 All right, everybody, my name is Barton George. I'm here from Dell Technologies and I'm here to talk to you about driving innovation at a large company and the five lessons we learned, some of the hard way. Let us get going. So here's the same thing, just with our template. So how did it all began? So about eight years ago, we were looking at ways that Dell could raise its profiles with developers. Developers don't think of Dell as the first company that would come to mind. So what we started to do is try to brainstorm ways that we could up that image. And so we brought in Stephen O'Grady from Redmonk and we were tossing around some ideas and he said, you know what would be great is if you guys took one of your laptops, put Ubuntu on it, wrote the drivers and that would be a unique offering because nobody else is doing that. And we thought, that's an awesome idea but it'll never happen at Dell. And the reason is, is because the group that makes our clients or the laptops, the volumes that they deal in are just humongous. And to them, this would be a rounding error. In fact, someone told me, one of the BPs told me, yeah, that's what we sell in Belgium on Tuesday between three and four in the afternoon. So this was not something that they were willing to take on, so we didn't even ask. But serendipity struck and there was an innovation fund that was announced. So also during that time, you can tell the laptop got a lot of better looking with the XPS 13. So I thought, great, this is a way that I can go and if I can get the innovation fund, it will give me that sort of protection to go and work with the client group to get this launched. But I started thinking and I thought, man, I don't know anything about how laptops are made. And I don't know how to build a relevant business case to something like this. And I'm not even a developer. In fact, I failed at Fortran in high school. I did do all my business school essays in using the VI editor, but that's sort of where it begins and ends. So finally, somebody convinced me to go for it and I decided I would. And this is what I presented to them. I said, okay, wanna do a high-end developer offering. The idea is that we're gonna do this by out in the open and involving the community. And the main reason we're doing this is to change perceptions. So while we don't wanna lose money, it's more of a strategic play, as I said in the beginning, to get people to think of Dell more in the way of supplying to developers. These first two, these what and how, were pretty unusual for Dell. The high-end developer offering struck people as odd because even though we had been offering, I think we had over a hundred offerings with Ubuntu on our systems, those were on the low-end systems. And so people thought that made sense because you take a free OS, you put it on a low-end system, you get a low-end price point, and it's a perfect match. And yeah, that did work and we did sell a lot that way, but they didn't think that it would ever work on a high-end system. Why would somebody wanna buy a high-end system with a free OS? And then the other idea too, of this openly involving the community, that's just a different way. We always talk to customers, but that's usually you bring them in on NDA, you go out and visit them, and you learn that way. But luckily they deliberated after a month, they came back and said, okay, you get a little pot of money, $100,000, and they only end up giving me 40, and you have six months to see if this works, so go. Then the first thing I did, getting back to that whole thing about, I don't know anything about building workstations or coding or building business cases around something as nebulous as this, is I went and I formed a team. And so these folks were all people in the client group. Actually, the guy holding the beer at the bottom wasn't even in the client group, he was on the server group, but he wanted to get involved. So all of these people did this in their spare time. I still had to go and get permission from their bosses to let them do this, but it was more of a passion project for the team. So as I said, we did this all out in the open, but with the exception of the initial phase, which is the don't embarrass ourselves or don't look stupid. And the idea here is didn't wanna come out, announce something that would end up harming us more than it would do good, where people would end up saying, you know what, Dell totally doesn't get it. This kind of a thing makes no sense, they never can do it, boy, they don't even know what we're thinking. So to do that, first talk to a group of what we called our alpha cosmonauts for developers here in Austin, one actually who used to work for canonical, and they thought it was a decent idea. And then we went out to the West Coast, talked to a big search engine company, and then this company used to be a bookseller and presented the idea to them. And while they didn't say, yes, we'll take 50,000 of them, they said, that sounds like a decent idea. Why don't you come back to us when you have a little bit more, when you're a little further along. So that was all we needed to start working on the system itself. And so the main thing is that we work with canonical and the device driver manufacturer to get the device driver written. Oh, and by the way, if you're wondering why Sputnik, this is Mark Shuttleworth, the founder of Canonical and Ubuntu. And after he sold his company during the dot com boom, he paid $25 million to go up in space as the second space tourist. Of course, humans don't go up in Sputnik, they did have a dog, but someone like Mark wouldn't fit. And he went up in Soyuz, but that's not a cool name. So that's why the name Sputnik. All right. So we worked on the drivers, we got a lot of them done, we still had some key ones not done, for example, the touchpad driver. We were working with Cypress, but when we went out with the announcement, it wasn't available. And