 All right, what do we say we get started three minutes early here and anyone that comes in three minutes late or on time You guys will have to get them up to speed Also, if anyone knows Jorge Castado, he has his dribble badge here. You do? Yes, we have a winner You get Thank you, I think I was in here from the last session which apparently got a little crazy and some lanyards were flying around Okay, we'll welcome everyone to building your DevOps game plan Thank you especially for everyone in the front row Appreciate the love So we're gonna dig in a little bit about how you can bring about change to the organization or the team or the group that you're working in and Get into DevOps a little bit So I'll just do some quick kind of Like hand raising to to work it out a little bit So how many people here like feel like they know what DevOps is and are excited about it? Yeah, maybe a bunch of people cool Let's talk a little bit about like the organization size that we work with so who works in an organization like under 50 people What about under I'm not sure 300 people Okay, what about under like a thousand people Yeah, so pretty good pretty good distribution My background is mostly kind of at smaller companies startups and kind of mid-sized tech firms But hopefully what what we get into today will be applicable across the board So I'll tell you a little bit about myself. So my name is Nick Stilao I do engineering at Pantheon as one of the first handful of hires there so if you If you're using Pantheon at all which I would recommend you're probably using some you know some software system that I built If you're using Pantheon and it's working really well You're almost certainly using some software which an engineer that's on the team now Got it out all my code and replaced it with much better working code I'm at and Stilao on Twitter and stuff and I'm a believer I think I'm gonna say believer, but I'm a but I'm a believer to a Justin Bieber believer I think he's like the most kind of dev Opsie of like the teenage, you know entertainer kind of you know trainwrecks but You know, I'm a believer. I'm a believer that there's a better way to do things. I'm a believer in dev Ops I'm a believer in teams. I'm a believer in collaboration And I'm also a believer that you should ditch work if there's a really good powder day Like it was in this day So here are here's some recommended hashtags I mean, I can't like, you know, tell you exactly what hashtags to use like this like if you are if you happen to be tweeting about this session There might there might be some good good ones here believers for dev Ops 49 ums. I think that's that's my allotted budget of ums for this talk So I think I have about 46 left If anybody wants to count their their more than welcome welcome to So this is the introduction unfortunately anyone who wanted to go faster We're gonna start start the beginning or at least where I think the big beginning is so The TLDR is basically dev Ops is awesome and you should believe in it, but we're gonna get into it from the beginning So who is this talk for? This talk is for anyone this talk is for everyone This talk is for people who believe that there's a better way to do things Whether you're a project manager or an engineer or a CEO or or a custodian or kind of whatever your title is This this talk is for people who want to be agents for change within the organization and groups and teams that they're working for I Want to believe I am a believer and if you guys don't believe that there's a better way to do things I don't know how much I can help you If you really don't believe there's a better way to do things I think there's a session next door on why we should keep Drupal 6 forever and ever and ever That might be more applicable to you at this stage There's not but there's gonna be a good wake kind of funeral for Drupal 6 on Thursday, I'm catching wind of so I might try to hit that up So change agent I kind of like that word it sounds kind of badass and it also kind of makes it sound hard And it is a little hard. I found this diagram, which is Which you it's like your own DevOps kind of change agent superhero and you get to fill in Fill in what your super DevOps superhero Kind of some of the things they might have in their utility belt and while you're doing this You can kind of think about what's in your you know What's in your DevOps tool tool belt that you want that you think will help you bring change to the table so in this case You know our DevOps superhero has a burrito because DevOps is hard work and you might get a little hungry So you might need that burrito there Kind of an open hand to make sure you're kind of reaching across the aisle meeting new people Giving high fives and celebrating your wins A fanny pack full of gifts because sometimes if you can't say what you really need to say In a gif or a meme, you know, maybe it just can't be said at all and You know a smile to make sure you're having fun positive mental attitude To approach the approach the organizational problems you're looking at so your DevOps superhero probably looks a little different than this But you know, I think you can be thinking about what what you have and you know What your teammates have And what you think will be necessary within the context of your organization To to kind of make some