DROdio (CEO at Armory) and Kate MacAleavey (Head of Culture and Leadership Development at Armory) recently had a fascinating discussion on YouTube with Joseph Jacks about "open source culture", in the context of building a Commercial Open Source Software (COSS) company. Armory is a COSS company of the open source project, Spinnaker, which originated inside Netflix.
Here are some of my personal top takeaways from the conversation. (To go directly to the transcript, click here.)
- There's a lack of common vocabulary and language when it comes to talking about company culture broadly and culture specifically when building a COSS company. When rubber meets the road, company culture really comes down to the type of behaviors the company will put up with or not put up with. If you want an open culture, default to public discussion and don't use Slack DM (as an example) unless you have to. Be intentional about the details, because those actions will reflect what is and isn't important to your specific company culture than what you say about your culture. *(Aside: this takeaway is very much reflected in Ben Horowitz's new book "What You Do Is Who You Are".) *
- It is important to build a sense of pride, agency, self-empowerment, resiliency, and openness (especially in the form of vulnerability) into your culture, if you want a can-do, start-to-start spirit to permeate the organization. That being said, freedom and agency can be terrifying, particularly for people who come from other organizations, where most organizations' default mode is "command and control". Thus, they have never felt "freedom" before, which can feel uncomfortable. Developing an effective process and mindset to self-select the people who embrace your culture is important. Not everyone likes the responsibility of having "ownership" or the power to act and experiment.
- For top engineers of an open source project, e.g. maintainers or core committers, their reputations and skills "transfer" with the project, not the COSS company that is commercializing it. (Instead of being a Xoogler, you might be a X-Spinnaker-er...). This dynamic fundamentally changes how a COSS company treats engineers and builds an "open source culture", because the company and the project are decoupled. Thus, the "how" is an open question and new frontier.
- Learning from mistakes ("how can we avoid mistake in the future?") instead of blaming the mistake-maker is key to building a "growth mindset" culture and a safe place for experimentation. This mentality, from a technical level, is codified as "Chaos Engineering" in Spinnaker, because the technology is made to empower application developer to deliver software faster and safer. The Armory culture embodies this characteristic, which requires a high level of ownership and agency, as well as the psychological safety that must come with constant experimentation and the failures that inevitably accompany experimentation.
Below are links to reference materials mentioned in the video, the video itself, and transcript to the video. Please share your takeaways in the comment section (way below) or on Twitter with @COSSMedia.
(This transcript is mostly machine generated, so please check with the video itself for accuracy.)
*Joseph Jacks: *[00:00:00] Hey everybody. My name is Joseph Jacks. I'm really honored to be here with DROdio and Kate at really exciting company called Armory. And we're here to talk about open source cultures and what creating an open source culture really means. I'm a pretty big advocate of commercial open source as a movement sort of abstractly and as a company category. So excited to ask DROdio and Kate a few questions and kind of have a conversation about armories experiences and how open source influences culture really. DROdio, Kate you want to give quick intros on yourself when we can dive right in when you start.
*Kate MacAleavey: *[00:00:40] My name is Kate MacAleavey I'm head of culture and leadership development here at Armory.
*DROdio: *[00:00:45] I am a janitor sweep the floors whatever needs to be done. Also the CEO.
*Joseph Jacks: *[00:00:52] So armory is is a COSS company. Commercial open source software companies are fundamentally different in nearly every way compared to closed core fundamentally proprietary software and technology companies from the lens of culture. What are the top three differences that you've observed at Armory?
*DROdio: *[00:01:13] So I think what I would propose is that I'd like to I'd like to actually say let's start with the common vocabulary because everyone uses the word culture. I think maybe a little bit differently. So maybe we can talk a bit about how we're thinking about culture broadly as a term and Kate has a lot of experience here and then we can circle into at a cost company what does that mean for us. So maybe I'll set the stage for that and then you can take us deeper. You know there's this great saying that culture is read our link and a lot of people when I talk to other CEOs they talk about trying to influence the culture or impacting the culture. I think my favorite thing about culture is that it's like a shadow that follows the company everywhere that people everywhere and you can yellow your shadow what it's like that change anything. So really when we say culture I think we're talking about the output of something that's much more intentional and sometimes not sometimes it's it's it's done intentionally and some of it is just kind of happens but the result is culture. So I think about it more as a you know as a a very complex formula where there's a lot of variables in this formula that lead to a number and the number is what we say culture we say a company has a good culture because the number is X but really it's about all of the variables that define that number. So it's things like what is the psychological safety of the people at the company. How competence are they. How well are they being onboard it. How are they working together. How intentional are they in their actions and the ways they behave towards each other. There's there's all these variables that it's kind of interesting we don't really have a good language to describe the sophistication that it takes to that end up in a good culture. We just use these examples of this company social service lunch. So they have a great culture but that's just one of so many things that treat this like operating canvas. One of the reasons that we hired cadence armory was because we believe that it's important for us to be very intentional about building that operating canvas or creating optimizing those parts of the formula to end up with good culture. So I'd actually like to see if we can kind of start with a common vocabulary and maybe share how our race trying to be intentional about the way that we think about those things and then go into the cost piece.