so that's why I talk about number two, being a rough ISO that we offered, but we were very open and honest about this is where it's at and then we're working on it and told them too that this is something, this is a project, if we get enough interest, maybe we'll turn this into a real product, but we can't make any requests. But what we'd like to know from you is what would you like to see in your perfect developer laptop? Please go over to this other platform and if you could just put what you want and then you can vote them up or down and we'll see what most people want. So we decided to launch this on my blog because there was a certain amount of distance that gave from Dell, although the whole thing, I wasn't hiding the fact that this was from Dell and I worked with Dell's PR team, et cetera. You can see by the number of daily views that I had, and this is pretty typical in general, leading up to the launch, we didn't go with my blog because it had a huge number of followers. So average of 59, when we went live with it, first day was 6,000, second day was 9,000, third day was 15,000. And we thought there'd be interest, but I had no idea it would be like this. There's an order of magnitude larger. So with that, we went forward. And here was how it netted out as far as what people were wanting in the way of a developer laptop. And this kind of shows the words that the, I think the number of views showed how much pent-up demand there was for something like this, this shows how low the bar really was. So first thing, don't make it more expensive than Windows, which would seem obvious because it's a free OS, but that being said, in the past we had offered this and because of discounting, et cetera, things got a little wonky and it did at points that it did look like it was more expensive than Windows when we offered on the lower system. So that's something we need to address wanted to work with it. Just the vanilla image, nothing fancy. You'll see we kind of ignored that and nobody cared, but we'll get back to that. Eight gigs of RAM, which at that time was decent. No Windows pre-installed. Once again, something you would expect on a Linux laptop, no CD or DVD. And then they did want support in hardware support that it would come with. That's our pro support. So then we were working on the drivers. We got the touchpad driver working and so we said, okay, let's do a beta program. We put out a note going all around the world saying we'd love for you to let us know if you want to be part of our beta program. So they filled out this whole long list. Once again, we thought we would get maybe 300 people. We ended up getting 6,000. And so that's where senior management said, okay, this makes sense. Let's move forward and make this a real product. But when we went to actually execute on the beta program, even though we had talked to operations beforehand and they said, yeah, we can send this out around the world. When we came back to them said, okay, we're ready to go. They said, yeah, we can't do that. So what we had to do is we had to admit that unfortunately we couldn't offer this worldwide and had to do this in the US. So not a message we wanted to deliver but had to come clean with that. And then, launch time. This is eight months after I got the green light, which is pretty fast at Dell and I would say at most manufacturers that you go from okay to launch itself. We did use the same hardware but getting the drivers written was no mean feat. And I should also add that when we worked to have the device drivers written, a very important part of this is that we take these drivers and have the device manufacturers push these up to the mainline kernel so that other people can use them, which also adds them enables people to run different distros other than Ubuntu on our systems. Okay, so how was it received? We got a lot of great press. I think one of my favorites was this from ours Technica about how our substantial investment paid off. And as you saw, we're a very small team. We were part-time. There was no substantial investment but it was really good to see that we cast a big shadow. So then helping to make this succeed was the ability to stay process light in certain places. So I would meet with my executive sponsor every two weeks and I would put together an MVP, a minimum viable presentation. So whereas more traditional products or projects you would have to spend a week putting together PowerPoint, putting together spreadsheets, numbers, et cetera, he really liked the fact that a half hour before we would meet, I would go and I would write out some of the key areas I wanted to talk about, some of the areas that I needed his help with. And this was just a great way for us to sit together and discuss this and for me to get his help. And as I said, this really helped advance the project because I was able to work on this, spend the time that it would have spent on the presentation working on the project. So that being said, on the flip side, because you work at a large company, you don't wanna ignore it. And so things like, for example, the website, we weren't gonna go and create our own website to sell these systems. It makes sense to sell it on the Dell website where you get millions of eyeballs. And then we had this grand vision which I alluded to before about how people wanted the vanilla image. We thought that, well, that may be what they want, but they really want something more. And we're gonna build this cool profile tool to bring down language stacks. So we're gonna build this cloud launcher and it's gonna be able to take applications, send them to the cloud. It's gonna be awesome. And then we found out that this kind of stuff is really hard to do. We never did go with do anything with the profile tool. And there was a company named Docker that came along and did this piece and sold it for us. So that worked out. I think the other thing too is we came out with a