DevOps in a Microsoft World with Jessica DeVita and Jeffrey Snover

Matty: [00:00:08] Welcome to Arrested DevOps, episode 30, DevOps in a Microsoft World. I’m your co-host, Matt Stratton, @MattStratton on Twitter.

Trevor: I’m your co-host, Trevor Hess, @TrevorGHess on Twitter.

Matty: Arrested DevOps is brought to you by TenthMagnitude, a cloud services firm that is, as luck would have it, working with Microsoft Azure, and they dig DevOps. And 10th Magnitude is hiring. Learn more at arresteddevops.com/10thmagnitude. This episode is also sponsored by PagerDuty. PagerDuty eliminates the noise, chaos, and manual processes across the entire incident lifecycle to decrease resolution time. PagerDuty is trusted by companies like Etsy, Nike, and GitHub. To sign up for a free 14-day trial, visit arresteddevops.com/pagerduty. It’s all— we are also brought to you tonight by Datadog, a monitoring service for scaling cloud infrastructures that bridges together data from servers, databases, apps, and other tools. Datadog provides dev and ops teams with insights from their cloud environments that keep applications running smoothly. Datadog integrates with a variety of Microsoft products like Azure, IIS, SQL Server, and Windows, and is available for a 14-day free trial at arresteddevops.com. So we have 2 great guests joining us today. First, it’s my pleasure to introduce Jeffrey Snover, a distinguished engineer at Microsoft and the lead architect for the Windows Server and System Center division. Jeffrey, can you tell us a bit about your background with this whole DevOps thing?

Jeffrey: [00:01:42] Yeah, so I’ve been doing management for quite a long time, the majority of my 30-plus years in the industry, and at some point I heard about DevOps and was lucky enough to find John Willis’s podcast and, you know, him and Matt and Damon Williams, excuse me, I just consumed that stuff up. And then I found Gene’s book and just fell in love with that. And, you know, a lot of this was just stuff we had been doing, but they gave it a voice and I just fell in love with it. So anyway, that’s how I got involved with it. And, you know, I think if you take a look at, you know, I started PowerShell and I wrote the Monad Manifesto in 2002, and it just felt like, it was a bunch of the things I’d been talking about there were finding a voice other places in the industry. Let me be clear, I’m not trying to take credit for DevOps. I’m just saying we’re very, very akin. We think, you know, great minds think alike. I was just ecstatic with that. You know, the idea of automating everything, the Taguchi method of quality control where you repeat things, make small increments. If you like the change, keep doing it. If you don’t like the change, back out of it. Just wonderful stuff. Actually, by the way, a lot of this stuff hearkens back to the early quality days. So in the 1980s, I did a lot of, you know, manufacturing. So I worked at Storage Technology where we did disk drives. And so there’s a— that was when the, you know, the Japanese quality revolution was in full swing. And so a lot of it was taking those concepts and applying it to IT.

Trevor: [00:03:23] Thank you for joining us. We’re also joined by Jessica DeVita, DevOps Evangelist at Microsoft. Jessica, what is your role with DevOps?

Jessica: That’s a good question. I’m a Technical Evangelist at Microsoft, which is a really fun, a really fun job. I get to bring the good news of DevOps, right? That’s what an evangelist does. And I, much like Jeffrey, really kind of was amazed at how articulate people like Gene Kim and all of these, John Willis, all of the greats. We all stand on shoulders of giants. How articulate they make the story of DevOps and what it does for IT people. And so hearing Gene talk about how he wanted to improve the lives of 1 million IT people, just really, just blew, I want to be a part of that in any way. So as a technical evangelist at Microsoft, we do fun things like IT camps and during the lunchtime, We spend a little bit of time talking about configuration management. We talk about version control and all the kind of topics that are, you know, maybe new to traditional enterprise IT folks. And so really, I feel like my goal is to just help them kind of start and see where, you know, see where they can end up with it, if it’s something they can put to work in their organization. And a lot of times these things start off very small with one person hearing, you know, the, the potential potential of this work and going, yeah, I want to try that. So that’s kind of what I do at Microsoft.

Trevor: [00:04:59] Great.

Matty: I just want to repeat because Trevor turned into a robot right when he was saying your name. So that was Jessica DeVita, who is also joining us, not brrr brrr brrr brrr brrr brrr brrr, which is how she was introduced.

Jeffrey: So—

Trevor: Leave me alone. I like my bleep bloops, okay?

Matty: Yes. So I have a question. As someone who spent a fair amount of my career as a Microsoft sysadmin, or as known in the Microsoft world as an IT pro, is kind of the way that that’s approached in Microsoft lingo. What are some of the challenges that you see that traditional IT pros face when looking at moving to a more DevOps approach?