*Kate MacAleavey: *[00:03:31] Yeah I think about it really is that the norms behaviors languages systems really the collective actions of the community and what we put up with and what we don't like at the end of the day. JJ And just like what are the behaviors that we want to have and what are the behaviors we're not OK with like really coming down to that and like really what small tweaks did he consistently make with each other and with our culture to show up in a way that makes us feel like we're representing ourselves that we're proud to be who we are. And also that we're you know we're unique we're not trying to copy other people like we may study other cultures and look at other instances of success but really it's like how do we how I think about it is how do we leverage the strengths and unique content within this organization and bring that to life and be really clear and intentional about it. And I think armory does an exceptional job that every detail really being thoughtful. So that's just kind of my thought around it loosely.
*DROdio: *[00:04:25] So the reason this matters so much to us is because like you said JJ we're commercializing an open source project that project the spinnaker and Spinnaker exists to help the world innovate faster it's a software delivery platform and it was built by Netflix and Netflix has a really interesting culture. You know at Netflix you know developers have access to production and that's not the case in most global 2000 companies. And so we we had this term of like you know a great culture at Netflix. Well what is it what does that really mean. It enables that. And how can a company like armory bring that really that culture that is Netflix that is presented in this project in this platform into other large little 2000 organizations and have them be successful with it. So the reason that we are trying to be very intentional about the way that we build this culture that we that we make these choices is because we believe we have fundamentally a bit of an extra special responsibility to live these values that were then product pricing and bringing into these large companies. And so to talk about that a little bit the I actually think of it as two to kind of flip sides of the coin again and gang. So there's there's Spinnaker the platform and spirit for the platform the way that I really would describe it is it is a sophisticated way to manage risk in production that allows large companies to make things safe to try. That would otherwise not be safe to try to get a little bit technical for a second. Let's just take the example of for example canary deployment. So in most large companies if I'm delivering software to production I'm terrified that I'm going to nuke the user base with that feature change. What that does is that I end up with a culture where I have a lot of integration testing unit testing many could automate a manual judgments. VP saying this is not going to go out until I approve it because the cost is so high and having failure in production a platform like Spinnaker allows application Roman teams to get access out of the box to their deployments which has the ability to limit the blast radius of a change. So when I'm a developer and I write code and it goes out into the world maybe only 1 percent of the population sees it. And so with a platform I'm starting to codify the culture of enabling the application developer to have more autonomy to them for them to have more freedom for them to go to get the really the art the code that they're writing out into the world with velocity while the company is able to control the damage or the blast radius that that one developer can do and so obviously this is Dev ops right. It's it's aligning Dev and Ops but that platform is what gives the company the ability to say I think I feel safe now to change my culture and give application of others more control and more of an ability to have what they want which is the ability to impact the world while still maintaining you know not nuking the user base. So the platform gives the company the ability to manage risk and production more restricted ways which then gives license to the company to start to change the culture. So we see these two things are incredibly interconnected things which is why we want to be so attention about the way that we build our culture. I think this is one of the things about commercializing an open source platform that's different from other companies. There are so many ways that we have obligations to people that most companies don't and those obligations also create opportunities. We have patients to our employees we actually castles a tribe. We call ourselves a tribe because we believe we really need to invest not only in the professional but in the individual as a whole and not only in the individual as a whole but the person that individual goes home to every day because if someone's going home and their spouse or their family member is saying wiring inside this company I don't think you should be. It's really hard to come and do their best work. And usually at most companies they don't care that much but that at a cost company that might be an engineer that's a core committed to a project and that core commuter ship goes with the engineer if she or he leaves the company. So we have to be thinking about how to empower the edges of the organization and have how to make them just really invested and happy in their work and how to be recognizing them as people so this gets to the work that Kate is doing so that ordinary and also in these large companies that are trying to figure out how to innovate faster and transform because eventually we're doing it for them. So I don't on occasion to talk a little bit about how we're doing that.
