprise DevOps leaders (https://techbeacon.com/25-must-follow-enterprise-devops-leaders-twitter). Regularly consults with large enterprises on how to start and sustain DevOps transformations.


Co-authored "DevOps Case Studies" (https://itrevolution.com/book/devops-case-studies/) and "Expanding Pockets of Greatness" (https://itrevolution.com/book/expanding-pockets-greatness/) published by IT Revolution.


Presented at all 4 Gene Kim's DevOps Enterprise Summits 2014-2017 (see videos below) and at the IBM Innovate conferences (2012-2014), InterConnect conferences (2015-2017) and others (ITSM Fusion, DevOpsDays) on the application of DevOps to improve software development across all Technology spectrums. Contributed to white papers and webcasts for Global Rational User Community and was the 2014 GRUC Platinum Award winner. Was the keynote presenter at the initial DevOpsDays Ohio in 2015.


Previous experience at Bell Labs with a focus on high performance, highly available telecommunications and intelligent network systems and Behavorial Healthcare experience in providing billing and clincial systems at CMHC/NetSmart.


Holder of 3 patents in software engineering.

CD

Carmen DeArdo

Senior VSM Strategist, Tasktop

Transcript

00:00:04

Hi, everyone. Um, this is not the talk that was on your program, so if you're surprised about that, you're probably not as surprised as I am, but, uh, apparently, um, I'm a fill in. So hopefully you know it, you get what you came for, right when it's a fill in. So anyway, thank you very much. Um, that's me. I'm senior VSM strategist. Don't ask me what that means, a task talk, but you heard Mick Kirsten's talk and, um, you know, we're task talk's really focused on, um, helping his talk come to life. So if you are interested in that, let us know. So Dickens, right? What tale of two cities? What does this have to do with DevOps? Um, as I was going through a journey as a leader, right, for whatever reason, maybe 'cause my daughter is a big Dickens fan, this would come up to me, you know, as it's the best of times, it's the worst of times, right?

00:01:07

It's the age of wisdom. It's the age of foolishness, because that's what it felt like. And I'm gonna take you through some of those ebbs and flows in this presentation. So first I'm gonna start with, well, you know, what diggins know about DevOps? What do you know about BMWs, right? Because sometimes I think se as Seinfeld has said, or maybe it was Ray Barone material just presents itself. So here's a few days ago, and that's me. If you can't see it sitting in the BMW, right? Everything's great, right? How can it be better than this? And then a funny thing happens last night when we're trying to get the car out of the building. It depends on a, has a single point of failure, which in this case is a freight elevator. In the freight elevator is a hundred dollars chain probably, and it breaks.

00:01:59

And here we have the most advanced car on the planet stuck next to the fruit loops, right? <laugh>, we cannot get it down the fruit elevator. Probably not quite the best of times, but that's okay. It turns out okay in the end. So as you're going through a journey, right, it's kind of, look how wise we are, right? Incredible things are happening by our innovative teams. Um, you know, we're reducing times from a test week, from hours to minutes. We're going from deploying monthly to perhaps daily, and we can reduce the flow time of a feature for months to day. So it's like, you know, yay, we're great. But look how foolish we are.

00:02:44

These challenges remain. We still have over governance of change control requiring more manual escalations, countless meetings, right? Dominican de Grandes talks about this all the time, estimating. And if that's not good enough, let's re-estimate <laugh>, you know, re infrastructure changes that are taking a long time because of manual processes, right? Or, or teams just feeling like they're not in control, right? They want to make A-D-B-A-D database change. It goes to centralized team of DBAs. They have way too much whip that they can manage. And so it goes on and on and the team is stalled because of this.

00:03:27

So what are the specific challenges that you're gonna have to deal with? So one is local optimizations, right? Teams, people have great ideas. A lot of teams, you know, it could be a tool, could be a process, but if everybody just optimizes within their own silo, you're really not gonna achieve the first way of flow, right? You're not, you're not applying systems thinking, oh, I guess it was on this slide. This is what happens. We never practiced this before. Lack of flow first way, right? You know, teams are not really being empowered, right? I don't know if you saw John Smart's talk yesterday, but he, he talked about, you know, it's like you're empowered, right? You can choose A or B, but yes, you're empowered, right? I mean, teams really don't buy in and feel like they're empowered.