of that better way that you believe in a reality So why why should we DevOps? You'll see I kind of make fun of the word DevOps and I would encourage everyone else to As well a little bit turn it into a verb and different things like that, but why should we DevOps, right? So this might be something that you kind of are tackling late at night as you can't sleep and you're wondering what the meaning of life and you know How you're going to move forward and and you know what you were put here to do This might be something you encounter as you kind of start to have some of these conversations within your organization a Little resistance to change and you can have you know, there are a ton of different valid answers for this One being like dude because DevOps is awesome That might be applicable to your situation or something that you believe in Or more of the Jesse Robin style Operations is a competitive advantage and we want to crush our competitors It might be a little more refined something like well doing a value stream analysis. We've identified a couple wastes use applying the the system of the seven lean wastes and we want to improve the Cycle time and delivery time of our products So you can come at this from a lot of different directions, but the nice thing that we all have in our back pocket is that if we don't change Ultimately, we're all going to be beat out by people who are changing in organizations who are changing And so that's kind of a nice thing to know ultimately the people who are resisting change Will will find that change one way or another so We had a couple of hand raised about people thought they you know had a good handle on DevOps But you know not everyone in the room so wanted to go into what exactly is DevOps So these are a couple kind of marketing infographics include including one from pantheon about what DevOps is And none of these really make sense to me So the guy on the left has this really weird stance and he's kind of half and half The people on the bottom don't have eyes and not like that totally precludes DevOps But I just don't really understand what's going on there I think the photo on top they seem to be having fun Which I think is a big part of it So they're probably like that kind of resonates with me and then there's You know the unicorn and And kind of pantheons over there So I think there's all valid stuff here, but none of these kind of reach out and grab me and let like Are are the DevOps that I believe in So then we can say all right. Well, what is it DevOps? Maybe this is an easier easier way to approach the problem So this is like the Docker GitHub repo, right? So DevOps isn't a single tool like you can't install DevOps You can't like get clone and like run DevOps, right? It's not about the tools You know, this is a job description or this is um, this is like from another marketing website I love New Relic, but this but it's kind of easy to make fun of so these are their tools for DevOps And this page just keeps going down and down and down It's like every piece of software out there Everything is a tool for DevOps. And so maybe that's true, but I don't think we need a website for it You know and a DevOps isn't a isn't a role. You can't just kind of hire DevOps and make it happen although You'll see a lot of these and if you really feel like that's the best title for someone that you're hiring or in your You know go for it But none of these things relate to me our DevOps either So we really need to pin it down and kind of put a definition on this I like this definition from Adam Jacob who works at chef He's a really good hugger and he's also put a lot of thought into DevOps and what that means and Kind of wordsmith this pretty well There's kind of a link on the bottom chef has a bunch of good content that actually does ring pretty pretty true To me So if you're looking for more info look up kind of some of chef's DevOps Kung Fu and and other content But this definition is that DevOps is a cultural and professional Movement focused on how we build operate build and operate high-velocity organizations born from the experiences of its practitioners So I like I like this definition it actually kind of works for me I think the most the key piece that I like from this definition is the word movement, you know And I think DevOps is about movement. It's about the movement of value from real from ideation to Realization it's about the movement of organizational culture and norms from the status quo to the nouveau It's about the movement of diffs and bits moving out from dev to test to live to customers Ladies and gentlemen DevOps is about movement But you know, this is like we're kind of have to like embrace this the paradox of duality here dev and ops and dev and ops and DevOps In the words of Zen master suki. They are one and yet not one and I think if you spend enough time in DevOps you like you might hear some people kind of get philosophical a little bit I tend to do that so you can kind of get me back online, but I think You know so Suzuki talks about what it must feel like as a bit of water That's in a stream and then the stream launches over a waterfall And then it turns into these millions of little drops of water And then they reunite all again at the bottom when the water hits the river again And if you stare at this problem in the face