*Kate MacAleavey: *[00:08:44] Well I'm happy to also just bring like the emotional lens and some of that language because to me like that's where I'm coming from and like thinking about the three differences that you're asking about JJ you know what I think is so unique is we're not functioning from a place of fear right like what we're doing is bold and it's brave. And the feeling of pride is very different. It's a different type of pride right cause it's for the community. This is like hey let's lower our barriers let's be as transparent as we can with our culture and with the external culture right. As long as we are continuously learning and adding value and in sharing I just don't see why that we need to have our defenses up right. Like we're not trying to protect anything. There's nothing sacred here. It's like how do we make everything better not just armory. Honestly when I really was thinking about it a few days ago I like the amount of heart in this company and the people you just feel it you know. So I ask people like have you been a toxic culture before I was like oh hell yeah. And look you know. You've walked into it you've experienced it and you've also probably been in really great cultures and you just you know and you feel like you see how people treat other people and you see how they show up for people. And so I think about the three big differences right now versus that close company versus the healthy one right. It's just it's such a brave space one of our recruiters kind of coined this term like it's a really brave space to be in because we're opening ourselves up or more vulnerable taking that risk because we believe that the learning is more important than us keeping something secret to our eyes.
*Joseph Jacks: *[00:10:13] This is like really profound stuff and I was thinking about one element that separates closed core sort of like fundamentally proprietary completely proprietary technology companies from commercial open source software companies are sort of open core oriented companies and it's in it's like something DROdio was saying around the engineers kind of having that transferability. This the software or the technology in commercial open source companies is kind of decoupled from the company and it has that transfer ability so many companies can form and nurture and invest and capture value around Spinnaker for example but Armory is viewed as a highly credible trusted center of gravity and commercialization partner to enterprises and to companies looking to run spinnaker and complex environments and so on. How do you how do you sort of think about all the other functions even beyond engineering like people that basically work at armory that learn about Spinnaker or learn about the open source technology that themselves developed that sort of skills skills transfer ability that now obviously would be generalizable and maybe technology and sort of like working in the software industry or all quickly to your open your open core which is spinnaker.
*Kate MacAleavey: *[00:11:28] Yeah I think I was where you on your questions like What were the hardest things that we're trying to teach and have to train right. I think engineers technically have a better understanding of the growth mindset data growth mindset at all.
*Joseph Jacks: *[00:11:39] Yeah I know a bit about it but unpack that for for me.
*Kate MacAleavey: *[00:11:43] Growth mindset comes from Carol Dweck to work from Stanford and what I love are a few different parts to the theory but really it's around like really focusing on the learning and being OK and instead of being so obsessed with failure and money that is the thing that pulls you down. It's just like wait what did you learn how do we iterate. How do we build upon it. Right. It's really building resiliency into a system into a company. And so engineers I feel like have early training in that. So this what we're doing makes a lot more sense to engineers. But then you know like marketing you have sales and everybody else who's like terrified of failure we've been trained our whole lives be scared. And my hope is that we build like an entire community that's just like what did you learn. And like what how do you keep going. How quickly can you bounce back from that and add value to what you're trying to do. So it's me like that's one of the biggest nuggets of really implementing that growth mindset across the entire company. Because again ones are really focusing on the learning. You're really starting to speed up your progress. And then also just thinking about how to like contribute elsewhere. So it's not just like in your team. But how do you think more functionally. I think more globally how you think about the whole open source community. So the engineers definitely have a leg up here and that's why I'm always sharing with people like you. Questions like go talk to the engineers. Ask them bring them in. Like this is the fun part about the collaborative process but I think for me that's like one of the big things I think about a lot.