00:04:22

You know, there's, there's this view that, uh, we heard this week about the cost center, you know, model and the project center model. I mean, if it is just looked at as a cost, then there's gonna be pressure to reduce it. If it is looked at as a profit, then you wanna invest more in it. The number one killer is cynicism, right? You know, we're gonna change. People say, oh, well I've heard this before. It's the flavor of the month, right? I, I don't, you're telling me, I, you know, I'm not gonna do that again. I won't get fooled again, right? And you can see it in their face. You don't need the metric. You can just sense it, it's in the air. And of course, the hero culture, Brent, from the Phoenix project, and also one time results, which really aren't sustainable, right? So you, so somebody does something good one time and you try to react to that. It's a great thing, but you really don't, you really haven't put in place a system to capture that success and repeat it. It's not repeatable, it's not sustainable. Or it could just be a happenstance, right? As Deming would call a red bead. That really is, is not something that's gonna be systemic.

00:05:37

And of course, we all know that there's cultural impacts, right? Addressing cynicism is one of those. So, and I would come in and I worked at Bell Labs for a while, and then I went to a Fortune 100 company, and I would, I learned very quickly to stop saying, well, we used to do that at Bell Labs because nobody wanted to hear that. So, so I would get, well, that won't work. And I wanted to say, why do you pick up your phone? You get dial tone. I think this probably would work, but that won't work. Okay, so how did I address this? Well, this is where the community that Gina has formed has just been invaluable, right? Because obviously me saying something isn't gonna work. But if I bring in a video by Topo Paul, Paul at Capital One, you know, who's like, I think they're pretty highly regulated.

00:06:30

I think they have like a bank or something, then it's like, and they aren't Netflix, right? They aren't antsy. But then the next one was, yeah, but that won't work here, right? Because there's something here that's different. Now, I showed this deck like 10 minutes. There's things that sound good. It's kinda like when you write code late at night and you wake up in the morning like, who wrote that <laugh>? This sounded really good at 6:00 AM And then I showed it to somebody 10 minutes before the talk and they looked at me. So I'm taking some risks from, from 1980 movie references, but Tracy's gonna laugh anyway. So when I, I would hear this, if you, if you've ever seen Mr. Mom, and there's the scene where he is talking about how long it takes to make instant grits, right? And it's like, you know, he's saying it only took two minute, you know, two minutes.

00:07:21

And he goes, doesn't it take instant grits? Well, no instant grits, no southerner is gonna make instant grits. Well, grits takes 12 minutes. So then he says, so will perhaps the law of physics cease to exist on your stove? Right? So that's what I wanted to say to people is, so it won't work here. Is there something different? Does Java run differently in this building than it does at Capital One? I mean, what, what are we saying here? Right? So, I mean, you're gonna get that, right? And, and I'll tell you potentially how to deal with that. And then there's cultural impacts, right? Sometimes we would come up with ideas and, and they would be dismissed. And my team would think, feel like, well, that was a bad idea. And I'd say, no, that's not a bad idea. That's a great idea, but it's not ripe.

00:08:09

The environment is not ready for it yet. So, so if you have things that you think really will work, but they're not gonna work now, don't throw 'em away. Keep them in your backlog. Because at some point, when people are ready, when the culture is ready, then those ideas can be parted. The play, I already talked a little bit about that, you know, cost, cost. How do you get the C levels focused away from cost? But then another talk I would have many times was we'd be discussing an idea and someone would say, well, yeah, we should be able to do better, right? We should be able to deploy faster, we should be able to do this more quickly. And I'd say, yes, yes, it's not perfect, but don't make perfect be the enemy of good or better, right? We, it's gonna be baby steps. So we have to continue to just progress in those incremental improvements. And you have to have a safety culture. We've heard a lot about that. We continue to hear a lot about that. Teams have to be able to be truthful and they have to have cover to experiment.