curated list of tools and utilities preloaded. And when we found that when we removed those, along with when we ended this, nobody really cared. They really did want just the vanilla image. And so here we were thinking we were adding extra value and people were fine without it. So one thing they did really want, though, was a powerful counterpart to the system that we launched with. So we launched with the XPS 13. They also wanted the beefy system. And this is a workstation class precision. And we kept hearing it until one day Jared said, okay, I'm gonna go and I'll get Ubuntu running in precision. So he did that in his own time, posted a blog and we thought, okay, then people know how to do this and that'll be happy. But that only sort of made them hungrier for a real system. So about a year later, it became a real system. We worked to produce it. And today, this is what the lineup looks like. We have the XPS 13, which we launched with and it's now eight years later on its 10th generation. And the precision mobile workstation developer line is now on its fifth generation and there's six different models. I only show two here, but there's six different versions. So we went from one system and one config to multiple lines and multiple configs. And here's just some of the feedback we've gotten recently. I love this one from Linus Torvald saying, hey, I don't usually name names, but he liked it so much that he ended up buying one for his daughter when she went off to college. Another one here is the upper right hand corner where someone said on Hacker News, okay, I didn't see that coming, Dell building a better MacBook Pro than Apple as regards the precision. And then this under the red is probably my favorite and it's because it shows what we were trying to do, this idea of changing perceptions. Well, I did something I never thought I would. I bought a brand new high-end Dell laptop for full price, me, hell will be freezing over soon. And then not just publications and blogs, but you get people of the community who will pick it up and they will run with it. So here is a review done on tech pills and he now has over 300,000 views. So this wasn't anything we orchestrated, but it was something that he came across and decided that he would do a walkthrough of it. So you get the community love. On the flip side, you also will get some angry people sometimes. And this is where community management comes in. So this gentleman here is based in Denmark. We announced that the precision system was available around the world. He couldn't find it on the webpage. That's Denmark. Ooh, I guess you don't ship these models to Denmark. Bad word, not the first time it happens. This is why I never buy Dell products, never ship the products and see a bigger profit. So obviously he was very frustrated. He heard it was supposed to be there and it wasn't. So what I did after walking around the block and cooling down, I wrote him back and said, hey, it should definitely be in Denmark. Send an email to our team to find out what's going on. Thanks for flagging this, stay tuned. And then he just said, that sounds really nice. Thanks for a quick answer. And I think that's one thing you find often in the community is if you listen to people and you speak with empathy, then they feel much better and they become oftentimes they go from renters to ravers. So that was a positive outcome. And also it helped us because we didn't know that it wasn't available in Denmark. That was some glitch on our side. So we were able to fix that. And then remember I said it wasn't about making money. Well, that is true, but it's always good to make some money, particularly that way you can keep the project going. So the original $40,000 has now made over, well, tens of millions of dollars. So that's really a positive thing, like I said, because it ensures that the program goes on. So with that, here's the five lessons that we learned, which I talked about in the beginning or I alluded to. So let's start with number one. And the first one, you're good enough. No one knows it all, build a great team and take the lead. And so this is what I said early on, but I think this is probably true for a lot of us. You hear about something and you think, well, I don't know about X, I don't know about Y, I don't know about Z and all those are needed to do something like this. So I'm just not gonna do it. And you find out that if you get other people that you can work with who do know this and you can be the person behind it helping to push it, it actually can work out. Number two, get a champion, be a champion. So as I said, it was actually, it was Michael Dell's EA who was my executive sponsor. And so that carries weight around the company, but at the same time, there's always people who say, I don't care if Michael's behind this or not. They're never gonna find me. I'm just not gonna work with this. I don't have to be part of this because you have to work throughout the system. It's not just the team that I'm working with. As I said, you've got to plug into the machine. And what would happen is every so often some new department would learn about it and say, why the heck are we supporting this? And then you'd have to deal with that. And so for the first two or three years, I had to sleep with one eye open just to make sure that when somebody did find out about this and try and kill it, that you were able to bring the right resources or reach out to them and get them to feel a bit better about it. Number three, leverage execute. So as I mentioned in the beginning, this wasn't my idea. This was Stephen O'Grady's idea from Red Monk. And that really doesn't matter in that if you can leverage it and if you can execute it, then you can have a winning product or a project or idea because just an idea by itself really doesn't do much. You have to take that. Doesn't matter if it's not yours, but you have to take that and then make that happen. Start