*DROdio: *[00:13:04] I'll go deeper there with two specific examples. One is the way that we are codifying this culture on a software platform and the other is an integral to our Armory. So I'll start with the former. We were sitting down with Jeff Rothschild top engineer at Facebook was a you know early engineering leader helped scale Facebook and he said something really interesting about the culture that he really worked to create a Facebook which was when an engineer took Facebook down and he had to give the engineer the talk about the fact that the engineer just took the entire site down. The talk was how can we improve the way that we work so that you or anybody in the company can make this mistake in the future and it won't impact the organization. So it's not about claim it's not about being wrong it's what's it's what Kate's saying. How can we learn from this. And when you think about it in that way you're either succeeding or you're learning and learning is all about retrospective and being very good about doing blameless retros which I think a lot of companies aspire to do and a pretty typically not good at doing well with that that that that approach is codified in Spinnaker as cars engineering and cars engineering is literally the idea that you know you've got cars going in and turning off infrastructure and seeing if the application form falls over so these two things are just so interconnected in our world which is why they're so important for us to live. So the way one example of the way that we're living that internally is this idea of a common experimentation platform. It's actually kind of mindblowing to me. I don't know of any sort of like SAAS tool if anybody wants to build this week we would love to consume this but when when someone says the word experiment an engineer usually means something very different than like an account executive an engineer will typically be talking more about scientific method. I would also say that hypothesis lab measurable success criteria would have a defined end date. I'm going to look at whether or not the hypothesis was successful in retrospect on that a salesperson salmon run an experiment with pricing and that just means they're going to try 10 different prices and I'm just going to see what happens right. So we really given the importance of defining the word experiment in a consistent way across the organization. When somebody says I'm going to run an experiment then that means something really special to us. What it means is we've got a way of working let's call it the golden path. It's not the only way. It's not the policy. We don't say you have to work this way. It's just the best way that we know it's the best practice for whatever it is. So many of us ran an experiment on that they're going to go off writing off the golden path. So this might be for recruiting right. Like if you're a hiring manager you can use the recruiting team. That's the golden path but you as a hiring manager at aren't we we don't tell you that you can't go off and do it yourself but if you're going to do that to you as an experiment where you're going to define what you're looking to improve and learn so that we can then go back and improve the golden path. So in this way people are welcome to go try new things and if they don't if they're not successful in those things it's something they fail as if they learn something and one of those things that they learn so we can then define another set of experience around that. And so having a common experimentation framework within the company across functions is very important to us. We're having trouble. It's really a very kind of haiku thing we've got a trailer park which is like our experiment template and we can just copy that part right there right there. Experiments out. I'd love to have a size platform which modifies this I don't think that it exists. I don't know of it but the important thing here is how can we ensure that people understand that they have a license that they have to ask permission. They have license to go off roading try improving the default way that we're doing things so that they can then make me feel better and then we're getting really good at iterating on that. So a lot of smart steps to make whatever we're doing better with it whether it's recruiting or whether it's how we build the product like anything whether it's the way that we sell it. Anything can be improved. Make sense?
*Joseph Jacks: *[00:16:54] There's there's about four startup ideas in there DROdio. I want to I want to segway to something that you've touched on both that I think would be super interesting to a lot of people. How does an open source culture provide a competitive advantage for a company in today's fast evolving world?
*Kate MacAleavey: *[00:17:14] Yeah I think what is really interesting about it is like I said the word resiliency so you can be resilient organization. So when people are feeling empowered they're self empowered right. Like You're not relying on its reliance as we talked about engineering chaos right here Chaos Engineering maybe botj. And right. Like it's like if like the CEO is not in town or if all of the executive managers are gone. Can you still do your job. You still have the ability to make decisions right when you are like the one closest to the work and you're sensing something that's really big and important like hopefully you feel capable and competent to do something about it. And so it's really being able to respond really quickly to what's happening right. I think within the complexity of the environment that we're in it's so important to learn how to sense and respond. Like I say this all of the time I sense and respond and start by starting and when we define experience I really push for small moves over big right really small things make it safe to try and start as small as you can but you're gonna quickly start looking and finding data. But I think that's just really important. The resiliency as well as just feeling empowered to actually go make a difference or anything.