long enough, you might start to look like this guy But you know, this is this is a hard hard thing hard thing to do and even a even a hard thing to kind of think about sometimes But it's essential to DevOps that DevOps is never done. I like this quote from Damon Edwards another gentleman who's kind of spent a lot of time Thinking about this and working with organizations banished done from your vocabulary. You're never done. You're never done with DevOps Just like Sisyphus was never done pushing this big ball up the hill over and over again this big ball of DevOps and I pretty sure the part of the reason you guys are in this room Guys and gals are in this room is that your believers? There's a better way to do things and you're ready to take on the challenge in one form of another of bringing some positive change to your organization and your teams and So you've already taken the first step But this will be lots of work to roll this ball of DevOps uphill against all of the organizational The kind of the status quo and the friction and the questions But I think that's why we're all here there's a French philosopher Albert Albert Camus Who did a treatise on the absurd when he was trying to wrap wrangle his head around some of these DevOps questions? and He came to the conclusion that absolutely the only way it works is that Sisyphus was happy He was delighted to have this obligation to roll this ball This rock up the hill again and again And so I think like that happy Sisyphus if you guys are looking to bring about change You're gonna need to kind of find your own Zen and your own reason why these challenges are the right challenges to be fighting and bring That energy to the table every day So how does one devop we went over that we can't hire it and we can't buy it and we can't install it So all that's really left is just to do DevOps And the good thing is that the only thing you need to start doing devops is just to start doing DevOps And you're like okay, okay, but really tell me you know we came here to get some actionable stuff So I think that's probably the end of some of the philosophical DevOps stuff But I can't guarantee it it might pop up again So now we're gonna kind of get now with that out of the way I'm gonna get a little bit more into the meat of the of the duck So building your DevOps game plan So here's this triumvirate of a kind of aspects of life beliefs and behaviors and Systems and these govern a lot of the actions we take day to day How we interact with other people even kind of the way we feel when we're doing our job or going about our lives and These all are essential to building your DevOps game plan If you master not one but all three of these you will be on your way to DevOps Enlightenment and we're gonna dig in a bit Double animation solid And so Beliefs and behaviors and systems are all kind of interrelated and interdependent I kind of came up with this one kind of set of relationships But I think a lot of these are there's a kind of a lot of different ways you could frame the relationships Between these you know kind of the beliefs we have and form the behaviors and the actions we take every day Those behaviors can get encoded in systems and kind of automated and systematized and then those systems reinforce the beliefs that we had to be there that we had the beginning and If you're not when you're trying to approach Organizational change if if you're not able to do that if you're not able to change each of these you're not going to be successful and a lot of the So excited about the DevOps track this year. I think there's a ton of great content But you know, I think in general a bunch of the content kind of focuses more on the systems and the tools and so I'm helping to kind of round that out and make sure we kind of understand some of these other factors that are critical in achieving DevOps success, of course Gandhi You know Gandhi nailed DevOps with this quote When he's you know talking about the interconnectedness of these different aspects beliefs behaviors and systems where your beliefs become your thoughts your thoughts become your words your words become your actions your actions Become your habits your habits become your deployment tooling and your deployment tooling ultimately becomes your destiny So I think I couldn't really say it any better than that Yeah, so we're gonna dig into each of these a little bit so Beliefs beliefs are interesting because one belief can inform, you know, dozens or hundreds or thousands of actions you have So beliefs are a really a high leverage place to think about organizational change and DevOps in your organization if you're able to identify those beliefs and And kind of either change them or even just understand them You know if you're able to change them You'll able to be able to change the actions of you know, many people and many actions over time So it's a high leverage place to to focus Kind of your energy. I think one interesting thing with beliefs is that they have an interesting relationship with like Truth and correctness so beliefs can be absolutely true or absolutely false or probably true or probably false And it doesn't really matter as long as someone believes something as long as they're They are kind of background and experience has led them that that's an important thing that they believe