 So we are coming to the closing keynote of the day which will really round things out We're going to hear that one more talk Then we're going to close the day's content and then move into our evening event open in sparrow Which I'll tell you about afterward but right now it's really a pleasure to welcome Chris Kelly to the stage Good evening so it's been a long day, and I really appreciate you guys giving me your attention I know how much a You know eight or nine hour a day can be when you're sitting in sessions and talking Something I found very compelling about this conference is actually Normally when I go to an open source conference, I mostly talk about software almost exclusively And this conference has been so much more than that and that's been really interesting to me sort of a new thing for me and I actually think it works really well because the talk I'm about to give Starts from software but moves into the cultural ramifications of open source and what that means and so I really think this conference is Kind of something unique and something special. I've been to conferences all over the world and this one's quite different So thank you for inviting me and thank you for having me and thank you for listening for a few minutes As you heard my name is Chris Kelly. I work at github. I work in the outreach team So that means I mostly spend my time communicating with our community working on how to amplify their work and really so they're celebrating the things that they do I've got a lot of Awesome opportunity to do that. And so I really appreciate that and I want to thank github for sending me around the world to do that kind of stuff Today I want to talk about open source, but really I want to talk about it from a cultural perspective What does it mean when we talk about open source and broadly speaking outside of the world of software? To do that. Let's start back in 1968 1968 computer science was actually computer science. I think you know sometime in the 90s We kind of stopped doing computer science and doing some actual interesting things But in the 60s in the 70s, this is where like major innovation was happening in the in the world of computer science And most of us probably in this room haven't been born yet I wasn't and I really believe that it's important that we read the literature back from an era where we weren't born when This kind of work was being done and there's a really interesting people paper that came out in 1968 in a journal called the animation This influential paper was written by Melvin Conway and it's how do committees invent? What it was really about was what he was talking about specifically was that Software project management was sort of an emerging field and he was trying to figure out How do we decide what to build and what do we work on? But in the end 33 years later He actually was quoted that this became far more influential than he had ever anticipated when he wrote this It's a 3,500 word paper, and I think you should read the entire thing But the most important part comes about 3,000 words in and we're gonna jump to the bottom. It's two paragraphs from the back That's where his thesis ends up being and that thesis is what really kind of is this linchpin on the cultural Remifications of open source and the way we build groups So Conway he says Organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations It's a lot to unpack, right? It's a it's a lot of words, but what he's talking about is that communication systems Define the way our products are built our companies are built our societies are built our organizations are built the way we talk Defines the way things are built. That's what his general thesis is This is known as Conway's law and you might have heard of that you might hear through brought up I think it's one of the most influential thought pieces in computer science And so I definitely recommend reading the entire piece, but this is what's important to pay attention to so Conway's law is about communication and systems Anthropologists and linguists have been studying this phenomenon for quite some time Quite a bit of work has been established that our language actually impacts the way that we think the way that we understand the world There are tribes that have no notion of relative position. There's no left right up or down It's only north southeast and west that actually changes the way they see the world and the way they understand their world The way they describe it the way they think about it. There's my left leg is not a thing my southeast leg is That's an actual phenomenon So if we want to design systems broadly conceived and we're talking not just software We're talking about systems in general society's organizations teams What do we need to know about communication and how to define that what makes up communication? So tools define communication Does your company use email? Chat rooms do they have a lot of face-to-face meetings? What about instant messenger? How do these how do these tools impact the way you communicate if you're using? Chat rooms everything I say is seeable by everybody else in that chat room I I say things differently when I'm in that context if I'm doing something on instant messenger It's a silo that information is now trapped between me and that other person that can't be shared out just as easily Those things change the way we communicate. So tools are important to think about when we're defining the way we communicate Processes are another thing that defines communication Do you have to get approval to get work done? Do you need a do you have the authority your company to make decisions? Do you need to talk to a manager who then talks to another manager then talks to some other person who gets you information? And that information follows all the way back to you anyone related to this like have you felt this? How does that process change the way we communicate? Obviously, there's impedance there and changes the way we are structured the way we think about ourselves How does that make you think about the work you do the value of your work the things that you are that are important to you? Knowing the tools in the process is that impact our communication We need to be very intentional