small. And this was something that I'm glad people convinced me of because when we originally were gonna launch this, I wanted to have both the, as I said, the XPS 13 plus the workstations and multiple configs, and et cetera, et cetera. And that's where we got to, but one of the main reasons is because we started small and we went incrementally. Also, not to over promise. And I think that's one thing we were careful about from the beginning that when we announced it, we said, hey, this is a project. We don't know if it's become a real product or not, but hey, maybe it will if you can give us some great input. And then last, but not least, communicate, communicate, communicate. So constant contact with the community. Probably the most ways that I communicate is via Twitter, although they'll post on my blog as well as forums. When you do communicate, as I said with the gentleman in Denmark, you wanna speak directly and you wanna speak with empathy. Particularly that may not be your first thought, but if you cool down and you do that, it ends up being a win-win. And then when you don't deliver or you screw up, you need to be able to come out and admit that right from the beginning. And that will give you more credibility in the community. So that's the Sabatnik story, the five lessons. This is the place that you can reach me as well as where we have our Linux systems. And from the next trick, I'm now in the Corporate Strategy Group and we're looking at how do we put a larger developer strategy all across Dell Technologies. So with that, I think I've got four minutes. I think there might be, okay, we've got some questions in here. And I think the question is about why if you get a, if you buy a XPS 13 with Windows and then you add Ubuntu on it, they won't be, oh wait, sorry, that the images aren't available if you wanted to get that. And that just unfortunately happens to be the way that the contract is written between ourselves and Canonical. So it would be great if folks could get that image, but as I said, that's just the way that it was structured. The other question, so this is about the idea of backporting. In other words, if you have an older system, would you then think about taking, so what are we, 2004 right now, taking that to one, say three generations ago? And I think the answer there is because our team is so, what shall I say, scrappy, no, there's, we've got a great team, we just, but we don't, we're not big. And so we don't wanna overload ourselves with something that will deliver a subpar experience to our customers. And so we'd rather focus on keeping the systems going that we have. And we don't usually do this, but the last couple of times we have upgraded a current system with a newer version. So the XPS 13, the most recent one came out with 1804 and we only do the LTSs. And we were able then to recertify for 2004. And that just came out a couple of weeks ago. And we did that once, one time before. It just depends on when the systems come out and when the versions come out as well. So I think that's, let me just scroll, scroll, scroll. Oh, gosh, there are more. Like about a Chrome-based product, Acer is the only choice. I've been using it for some time and no longer own a PC. There are always things that we're looking for. And we do have Chrome-based offerings, but not Linux-based at this point. We do have Chromebooks as well as that enterprise-type Chromebook that we created. Okay. All right. They're going to quickly here. I saw the one from someone in South Africa. That's great. I think you buy it through resellers down there. Would be the way to do that as opposed directly from Dell. But with Mark, shuttle worth being from South Africa, you really need to have them there. Let's see. I'm almost out of time. I got one minute. Quick, quick. Oh, the lack of the fingerprint sensor support in Dell. It is, we've been working on that and it's just not supported at a place yet where we feel comfortable offering it. And it should be out soon that we'll be offering it as an OTA for folks to upgrade the systems to that. So let's see. I may be out of time. How do I motivate myself and the team? That's a tough one. I think sometimes you get really beaten down, but the good thing is that the folks on the team were motivated to begin with. It was self-selecting. Like I said, the gentleman who was in our enterprise server team, even though he didn't work for the team, he went and he worked on the laptops. But it's being a champion. It's bringing, when you do hear good things that are written about the system, you bring it to the team and say, look at this, look at what your work has done. And then I know in the beginning, I didn't do as good a job of that, sending it over to our engineers and saying, check this out, look at what your work is being, it's being valued. And then with regards to myself, it's just sometimes you feel like you're gonna give it up, but for one reason or another, you keep going. As my manager said in my review one time, it's like pushing a car made of mud up a hill. You really have to sort of stick to it. But there was a lot of great feedback and I really enjoy interacting with the community. I don't get to do it as much anymore, but that's the one-to-one contact with customers is something that I personally really enjoy. Let's see, is there anything? Arm-based PCs, we're always considering everything. Let's see. Anything else as I scroll, I scroll. I might have gotten them all. And I'm assuming we're now moving over to the Slack channel, but I need to make sure that I can get that set up on my end. Or do we have folks on the production side, do we have time to keep going on this one? I will assume until I stop getting questions that we can. Okay, we need to move Slack. Okay, so I'm gonna go and try and get that set up. So bear with me and I hopefully will be over there in just a little bit if you're still interested in talking more. Thank you.