in that It doesn't even matter if it's true For example, this guy hanging off a cliff. I'm not sure what he's believing right now. He might believe be believing. This is remotely safe Not crazy He might be believing he's gonna cling back to the rock But if I was in his shoes, I would pretty much be believing I was in a pretty tight spot and almost certainly flat shortly And so this quote about What people believe to be true is which is that which is coherent to their already established cache of truisms? I like this because it kind of represented that that accumulation of experience Leading leading to a set of beliefs. I also liked it because it has the word cash in it And as a nerd that makes me think of like oh like my varnish cash or something like that And we all know what happens with cached objects, right? Sometimes they get a little stale. They're not so relevant or fresh and We what do we need to do? We need to purge those caches flush those caches So we also have these like belief caches built up within each of us and even within groups of people and sometimes You know we need it. We need to flush that cash So behaviors People are creep creatures of habit, right? You do in action a couple times those become routine Routines become habits habits are hard because you don't really ever think about them at all, right? So they're not a place where you can kind of apply logic or something. They're almost kind of automatic They're hard to change. I think there's some good research that forming or Destroying habits takes about like about 30 days of repetition And so that's kind of one of the things you're going to be up against is these behaviors and these habits that have built up with the team members and within the organization And that's kind of an uphill battle Especially attacking it on the behavior level and then so the systems we have the tools we have this is super broad Right, this can be everything from the language that we use To the kind of you know the develop the tools we use to manage the development process Whether sprints or waterfall and all that let down to the kind of individual I don't know lines of code or libraries we use or anything else and I think we're all familiar with the adage of if all you have is a hammer everything looks like a nail and I think that's a really good Explanation of why tools are so important If you use tools every day or they're really familiar to you They actually can really frame your understanding of the world around you in a pretty serious way So basically we need to be attacking kind of all of these all of these different areas beliefs behaviors and systems when we want To introduce some positive change Okay, so this is a little audience participation to participation exercise You don't have to participate, but I would appreciate if you did and I'm gonna participate too So what do you believe that others don't right so these are kind of your differentiating beliefs These are things you can leverage because ultimately You know we talked about how belief that beliefs are a high leverage place to focus And if you're able to identify just some core belief you have that someone doesn't You can use that to your advantage So these are some of the things I believe simple can be effective You know it doesn't need to be complicated in a big workflow or something, you know You know have fun, are you doing it wrong? I think not everybody believes that but that's pretty core to what I do a couple other ones So the audience participation part now is that you pull out your smartphone And you can either like Twitter or like email and you can just like email to yourself or Or your teammate or your mom or me. I'm nick at pantheon.io and Just gonna just mean the subject or whatever you just say Just write down one belief that you have that other people did not everybody shares This could be about DevOps or deployment or the comfortableness of the seat you're in or or your feelings of New Orleans or anything else. So I'm gonna do this right now. There's no text in the subject I'm gonna send anyway anyone finish up. It's still doing that mine was that I believe this is going pretty well so far Better than I thought I hope you guys share that but I'm not sure everyone else does so I'm not sure how much I can lever leverage that for for future wisdom but But I think this is a exciting little exercise and it just Took 30 seconds in the middle of some talk at Drupal con And maybe you guys will hit upon a belief that is quarter who you are and what you can bring to the table to to Realize the positive future out there Cool gonna go quickly into some belief and behavior and system examples Like from my experience particularly a pantheon and how these might constitute some challenges for you know, basically a successful team So what one belief I held firmly, you know a couple years ago is That shipping rapidly is more important than shipping perfectly And I'm not telling you this is the right belief for you or this is a kind of a truism We already talked about how beliefs have an interesting relationship with truth But for me at the time, this is a belief I held pretty dearly Arguably, you know, maybe it was the correct belief at the time because you know pantheon was you know Just trying to just trying to make it happen and just a couple people And we didn't