about them Communication is an emergent phenomenon humans just it emerged out of the way we we work together It comes natural to us But we forget that we don't all communicate the same There's distance time zones Cultures all of these things impact the way we can communicate. I have a distributed team I have people that are currently working in Asia and I have people that are working in Europe There's not a single time time during the day that I can have my entire team on a phone call And we can catch base. There's not a logical time that that can happen that somebody's not up at midnight to do this I can't communicate my my team within a face-to-face system. I have to use other modes And communication is hard and so we need to be very intentional about it We need to craft how we communicate and understand that's how we how it impacts the way we work How productive we are and? What we produce Open source is often held as this like panacea for how we can solve this communication problem Why don't we just all work like open source? I've heard that a lot today, right? You know it's and we're talking about open source impacting all kinds of fields economics to software So what are the promises of open source? There's really no canonical principles. We talked about four of them this morning, right? We talked about transparency Participation collaboration freedom of innovation. I call that one creativity These are the things we're searching when we're looking for designing a communication system when we're trying to adopt open source principles These are the sort of baseline that we want. We want these in our system And there's good reason for that. There's lots of great benefits that come out of the notions of open source Open source means that you're accountable everything you do Means everyone else can see it if I make a decision you can see the chain and how I made that decision I have to explain myself. I have to Show my my thought my work we can see who's responsible for it. That's a great thing Openness is about accountability You also have to take the time to explain I have to slow down if I'm explaining something to somebody in chat I need to add a bunch of context so they understand the high fidelity of face-to-face communication Allows me to transfer a lot of information very quickly But when I'm having to do it in chat or an email or any number of other Text mediums I have to slow down and I have to explain myself. I have to explain my thinking That's a really good thing Open also means anyone can participate Right, it's open There aren't hierarchies of people for participating. There's not approval processes open mean you can participate You want to contribute you contribute? You aren't supposed to be barriers for participation in openness Which means open provides more diversity since everyone can participate it means that Contributions and collaborations come from a wide range of people we get a nice diverse set But this is actually true These are the things that we talk about in ideals, but do they actually hold? Jessica talked about it earlier today If this is a if this slide is a hundred percent representation of all the software developers that yellow circle are the number of women developers and software 12.6% I'm gonna I'm gonna use statistics Jessica was shy about them these are every paper you read how we'll have a slightly different variation the ones I read say twelve point six percent of women That's one in twelve That's not very many One in one in twelve again, so one point five percent are open source developers That's a fraction of the number of software developers that are women participating in software in Open-source software that's shocking. That's shocking to me your face doesn't look like this when you hear those statistics I don't know this is what this is how I feel when I think about open source and getting more participation And there's a lot of different facets. I'm not saying gender is the only facet of diversity. This is an example of one There's race. There's nationality. There's language. There's age. There's background lots of other diversity I'm not saying gender is the only one this one. We have the most statistics on so it's the easiest to use as an example So These are all ideals and obviously our ideals aren't quite perfect. We don't have great diversity. We don't have great communication What are the realities of open source? In the real world ideals rarely hold up and let me tell you about some of the things I've seen in the real world Open source means anyone can participate. That's right. This is also a negative Donnie Birkholdt an analyst from Redmonk has a an awesome talk called assholes are ruining your project You gave that in in Fosn in 2013 Because anyone can participate that means assholes can participate and assholes are the usually the loudest participants Um Brandon Keepers who gave a talk this morning has also identified that the very notion of software development what we see in open source communities is that most software developers in those communities are White men who have a very thick skin And by white men, I mean privilege as Jessica mentioned as well as well. And by the way, Jessica I did not coordinate on these talks This white man privilege with thick skin is a you know, it's homogenous Homogene is the antithesis of diversity. We simply can't have it So by the way we communicate in open source it creates this homogene Open is a bizarre and not a cathedral You're probably familiar with Eric Raymond's Book the Cathedral in the Bazaar. Hopefully you are if you're interested in open source You should read it if you have not it's from 1999 He identifies open source as this bizarre. You're familiar with bazaars in Various countries. They're loud full of traders. They're negotiating. It's chaotic. It's a marketplace Now there's a lot of benefits to those notions But what happens in those marketplaces? Have you ever gone to? Marrakesh and gone through one of the bazaars and try to negotiate if you are not a loud confident Sort of aggressive person you're going to pay more for the things you're buying than the person that is loud and aggressive It's just the nature of it So are bazaars necessarily the things that