00:09:16

So here's my example. I used to use another movie, I should have set this up. Oh, well, anybody ever see Mr. Mom <laugh>? I used to use this. Michael Keeton went for my high, went to my high school, and, and you know, it's like, hey, 2 20, 2 21, whatever it takes, right? We don't have to be perfect now, an electricity probably wouldn't work, but it's like, let's just get better, right? Whatever it takes to get better. So current state, um, Mark Schwartz gave a great talk in 2014. It was the first DevOps Enterprise Summit that just blew you, blew me away. And if you haven't watched it go back, it's on YouTube. It's, you know, the topic was how the DevOps will save the federal government. I'm right now who's not gonna go see that talk, right? And what, what Mark stressed in it was, whatever your current state is, you can't be too critical, right?

00:10:15

It got you to where you are, right? You're, you're probably a successful company, right? It's paid, you know, it's paying a lot of people's, uh, bills, right? There are reasons that you got there, right? People, people had good intentions when they designed current state, right? So no matter how absurd some of the processes you're executing are, they were done with good intention based on what was the situation at time. So we can't be too critical, but we also have to be truthful, right? Marco talks about this, right? What It won't keep you there, right? Yes, we may be a Fortune 500 company today, but Mick showed a slide that showed what 50% projected of the Fortune 500 companies won't be Fortune 500 companies in five to 10 years.

00:11:07

And again, there's a reason behind the current state. So it got, it really got down to this, right? Don't call my baby ugly, right? Nobody likes their baby to be called ugly, right? So yes, we have to be open to it, but if you're talking to somebody who's had an vested interest in a current state, we have to do it in a way, right? That doesn't feel like they're calling their baby ugly <laugh>. But we also can't be defensive, right? We have to balance this, this being critical with also, we can't be defensive because we have to be open to continuous improvement, right? Going back to the shoe, heart Deming cycle, whatever you call it, wherever you work, this is the way you're gonna change the culture. You're gonna go through a journey over time.

00:11:59

You have to be open to experimenting, right? Um, so, you know, we've heard a lot about that, but I really believe, and we saw a lot of the work that I did with the nationwide folks are up here. It's running the experiment, right? Limiting the risk of running that experiment to, to executives so that you can show, I think one example that works and then that becomes the example that actually works here, right? It's no longer it Etsy or it's no longer even Capital One. It's, it's Jane and Kathy and John and that team on the fifth floor, and now they can do it. Well, that makes it real to people, right?

00:12:43

And that I think, ties into the idea of quality circles and what Deming talked about is, I mean, a lot of transformations I've seen are imposed, and John Smart did a great job talking about that yesterday. People are not gonna feel bought into things that are imposed upon them. Deming talked about the people doing the work, no mess best about how the work's done. Your teams know how to get better. If you, if you actually give them a safe environment and let them talk about it, they will give you ideas that'll blow your minds on how they can get better, right? Jared was a great example of that, how I was working with, at Nationwide, things that, the things that the leaders would've never thought about. That's not, so you can't impose this. The, the teams need to be able to suggest ideas and experiment again safely.

00:13:34

You have to give them that safe zone where they can run experiments, they have some buy-in, in that area with the business, with their leader, that they can actually run the experiment. But then what the leaders need to do is take those results and apply systems thinking. And I think this is one of the things that people really do not understand. If I was talented enough to teach a class to executives who would be on systems thinking or, or leaders in general, right? You have people that can do the work, you pay them to do the work. Once you get those results, then you need to figure out what am I gonna do with this? Right? So they may have a great, something great that worked for that team, right? Now, how do I operationalize that across the enterprise, right? Does it even make sense to do it? That's, that's what the leaders need to do.

00:14:28

And I think you need to avoid that lore of, I I hate when I hear low hanging fruit, right? Some of those things are really red beads that are just disguised at low hanging fruit. They're one time results that really aren't sustainable. And if you haven't ever know of or, or seen the videos on red, the red beat experiment, again, I would, I would, um, encourage you to do that. I think a lot of the non-value added work that we do is really some idea of a red bee experiment, right? And I put, you know, one of my favorites estimation up there, right? But how many things are we doing that when you actually look at the results, they have no impact, right? What, you know, Bezos talks about, you know, measuring proxy metrics, but it's really like measuring activities, right? If we just do these activities, somehow we'll get better. Well, will we, is estimation estimating five times really making us better? We have to start thinking about that and getting, getting those things exposed because they, they are the things in our environment that aren't leading to successful results.