want the kind of perfect to be the enemy of the good But this affected pantheon in pretty drastic ways. So this affected who we hired What roles we hired for and when what kind of skill sets we hired for and what even you know Hiring other people that believed this This affected how we our project management tools. This affected our deployment strategies. This affected Like I think this is a good example of one belief Determining that outcomes of many actions And so I'm not sure, you know, like there's also the question of Is this the right belief for pantheon at this time? Yeah, probably less so certainly less so than it was a few years ago But That's where we were and that did shape a lot of kind of who we are at pantheon, but there's still room to change And actually need to change So I was thinking of this one how many people like log on to a server and do sudo sue ever Right and this is one of those ones that I think is a great behavior example because this just like flies off your fingertips If you're used to doing this, it's like SSH sudo sue bam And you don't even you're not you're classically not thinking about what security you're kind of circumventing or you're not even thinking about security at All it's just your fingertips like flying on the keyboard Really ingrained behavior and habit I do this with like a lot of command line stuff if I'm like writing a blog post or like sending a text message or whatever Like if I zone out and zone back in I'll just the my fingers will type LS enter Like I'm what directory am I in when I'm in like a word doc or something, right? And I don't think I can ever really change that for my self and that's you know I don't think that's a terribly destructive behavior, but but it is a behavior that's hard to change And so you know early at Pantheon it was like David and Josh and me and Ray and like we'd you know log on You know like didn't have any customers whatever so we're like log it on and you know You log on sudo sue do some stuff on a server try to move it forward and this became a pretty you know Kind of ingrained habit SSH sudo sue You know then we hire a couple more people. Hey, we trust them. They're like, oh, how do I do this? You're like, oh, okay. You just as the server. Okay. Yeah Now you need to sue to sue just get yeah, just get rude and then you know go around and then you know We continue to grow the organization and like kind of look around and we're like, oh my gosh We actually not only do we have this behavior that's totally unthinking and is like kind of bypassing a lot of like good security Practices which are like totally feasible In Linux and and whatever We've kind of trained like a bunch of people to do this really dumb thing And then when we had to kind of take that away you know it was it was It was that was a hard organizational change, you know taking stuff away from people is always hard But I like this example of the behavior because some of the stuff you do on the command line or something You're totally not thinking your fingers, you know Maybe you're just clicking your mouse or whatever and some web UI or just kind of flying through something and not even really thinking of the repercussions And then worse you're kind of like training up new people to do the same thing Currently nobody does this at Pantheon anymore And then like a tools think I have a good tools and systems examples So Pantheon is a employees kind of all over the world and France and Romania and Montana and Oregon and Minneapolis and San Francisco and Cape Town and Prague and New Zealand And so, you know, we believe that's an effective way to great get kind of amazing Developer and engineer and otherwise kind of skill sets and people on the team But there's a lot of friction in making that work. We have tried so many combinations of kind of SAS tools and different microphones and wide-angle webcams and like all these things To try to make it work. We have a pretty good solution I can like share if you guys want later, but but the experimentation isn't done, right? We've done expensive things like buy a robot where people can like tell it whatever Tell it in and like drive it around the office and stuff Which is pretty cool, but I but Doesn't get used a ton We've done other like super cheap stuff like in our slide templates where we know we're gonna have a bunch of remote people We actually block out an area at the bottom that you don't put content in so that we can keep the faces up So you can see the faces of your like remote team while you're looking while you're kind of watching the presentation And so this is an example of how just like the friction of like you'd think this stuff is a solved problem but it's definitely not and This is something you need to kind of keep, you know Keep pushing that AV AV rock up the hill until you get it right or it's never done Right until you get a workable solution move on from there. All right, so building your DevOps game plan This is a quote from Eisenhower that I like to butcher the one I like to say is Plans are useless planning is essential and I guess this is what he actually actually said But it's important to talk about planning for a little bit. So if you If you want to just kind of break out your laptop and start developing like by all means do that You know, you'll have some fun Maybe