*DROdio: *[00:18:24] So I will I'm going to build on what you're saying here and I'm also. This ties very perfectly back to the company that we're building. So the reason armory exists is to help the world innovate faster through code. We we believe that software is the highest leverage way to improve humanity when software drives our cars were a thousand times safer software's not flying our planes. I think we're we're all starting to see the impact of software in our lives. And so we want to help the world shipping our software faster. We are commercializing Spinnaker to help companies go from idea to future in production in minutes instead of months. This really means that in order to do this companies need to be able to think of it like strip mining. They need to go to churn through ideas rapidly. It's soft the software is very not obvious it's not obvious that it's a good idea for you to let me into your house when you're not home rummage through your fridge in your pantry use your shower and your SO sleep in your bed and then pay you one hundred dollars. That sounds like a pretty terrible idea. But it's Airbnb it's a 30 billion dollar company and because software it's the because the next great feature is not obvious what a company needs to go to get very good at is churning through ideas quickly like a strip mining machine instead of panning for gold where you're just looking for one or two calls like it's going to get through at ideas with velocity to find the good ones. And so this is why devops matter so much. This is why having that deployment velocity and Netflix deploying seven thousand times a day instead of once a month is such a big like strategic competitive advantage. Well in order to do that we believe you need a or like Spinnaker that does what we said earlier to manage risk and production in more sophisticated ways so that you can churn through ideas quickly instead of having DP say I'm going to not allow this to be about production until I approve it. So you can find those golden nuggets which are good ideas but you also need to make sure that the employees in the company and specifically the application developers are empowered and have agency and have a vested interest in digging into that innovation and those ideas to to to innovate to then figure it out. And so we really believe in this concept of empowering the edges of the organization like that is our litmus test should something be centralized or decentralized. What's going to empower the edges of the organization the best. Should something be transparent. How do we most effectively empower the edge so the edge is the person that has the most contacts because it's not obvious they're the ones with the most data. So they're the ones who will be out to make the decisions the most effectively to innovate the most effectively. So from a cultural perspective what this means is a great book called "brave new work" not brave new world brave new work by Aaron Dignan and this book talks about a stoplight versus a round about culture a stoplight culture is a traditional command and control corporate culture where employees are told when to stop when you go when they turn left they just wait for instruction then told what to do in a roundabout about culture the employees if you've ever been in a roundabout in Paris it's a lot more stressful to be in a round about because you have to merge with traffic without colliding with traffic. So in a roundabout culture the employee has a lot more agency that a lot more control more power and more responsibility as well. So when you merge a software platform like Spinnaker it allows a company to churn through ideas faster with a culture that has the employees showing up caring about what they're actually building and and being curious and wanting to innovate. That also means that the culture has to not be command and control it has to be executives that are saying I don't know the best answer because it's not obvious because it's software innovation that's a knowledge economy it's not building a widget right know how to build a widget that's effectively it's ideas and innovation. So as an executive as a manager my role moves from telling the edge of the organization what to do it to instill empowering the edge to figure it out. So I'm more of a coach instead of telling them what to do. I'm more of an enabler. I'm more of a storyteller and trying to rally everybody towards these common March objectives so every everything changes. And this is why it's a yin and yang because you need that platform to enable that strip mining machine need that culture to get people to actually use it. Does that make sense?
*Joseph Jacks: *[00:22:33] One hundred percent and I think that maybe segway into in terms of like this what it sounds like is like kind of a new organizational design because we've exchanged a little on this journey others like her locker sees there's things in between you know democracy Marcus's representative democracy's all different types of governance models. But what you kind of touched on is round about culture and what we're scratching at with like an open source culture kind of has a religion poorly defined yet and the industry is kind of you. So in doing that you're always pain points to embracing new things. So for armory and your experiences and just general observations kind of committing to this path. What are the most sensitive aspects or it's sort of tricky things about embracing an open source culture with the financial compensation incentive strategy. You talk about that a little bit.
*Kate MacAleavey: *[00:23:24] Yeah yeah I read that question JJ. And to me the most sensitive is actually so many people haven't been empowered their entire lives so to rock organization maybe after 30 years 40 years and just sit here and be like oh you have agency and you have freedom people will get terrified they don't know what to do. They feel really out of control they don't feel supported. But there are a lot of negative emotions that can come with it. So we try to really clear about what we're expecting and what we're trying to create here so people can really sell slept in and out. So we've had just hires that like once you know we were going really clear about our environment they're like oh yeah I know but I love commanding control like I just want to sit back I want to be told what to do. And that's totally OK. Good for you. There are so many companies that will help you and I and I want to point that out like it is OK. That's just like you need a lot of structuring and you just need that kind of management style totally. But that's just not what we're trying to create here and so I think that's just something that's really hard for people to kind of get their arms around right. And also how do we teach that it's so easy to say agency and empowerment what does that look like. What's it feel like on a day to day basis. How are we integrating that into our conversations. It's like the smallest behaviors. So it's me that's something that I think about quite often. Abbott's probably more sensitive than like I haven't heard a lot actually buy our financials and everything else. I'm sure we'll have our moments. We're certainly not perfect by any means I don't want to paint a picture like that but for me that's what strikes me.