00:15:36

And again, I really believe out of the crisis, which is probably written in the seventies or the eighties, is a classic read on leaning continuous improvement. So I would encourage you to read that. So what are some of the patterns for success as you're trying to go through the ebbs and flows of, of the Dickens novel, right? Model lines and experiments, which is what Nationwide did pick start Smalls talking to, to someone who's leading a DevOps transformation this morning. You know, there's a lot of stuff you could do. There's a lot of tools you could do, although it's usually not around the tools. Whatever it is, start small. Just get, just get one or two areas who are enthusiastic, who have support, who you can provide that safe, safe experience for, and let them start to do the experiments. Don't fall in love with a process or technology. Fall in love with getting better. It's, that is what's gonna get you going. It's not a tool, it's not the shiny object, it's the mindset of continuous improvement and getting better.

00:16:46

And, and in order to make that transition to where you are, to where you want to go, sometimes you can draw the picture of where you are. And sometimes you can draw the picture of where you want to go, right? So in the BMW's case, are the cars on the showroom and the car's here and where we want it is back in the dealer shop. How you get there though may not be a straight line, right? So you have to do things like strangler patterns, right? If you want to get rid of a tool or you want to get rid of a, a process, you probably just aren't gonna say on Friday, on Monday we're gonna start doing this new thing. How do you strangle that out of your environment?

00:17:30

And also, I would call it building scaffolding, right? There were things that we did that I knew I did really didn't want to do <laugh>, and, and I didn't want, I wanted it to last only as long as it needed to, but it was the only way to build that temporary bridge to the next step. Now, you know, one of the catches is that if you call it scaffolding, probably you're not gonna get much help building scaffolding, right? So you have to be able to sell it as, as more as a journey. But you also have to be willing to say, I mean, we would talk to people about the fact that we had done something and now, you know, and they would say, well, wait a second, you're telling me in six months or nine months we're gonna do something different? Yeah. Well then why would I wanna do this in the first place?

00:18:20

Well, you know, it's a step on the journey, right? It's, it's a necessary part of where we have to go. So building that scaffolding is necessary because the gap between current state and future state is gonna be so big and so overwhelming that unless you have that scaffolding and can adopt those strangler patterns, it's gonna be very hard to get there. And then I think maybe the most important thing is, is, you know, having a centralized team of experts take credit for results is not going to change the culture. It's, it's that team in the corner. It's, it's the Katie, the Peggy, the Jared, the Fred Ed team, right? They need to tell the stories. Nobody wanted to hear me talk. You guys are tired already, and I still have 10 minutes to go, right? <laugh> me saying something to a bunch of executives, it's like, Carmen says that all the time, right?

00:19:20

Here comes another Capital One video. They don't want to hear that. When the executives came and heard those stories, I never saw, I would just sit there and watch and you could see their face, they could see the excitement of the team that sold them, that sold 'em better than any PowerPoint deck or set of metrics. They could feel the enthusiasm sometimes I to beg the teams to tell their stories. Like, please tell your story. Okay? That's what's gonna change. Because when a team sees that they want their story, they, well, if they have a story, why can't I have my story? And then it starts to become infectious.

00:19:56

And again, you have to look at what you've already accomplished, right? I, at times, I, I know on that journey from current to future state, it can just seem like we've been doing this for six months, a year, a year and a half, two years. Why aren't we further up the mountain? But then you have to pause and reflect and say, well, look how far we've come, right? So let's celebrate how far we've come, not let's focus on it seems like there's a mountain we still have to get to. And then another thing is you have to be truthful, especially with the business, right? When we did these experiments, we, we showed them, we made, you know, Dominica de Grand, this make work visible, this work showed up on, on their K bond with all the other work they were doing, right? It was on four week iterations instead of two, had the show and tells it wasn't hidden. Now sure, you had to manage it, you had to have whip limits. You know, we, we didn't wanna put critical business deliveries exist, but the team was so engaged and so excited. And I'm not saying excited, like they're working 50, 60, 80 hours a week. That's not what you want. But they were so excited and bought in and there were no business deliverables that were missed during this period of time.