you'll install like some chef stuff or some like, you know continuous integration stuff and it's gonna be awesome But if you really want to and I think you know engineers and I'm an engineer and stuff like that feels really good To like hack on stuff, but if you really want to bring change your organization, you're gonna need at least a little bit of a plan So that's what we're gonna go into of course after you make that plan as Eisenhower said it's gonna be totally useless But it's still essential or indispensable So what makes a good game plan right it should be actionable right the goal is you're actually gonna act on this It should be you should be able to communicate it. Well whether that's getting advice from people or sharing out to your team or Getting to your remote employees, you know, who might who might not be in the room when you're kind of developing these ideas It should be adaptable to be able to change to the circumstances that arise It should encourage kind of long-term alignment, but also set some real goals Notice that in your game plan, I didn't say anything about having The rock in it or cute dogs or people in tutus and it turns out that was kind of a crappy movie anyway But okay, here it is. So BAM pow wowza. This is how we're gonna build our DevOps game plan So this is kind of a modification of the Toyota Cata. Cata is a word. That's root meaning It's basically a form. So this comes from martial arts It could also be for dance or cooking or in this case an improvement the on business processes This came out of analyzing the Toyota production system the precursor to lead manufacturing And so I think this is pretty genius It's and this goes back to like one of the beliefs that I hold is like simple can be effective And this is really simple and it can be really effective It doesn't you know, it needs some like complex plan This is enough to get you going and we're gonna dig into it a little bit Basically as these four quadrants and we're gonna go in a little more depth on each one And if you're gonna need plans within plans You're gonna need some big plans some little plans You might you know, your big plan might be like do DevOps and make awesome and then your medium plan might be like Okay, reduce kind of cycle time or delivery time to two weeks and then your little plan might be like Oh, let's get our tests running and you know under 20 minutes instead of under an hour So you're gonna need these kind of different fractal levels of plans And now we're gonna deconstruct the Cata a little bit and go into kind of each quadrant and how these can help you so The top left quadrant one of my favorite four quadrants of the Toyota Cata Is I like summarized as the now slash problem So this helps you to find the problem space before thinking about the solution space So we are all great problem solvers a lot of us are paid to solve problems in one way or another whether that's with code or people management or anything else and One thing that that I think all of us fall prey to a little bit and certainly me Is thinking about the solution and not defining the problem So it's really critical in the in this Cata that the problem the now slash problem Section is separate from the solution Within your org this is important because this helps kind of remove some of the emotion of this So like if this was you know my Kitchen counter at home There'd probably be a lot of motion emotions. There might be some words like totally disgusting And stuff like that And those same emotions are you know present in the workplace and maybe you'd be like all right Okay, I get you are disgusted, but let's let's try to think you know realistically here. What's the actual problem? we're trying to solve and This also has a little hack that the sooner you start agreeing on stuff as soon as you start agreeing on what the problem is You've already kind of driven some level of alignment Moving on to the definition of awesome. I can't remember. I think this is like Spotify's take or something on the Cata But I like these names So the definition of awesome is where you can think about the kind of constraint free Unbridled fantasy world what this thing how this thing would be solved? This is this other little game. I like to play here, which is like how much can we agree about in 15 minutes? Right, don't worry about when you don't agree worry about where you can agree and get as much of that down This is vital again. It's that hack where you start agreeing on something and secondly If you can all agree in this like far distant future of where you're going you can kind of line up Solutions and kind of implementations towards that, you know If you think we're going to Australia and you think we're going to the moon and we don't understand that upfront We're probably going to end up building different pedal based solutions to get there Yeah, and this that can be fun too because you can say no, no, no, don't worry about those constraints Don't worry about time. Don't worry about people. Don't worry about money Let's just think about whoa and what we really want here if it was totally perfect So the next segment is the next target condition So this is where we start to apply some constraints. This will change for organization. It could be I think a kind of good general rule is like Although totally depends like