*DROdio: *[00:24:54] Yeah. So I completely agree that this is undefined. Like I said at the beginning of this I actually I don't think that we even really had the common vocabulary and language to even have the conversation because cultures mean something different to everyone it's usually just the output of all of this. So you knew all about organizational design. I called it an operating canvas which I got from brave new work. So yes this is new and different. And if even calling it an open source culture is something that's really interesting to just play with and kind of see how well it fits I'll tell you the things that we believe to be true are things like defaulting to transparency that that doesn't mean that everything is transparent and means that we have to work to make it not be transparent. So we're flipping and we do that with everything we do. We do that with our comp so inside our way our comp is transparent to everybody at Armory. We just haven't found a reason to make it be not transparent I have so many CEOs that have told me that they don't think that's going to scale. And I didn't think there was even going to scale to the size that we're at now which is Series B but it's been a complete non-issue. And we do it because we do it for a couple of reasons. The first is it's our culture to default the transparency unless we can justify not doing it. But then we get all these beautiful benefits like we can ensure that we are paying you know of a woman and a man the same amount to do the same job and everyone in the company is holding each other accountable to that. Negotiate harder and get more pay. So you know that there's all these benefits that we get from defaulting to transparency but also evidences in the way that we work in Slack. We we we don't use email because in email conversation is lost in e-mail it's only the people that had the conversation. We can see that we don't use dams and stock unless we have to. So instead we always try to sort of conversation in the most public channel that we can and then we add mentioned the people that need to be a part of it. And we think of it as a push versus pull model like there's a river of information that's flowing you can dip into that river if you want the transparency but you'll be out which you need to know. That can be really overwhelming for people that aren't like full stock employees that really know how to use technology well. So this is what Kate's talking about really self-selecting in. We give our investors a lot of transparency into the company financials we have a API dashboard that all of our investors can see as we're updating it on an ongoing basis so that there's just all these little little little things that we do to really empower the edges of the organization to have as much data as possible because you can't make the best decisions if you don't have the data. And I think that's really where it comes down to again thinking about what's most effective at the edge of the organization because it's the edge that's innovating because it's the ice that has the most context.
*Kate MacAleavey: *[00:27:37] I think like building off of that we're continuously steering right here just like steer once when you are a sailboat in the ocean hopefully down here I don't know where you're gonna end up exactly right. We're constantly responding again to the environment and what's happening around us instead of like committing to something and be like matter what we're going to summer to stick to this path even if we only go under the rocks. Right and a lot of companies do that. So there's a lot of anxiety and uncertainty that comes with this type of work in this type of environment. So if that is something if you're dying for certain I say the company all of the time that I'm OK with your discomfort like I'm comfortable with your discomfort because we have to sit in this ambiguity that's very very high. Right. If you don't find it exciting if that doesn't light you up. That is like bring that spark to your eye. Probably not the right environment for you. You feel like you have to be excited by that otherwise like you're going to feel really crappy when you go home and like the human brain Rob certainly we were looking for patterns right. Like we're designed in a certain way. So there's so much you can hold on to. And right now there's like a craving especially as you scale right our structure structure structure and we have to push back against these natural inclinations that we have to make sure that we're staying like Agile mean and responsive and not getting into these like very primal parts of our brain that are dying for certain.
*DROdio: *[00:28:51] It's always easier to say no. It's always easier to say no than to say how can I start from a yes.
*Kate MacAleavey: *[00:28:57] And so even though we have to figure out our org structure right like right now we're very much like dealing with art. We have a typical hierarchical top down structure right. But our culture and what we believe in is not that so we're feel the real tension of our powers our structure really reflect our culture. And it's a conversation we have to have and we have to figure out we have to iterate on. But that's just like the that's the reality of what we're trying to do.
*DROdio: *[00:29:20] I think that's a really important part here which is we don't believe that we've figured this all out. Like you're saying this is new like COSS is new, open source cultures are new. And How do we need to iterate to be very good. How do we be very good at iterating so that we can continue to be improving. This is where the common experimentation culture comes in to make sure that if somebody wants to try something differently they can because they might want to make it better. We don't have all the answers. And it's interesting when when a new tribal some where a tribe we call ourselves a tribe so an employee is a tribal armory when a new tribal starts the first thing I say to that person is you know you're used to entering a company where you're not allowed to do anything until you are flip that and understand that you are empowered to do whatever you see needs to be done from the very first day that you're here. And in fact we encourage you to do that. And if you do something wrong then we're going to learn from that. And it's OK for you to cause a problem when you're trying to make something better because that's an opportunity for us to learn and improve the organization. That's a really empowering exciting and like harnessing scary kind of all wrapped up in one thing when people walk in.