we want or a cathedrals things that are structured and designed and long long lived Or those are things that we want Open can lead to a argument culture If you've ever read a comment on hacker news or you've been on reddit or basically any kind of thread on YouTube You've seen argument culture This is the sad reality of open source. It's it's definitely not what we want, but it is the truth argument culture is the primary thing hurting diversity in our industry right now and My thought has been Significantly impacted by a book called the argument culture by Deborah Tannen and also if you want to get introduced to this concept Kate Hettleston has a great blog post entitled argument cultures and the unregulated aggression So Arguments aren't about dominance. They're about dominance not ideas. You think arguments are about ideas We think in an ideal world that we have this sense of intellectual Darwinism Where we cut down the weak ideas and the best ideas flourish. Well, that's great in an ideal world But arguments aren't about that arguments are about dominance and power Sociologists have identified this phenomenon as the argument argumentative theory of reasoning Massive studies have been done to show that argumentation is actually has very little to do with logic and fact and has very much to do with emotion What we're concerned about in arguments is winning We get emotionally invested in our ideas and we want them to win no matter how true the irrefutable evidence comes What we want is to win no matter how true the facts are and The desire to win is an emotional response. This seems to be hardwired into our thinking Sociologists have studied this over and over and over again arguments are never about logic Are you insure about emotion in Think fast and slow Daniel Kahn man documents countless studies that show how people are not rational or objective in their thinking Their decision-making is driven by emotion and emotional loan. He shows that people are emotional beings when they make these decisions this is a problem this problem is because arguments have a dramatic effect on the quality of work and Productivity that our groups are arguments are detrimental to the work. We do arguments impede creativity great creative companies like Pixar and IDO have studied this and understand they've taken out argumentation in their ideation phases that you are not allowed to say negative things in brainstorming phases They've understood this and they are some of the most creative organizations in the world arguments impede diversity arguments mean the loudest or the most aggressive people win Like in open-source communities this creates a homogene All the loud people are the ones that win and they were ones that stay and now we've got this group of Very like-minded similar people all hanging around together Studies in biodiversity refute the fact that homogene is actually beneficial Diversity is the thing that is strengthens an environment Coral reefs are a great example that biodiversity is hugely dependent On their survival and once homogene sets in there's a high risk of collapse So we don't want Homogene in our systems. We want diversity arguments are actually particularly disadvantageous to women because an aggressive Woman is often perceived as pushy or bossy. We by very nature The way our cultures are designed we call that And I don't that's it shouldn't be that way Simply recognizing arguments as emotional means that we're pigeon-holing women into a negative component so Just the very nature of an argumentation culture where emotions are the drivers and not the facts in the logic make women at it women at a disadvantage So let's apply Conway's law to this reality Your communication structure will be your culture What is your communication structure look like when you have an argumentation culture? Do you have a culture that's creative? diverse flourishing or Do you have something else? Simply accepting the notion open-source principles are the solution to anything without consideration Means you get the good with the bad you do get the openness and the accountability But you also get the the the natural evolution towards argument culture Open is hard. It takes a lot more work than being closed It's a lot easier to have a face-to-face meeting and and get a solution solved and answer resolved Then it is to put it online and wait for an answer wait for the things you need It's easy to abandon openness when things get busier things get hard Open is a culture culture can be crafted we saw that We know that we can be intentional about the way we do things culture is a network of systems It's not a list of beliefs It's about having the right people the right values the right tools and the right processes Maybe open needs to be more like the cathedral and we give it credit. Maybe the bizarre isn't what we actually want Open requires collaboration You need a place that you can work together not in the silos where a single engineer or single person solves this problem and Sends it up the ladder and calls it a day. You need genuine collaboration where ideas can start at early They're accepted at their face value and that they give the space to flourish That's a github. We do pull requests very early so that we can start talking about these things very early Not before the idea is finished not when it's all cinched up and we found all the answers But was we know we need to collaborate? Open is about community. It's not a free-for-all. This isn't just some Marketplace of ideas. This is actually something that we need to think about and work on open needs people that care deeply and want to make it work Open requires communication. We need healthy communication not the argument culture It needs the ability to stop arguments and when they do happen how to resolve them because they are going to happen So make sure when you're designing your systems that you actually do put in place mechanisms to solve these problems Open requires being intentional What happens when you remove the argument culture? What does your communication change instead of saying no to everything? What if you approach communication with a disposition of yes? How can systems cultures and society be changed with open? Thank you