is this gonna take like like five people for five weeks is like kind of an amount of time That you could actually realize some real change But it's really important that you get to a stable state, right? If Sisyphus is pushing that rock uphill He's not gonna like stop on the super steep part where he could if he like to stretch his back where the rock It rolled out, right? He's gonna get it up to a little plateau and then, you know take a a Powerbar goo and Hydrate a little bit and then get ready to push back up So I think these two things one really important to understand what a stable state is again We never know what's gonna happen even with the best laid plans But if we get our you know our technology or a process or our team or whatever it is to a state where it won't like Automatically regress if we can look away for a few minutes that lets us be agile and kind of turn around and focus on something else That comes up if it needs to and not lose that work Like the idea of conceivable effort, you know conceivable is like can you actually kind of conceive of that in my small human brain? Like yeah, five weeks. That's like after you know my wife's birthday or whatever like yeah, okay Yeah, let's let's give that a shot And again the specifics of the constraints you apply will be specific to your context in your organization bigger orgs You might be like hey, where are we getting it? You know a year Other people might be like hey, we've got two hours after noon Let's see if we can we can stable get to a good state commit the code or whatever agree on the change Although you do part of this is you do want to break these up into like pretty small like basically small chunks as you can Just to reduce the risk of moving forward You can always fall back to the definition of awesome, but you're focusing on the next possible stable state basically And then lastly we're going to talk about first step. So what can we start doing right now? That will help us start moving towards the next stable state towards the long-term long-term vision Just like this cute little duck waddling off into the sunset You know, they've proven that ducks are the funniest animals And I think duck butts might then are even funnier So there you have it This is the kata. This is the form that we're going to put into practice and continue to kind of work with Within our work and work through over time Reinvent constantly as needed You're welcome to kind of like this is a good starting place You're like welcome to kind of tweak it as you're you know as your circumstances need But this is the basis So we're going to go into some into a couple examples and There's like an opportunity for improv in here if anyone's really into improv Or we can skip that do that offline get a boff But so we're going to go through a couple of these doing the dishes which is like one for my home life, you know Trying to get some positive improvement there looking at our on-call rotation Which is something that we're actively kind of discussing at Pantheon And then and then we'll see where we go from there. So You know, okay, so my wife and I are like, okay, you know, she's pissed. It's disgusting. All right, let's define the problem here Okay, I totally agree. Nobody could miss it that the dishes are covering the sink Yeah, and the and the utensils, you know are dirty and she's like and moreover when I want to go my cookies I can't and so now I kind of pick up. I'm like, all right. That's totally true I totally agree this lack of cookies is totally untenable And so then we're like, okay, agreed. Boom. Let's go to let's like what would be awesome about, you know Our kitchen and our relationship with each other vis-a-vis the kitchen Cool. Yeah, the pans are clean. Yeah, things are put away. Oh that because that makes that makes okay We have we can get homemade cookies pretty much anytime in 38 minutes Yeah, that would require like the cookies put away or the utensils like put in the right spot So we know where they are so we can streamline that whole thing. Okay cookies in 38 minutes. That would be awesome Yeah, and then okay realistically where are we gonna get from here, you know, this is kind of where you can break it down further Okay, so maybe we don't need always need a new batch of cookies But it would be if we make a big enough batch like we can always have cookies around So I think that's like a pretty good stable state that like in our lives We just introduced that there are always these homemade cookies and that will have like some backstop that like at least our kitchen Operations are good enough that you know, we're able to make a batch and we need to replenish the batch that Somebody's gobbling on And so I think another like I wasn't able to like really Manage it in this talk, but I think you know metrics quantitative goals are really important So even like some of these will be qualitative and that's cool But as much as like quantitative goals and metrics as you can put in here That really just like helps. That's another tool for alignment And probably worth a whole talk on its own Then we're like, okay getting into first steps and I'm like, well, you know, I would like honestly I would do the dishes, but I don't like the way my hands feel when they're all like soapy and stuff Alright, that's easy Amazon gonna get you some big