*Joseph Jacks: *[00:30:36] I think it's incredibly high leverage way to build companies and companies at the end of the day are just a group of people that get together to build a product or service and vast majority of companies are the exact opposite of what we've been talking about here. There are few people that have to be told every last detail what the definition of their job is and how to execute. So I have two more questions and I think there's sort of both broad questions but they build on what we've been talking about over the last few minutes and then I'm going to wrap up. But the first question would be you mentioned your store is like evolving this as Armory grows your series B Company. Now you're not you know a seed stage a company that fuels a few people. So what is what is this kind of open source culture look like inside of an early stage company as compared to a growth stage company.
*Kate MacAleavey: *[00:31:26] I actually think a similar and different. Of course the smaller you are. Look I actually think that early stage startup is probably more open source perhaps than any other stage. Just because you're you're greedy you're hungry you're willing to try anything that engineers are selling. They're flying around just talking to people like it's just so that everyone's willing to be so hands on. And it's like those responsibilities are blurred. So it's really like doing whatever it takes. And there's like such a rich history. This is why people are so obsessed with startups right. It's fun and it starts working. So I think actually probably the earliest age of a startup might be the most the best representation of this because as you get bigger more voices you know more organizational that more technical debt is piled on no matter what. That's just part of the growth process. And so you know where we're at right now is really thinking about you know what it's going to look like when we're a lot bigger and how do we educate people as quickly as possible. How do we ramp them up. How do we onboard them. And it's always like in the back of our minds. And so you know I I can't even tell you. I think from week to week right now maybe months and month but like a lot of we do end like there's also a lot of companies that plan excessively. I think what I'm dreading is first questions is like how long is it gonna take like transform our entire organization. And I was like as quickly as you want to go right like it's up to you. I can't do this work for you everyone has to transform. Everyone has to change their behavior their way of thinking. So it's really up to every person organization and we have to think about how do we get people experiencing this understanding embodying it as fast as possible. Do I have an answer for you right now. No but we can definitely talk later.
*DROdio: *[00:33:06] Well I guess I guess the way that you're representing a cape is actually a great way that we could try to figure out how to measure this which is it. It's about the magic that companies lose as they scale. So startups are these magical environments where there's very high trust is very high trust has a very high context and because people do believe in each other as individuals and they know the person next to them you know they have the best intentions they they know that they are competent and capable others. There's just a lot of psychological safety in these small environments because everything's so scary outside. We have nothing to lose so it's you know this small group against the MP the possibility of what they're trying to get done then fast forward to a company it's 10000 people and its management is saying the employees on what they're doing employees are saying that management's not providing leadership and there's no trust. Not give developers access to production. So so really what we're trying to figure out is how do we not lose that magic as armory scales and then not not only that that's hard enough right. Like I don't know that any companies ever figure that out. Not only are we working to figure that out how can we then bring that into these global 2000 companies that are our customers like we have unfortunately fortunately thank this customer we have a fortune 50 healthcare company like all these all these big companies that are realizing that they need to be more like a startup in order to be able to thrive. They need to be able to innovate faster. And so it's this really interesting dynamic where we're trying to figure this out as we scale and then product ties this into a platform that we can bring into this company so they can have it themselves. And I think in this case saying we don't have the answers we're still figuring it out. One thing that I think has served this very well. So we have three core values and the third core value is that we have a culture of experimentation including experimenting with our culture and that's really the second part is the key part which is that we it is safe to try. We are OK saying we're not going to be dogmatic about it. If if salary transparency doesn't scale we're not going to kill the company and say that we have to do that. We're going to figure out how to solve that by experimenting through this rapid iteration framework that we talked about earlier. So it's really about being good at that.
*Kate MacAleavey: *[00:35:16] And we also we're trying to create as few policies as possible. Right we're trying to be as like we all want to be restrictive at all with our environment and what we're trying to do because that's actually not a roundabout culture right. Like fewer rules the better. But JJ I've been in culture transformation for a bit now and what this requires is a lot of maturity a lot of self-awareness a lot of ownership a lot of responsibility on everybody's part. And again a lot of people are coming in where they just have never taken that kind of accountability onto their shoulders and not like an accountability way that feels badly. But like the pride of being is fully mine I fully own it and getting that excitement. And once that happens when people pass that threshold but I know that they get it. You see them but really light up and being like oh this is mine and I'm OK if this goes bad for you I'm like Hell yes like there but like those are really tall orders for anybody any adults. Right. Like that level of self-awareness and maturity is big. And so I think that's something we always have to think about as we scale. But again we're not trying to build a lot of policy or thinking about like what's H.R. going to look like here and how do we keep aside from the legal parts of it. Like how do we build something that's unique and different that really just fits who we are and really reflects the open source community that we're looking yeah.