00:21:15

And again, systems thinking. So if I look back on this, the biggest reward personally that I got was not any metric that we reduced, you know, the, the flow time from 40 days to 20 days, or we implemented docker or, or this team could deploy more frequent frequently, or even, you know, getting promoted even though I didn't get promoted. Um, it, that wasn't the biggest reward. The biggest reward is when you could look in those people's eyes before, right? And you saw despair and you saw hope. When you see people start to go from cynics to being hopeful, then you know that you got something good going right? Then you know that you're on the right path. Because now those people have changed their mindset. And if you can get people who are maybe the biggest cynics to become some of the biggest stories or storytellers, well then you can relax a little bit and, and let them go out and tell their stories. Because now you know, you're on the right path.

00:22:32

So if you look at Dickens' hero's journey, right? What, what did he go through? His motives changed at one point, right? He was not a good guy at the beginning of the novel. And, and he became a very heroic figure, right? He faced the guillotine well, okay? Um, you know, I don't, I'm, sometimes there are things that feel like the guillotine, but I mean, he faced the guillotine and selflessness became his key leader characteristic as he was going through this journey. And I used to say this a lot, right? Patience and perseverance, right? You're gonna have to be patient, you're gonna have to persevere. You have to, you have to count those little victories to just keep going and keep going and making those incremental improvements. You know, you need a level of courage. Gene talks about the courage. Um, you know, I've heard talks about some of the great who's, I'm gonna ask to make a question.

00:23:36

Sorry, I, I forget her name. The target, the original, um, Heather Mickman. Heather Mickman, Heather Mickman. I have so much admiration for her. She, I mean, she didn't face the guillotine, but man, she put it out on the line. She had operations people, executives coming up to her and saying, I hope you fail. We hope you fail. Okay? And she persevered with her team, tremendous courage to make these transformations. But again, I wish you a better result than the guillotine, right? That's not what we want. So, so coming back to the BMW story, right? But she got promoted. Yes, she did get promoted. I should let, I should give what? To me, Heather did get promoted and she did go on. That's right. That's the rest of the story. As, as Paul Harvey would say, she did get promoted, right? Her courage paid off. It paid off both for, um, the company and it paid off for her.

00:24:37

But, you know, there was tremendous courage that was displayed there. Courtney Kissler is another one who's displayed tremendous courage, right? Because she's gone into situations where her message maybe wasn't so well received. And now you see what she's amazing things she's doing at Nike. So, so the, you know, people like that are role models to me because you know, the things that Topos do because, you know, they show how this can be done at large organizations if you have that patient's perseverance and courage. So getting back to the BMW, I think this is the next slide. I don't have my, and I forgot what I put <laugh>. Um, so, you know, here we are. Last night, we're at the freight air elevator, you know, the freight elevator and what are we gonna do, right? And laurel's here, she's, she is doing all our, um, coordination.

00:25:29

You know, it's like I saw there was a roadside assistance. We open up the going door, there's a roadside assistance, say, well, maybe we should just call that 800 number, right? May, I don't know, <laugh>, but you know, we took it in stride, right? People weren't panicking, you know, people weren't worried about liability except maybe the hotel folks. Um, you know, Laura was keeping it all together, kneeling everybody, you know, Renee's there having a good time. And the kicker, Renee goes over, he goes, that elevator's German. So <laugh>, I don't know what to say, Renee. I mean, yeah, German card, German elevator. Who knows where that chain was made though? I don't know. So, you know, let the journey begin, right? Again, start small, empower teams, experiment, show and tell with executives, celebrate success, right? Your teams are able to do this. You can do this. Your teams want to do this. Celebrate those successes and then, you know, continue to scale and sustain it up the mountain. I need lots of help, um, as you can tell. But, you know, share your experiences, right?

00:26:43

I'm sure you have. That's what this makes. This conference. Great. I cannot tell, I was telling Gene this, I saw him for like a millisecond because it's amazing all he does in this, this week how much this has benefited me. Other people having this network. Because when you're doing this, it can be lonely, right? You may be one or two people at your company trying to do this. You may think you're that lonely voice in the wilderness, right? This is an, and I think maybe Jeff Galloway said that, right? This is a place to reenergize. Make your contacts here, but don't just take that business card and never call that person again in talk, talk during the year. Share stories, encourage each other. I mean, without this community really, I, there were days I don't know that I could have gotten through because it gets hard, but this community is just such a blessing. So that's what I'm encourage y'all to do. So thank you very much.