Gloves for doing all these dirty work kind of a couple other things. I have these big gloves They're like so thick. I think they're from Japan and they're like fishing gloves or something, but they make me feel like, you know, I don't know totally capable of getting in there So that boom we're on our way. We have a plan. We've throughout this process We've gotten a high degree of alignment on where we are now Objectively why that's a problem where we want to go in the far distant future of our, you know kitchen operating Vision we know we think we can get to a stable stable state which will be Easy to show whether we've made it there or not as well as some some first steps Yeah So the next one is a little more relevant like to work and stuff like that But so we have this problem at Pantheon We do have people all over but kind of on the EU time zone We have a lot fewer people and so we have a lot fewer people there that in particular They're on the on-call rotation And then that's hard for them because they end up spending a lot of time on call You know We want to treat on-call is a top priority to make sure our systems are reliable and you know sites have the Best of time that we can manage But for this team in the EU they're spending like an inordinate amount of time kind of being a little more reactive and not enough time being proactive we're getting more where you know hiring more and more great engineers and And so the rotation is getting bigger now We have like eight or ten people in the rotation in the US alone, but that means the distance the time between When you're on call is is spreading out And so that means there's kind of less of a good feedback loop where we're actually able to fix problems and just end up kind of break fixing what we see There's so much tribal knowledge So it's really hard to train new people on On you know what's it what the expectations and the tools and the tricks and the everything else that goes along with kind of debugging our production systems And so like as a manager my team is kind of you know like I've only been a manager for a few years and a Couple times my team is very vocally kind of said dude You know we're here to solve problems You can't solve all our problems and if you do if you try even though it looks okay here Like you're not gonna we're not gonna implement the best solution that we have because we know a lot about this stuff Because we're doing it day-to-day So this is kind of a case where using the Cata to help break out the problem space from the solution space and letting them handle that Working on definition of bossam we have clear expectations about what's being on call It's not relying on judgment. So that means kind of Our playbooks anyone can follow a playbook like if it's your first day You could be on call and still follow a playbook, and you don't need to know a ton of this stuff about how the system operates to be able to be effective reducing the number of alerts and and I think you guys can kind of take away the rest So This is where I was gonna do the improv one, but I'm not quite sure we're up for it today. I Don't think there's any huge improv fans. So you can go through this on your own And with that wrap things up a little bit So hopefully I've convinced you and I hope everyone here is in the room You know believes that there's a better way to do whatever it is. We're doing You know, we're excited and we're ready to be an active fearless Change agent with on our team and organization We have some tools now to help align on the kind of the collective vision in the future And we're gonna focus on some actionable next steps to deliver change And Lastly but not least Everyone here has to draw from find and draw from sources of inspiration This is one of it's one source of inspiration for myself This is a picture of Franconia notch in New Hampshire on the White Mountains and and this quote always rising never steeply and This is something I've repeated kind of a mantra for myself for a while You don't need to change the world and kind of big steps and bounds as long as you're getting a little bit better every day and Jay Rainer Edmonds was one of the heads of the Appalachian Mountain Club Towards the end of the 1800s who had this vision about recreation in these these mountains about creating a system of huts and trails There was always rising never steeply and in fairness. There's like quite a bit of steep stuff But this is always something I've been drawn to to kind of wake up in the morning and Get out of bed and you know Just try to make the world a little bit better of a place One smile one laugh one shared goal one act of collaboration Yeah, so that's kind of what I'll leave you guys with And if you want more of this another one of my DevOps inspirations Michelle You can't say crechey without see I crechey is talking Thursday About DevOps in the chocolate factory or something like that If anybody wants to rate me I would appreciate the feedback and Some other content that might help you guys so Thank you so much everyone for coming. Hopefully that was useful. Happy to dig into any Questions So are you still scared of DevOps in the back? I think we might have scared him out entirely But yeah, any questions I can answer or we can take him offline Yes The question was can I back up a slide? Yes Cool, thank you