*Joseph Jacks: *[00:36:30] I guess what was final question since we've been going for a bit here would be like just kind of framing you mentioned a lot about internal transparency and open open kind of data and access to a variety of things. There is also the external dimension to this sort of your sphere external stakeholders your your customers partners types and competitors. The overall ecosystem if you were to sort of think about framing inner source as referring to something like this kind of open source culture around sharing everything internally and then open source externally where where you're essentially like me basically sharing everything with the world or everyone outside of the company. How do you think through the differences and sort of like what would be like a transformation towards either end of the spectrum or embracing both. When you see companies like things with social media management coming up buffer in New York that was little I mean they published lots of their internal company data and financials and copied stuff publicly. We've seen you know we've seen companies like get lab which is a alumni of why see like armory very transparently sharing product strategy roadmap pricing everything publicly through an open source company handbook. There's lots of different approaches to kind of sharing things externally and internally. How do you see the balance of those two things kind of evolving over time. That's the final big question.
*DROdio: *[00:37:59] So I mean what's what's also various thing about your question is that we are a company that's commercializing open source right so we are bringing that kind of the specialness of an open source project like Spinnaker into these large global 2000 environments and we are building a profitable business off of that that that core project that is providing so much value to the world. And so I think this is also where we don't yet have the language to really talk about this in a defined way. Here's here's how I think about it. I actually just shared with you JJ a screenshot that Kate shared with me. I've put it into this document that we've been looking at for questions called even overstatements and Kate shared these guidelines and principles progress of imperfection small moves over big moves less of or more experimentation over planning open over closed consent over consensus freedom freedom over control. What I think about what we're trying to achieve we are trying to help the world innovate faster and we are trying to build an operational canvas which results in a culture that is the most effective in doing that. The thing that I know to be true is that the old command and control the inner the complete inner source close to do that is not the optimal way to do that. Chuck just like with the fact that we are commercializing in every source project if you go to the other extreme like like buffer I also don't know that that is the absolute best way to achieve that objective. I don't know. We're figuring it out. We we are doing it through these there's this concept of these even overstatements we want to empower the edges of the organization to innovate. We want to product ties that we want to bring it into big companies. How do we iterate to figure that out. It feels like there is this this term open source culture which is really applicable in that journey. But I don't know where the boundaries of it are. Like for example we haven't made armories comp externally available to the world that would be like truly open source. We chose to make the boundary you know when you are a tribal at Armory and you know and the reason was because it was going to be just like we've seen for buffer there's there's there's a lot of friction when you do that. And it felt like more friction than the benefit that we would get from that friction. But as I mean it's the right it is the beginning to experiment around that and see if we can be more open but I think that's the thing it's like we're always trying to figure out how we can be more transparent be more open empower more buy more access read more empowerment I mean maybe that's the thing that open source culture is the aspiration and we're just trying to push those boundaries as far towards full open source as possible. But we don't really know whether that is in most companies on the other side of that spectrum.
*Kate MacAleavey: *[00:40:45] Now when I met DROdio, I was like you know it was just very clear you know we got along because like we're both boundary pushers. I was like Look I like to push the envelope like as much as possible so let's find all of the edges here and this is just completely unexplored territory and we're totally down to go find that like in a safe way right. But like we'll learn we'll find out what's going on. We'll probably fall on our faces a few times. We're okay with that but like we're for instance we're gonna be setting up a podcast like talk about what happens in our culture and what we're learning and to me at the end day, JJ, again it just comes back to the learning and the fact that like this is such a communal effort to me like I'm really land on where my philosophy is is that we're better together than we are apart. Right. So when we're working with Google we're working with Netflix or having these huge conversations about how to improve this community and really putting that as the centerpiece of what we're working towards. I just think that as long as we're adding and contributing and sharing our learnings there that I think that's what makes us up and that makes us really happy. So how we're going to do that I guess we will be sharing as we go as we figured out.
*Joseph Jacks: *[00:41:48] This was a super awesome conversation. Thank you both so much for your time.
Top comments (0)