Jessica: I am in the field, and so that is definitely a unique perspective. I’m sort of like on the you know, the front end, the front lines, if you will, hearing what IT pros are dealing with on a day-to-day basis, and they’re frustrated, and rightly so. You know, they’ve, they’ve been promised a lot of solutions that, that haven’t worked, I think, in, in some cases. And so I think they’re frustrated, and a little bit hard for them to tell, you know, the wheat from the chaff in terms of what tools are really going to make their lives easier. So that’s a big part of it. The other part of it is, you know, cultural challenges, which you guys have talked about on the show extensively, but, you know, merits as much discussion as possible is how, you know, what are the work environments like of these IT pros? And, you know, are these generative environments that are really supportive of, you know, the human, you know, in front of that console, in front of that command line? You know, so those are some of the challenges I see.

Jeffrey: [00:06:42] Yeah, for me, what I see is, you know, first the traditional DevOps challenges, right? IT pros who are in charge of making things run and the wall between the developers and the operators, but that’s kind of well-trodden territory. I see that a lot in our world. However, I think especially in the Microsoft world, what we see is because Microsoft has such a strong history of great GUI tools, you know, everything’s greatest strength is also their greatest weakness. And so we have this great strength in GUI tools. You walk up to a machine and it’s, oh, it’s just so easy to use, as opposed to the traditional Unix experience of you walk up to a machine, you see a prompt, and you say, what do I do? And so we have a very deep bench of people who are very good at using GUI tools, but that really works against the DevOps mindset of, I want to automate something, I want a repeatable process, I want to join a community and contribute to that community and use the products of that community, it’s very hard to share a bunch of mouse clicks. If you take a look at some of the deployment guides in our world, you see these incredibly large books and you’re like, well, what’s in that book? That’s a very scary thing. What you see is a bunch of screen captures that says click here and click here and click here. Then when you replace those books with, oh, here are the PowerShell cmdlets, it’s this tiny little page and a half, thing. Yeah. So anyway, then, then we have the challenge of, oh, can everybody do that? You know, oh, a command line interface, I’m frightened of that. You know, I, I’m really used to just clicking next and that having working out for me. So there’s really this cultural challenge. And, you know, some people just love it, you know, the, they, they get it, they adopt it, they’re having a blast. And other people are like frightened, like, I don’t want to change, you know, I learned my skill a while ago, and I was sort of hoping I just like be done with learning, right? Like if you want to be done with learning, you should get into like forestry, right? Like my best friend’s in lumber. I think what a great job because like the guy learned his trees like way back when, and when was the last time a new tree came out, right? I don’t think there have been many recently. Anyway, so if you don’t want to learn anything new, get into the lumber business. But if you’re in the IT business, it is a, it is a You’re riding the tiger, man. It’s constant renewal. You gotta constantly learn new stuff.

Matty: [00:09:17] That’s, yep, that’s one of the Trevorisms is always be learning. It’s like a shark. Gotta keep moving forward.

Trevor: Bridget gets credit for the shark though.

Matty: Yeah, that’s true. That’s true. All right, well, that being said, this makes it easy. Trevor, why don’t you go ahead and move to the next question?

Trevor: So Microsoft recently has been making a lot of moves and it would— to adopt and integrate with more open source projects and products. And especially, you know, so much is changing philosophy to embrace open source software instead of kind of fight. To quote one of my favorite people, what up with that? Which he probably quoted from somewhere else, but I’m gonna just say it was his and say, what up with that?

Jeffrey: Yeah, you know, I like to say that, you know, organizations are like gymnastics. The body follows the head, right? So we changed heads. We got a new leader and the body’s following the head. New leader, new approach. Satya is just a wonderful CEO. I mean, he really is. I just cannot tell you how enthusiastic and optimistic I am. And part of it is a very fresh approach. It’s like, hey, let’s go do what it takes to make our customers successful. Let’s make sure our number one thing is to make sure people use our products and are successful with them. And you know what, we got smart guys that can figure out how to turn that into money. Don’t let that be your focus. Just make sure things are doing the right thing. So too, he takes a very fresh approach with open source. Hey, let’s go meet our customers where they are. Let’s take advantage. Let’s contribute. Let’s be a member of the community. And for a lot of us that have been trying to do that for years, honestly, you know, wasn’t really a conducive, friendly environment to have those types of conversations. And now it very much is. It’s amazing how sometimes, you know, it takes forever to turn the ship, and then sometimes it could just turn on a dime. And this is one of those examples. Now, I will say there’s still a lot of things, steps you go through, et cetera, but the attitude really has just turned on a dime, and it’s wonderful.

Jessica: [00:11:31] I have to say that is absolutely the case. You know, it’s really fun to have joined, I joined Microsoft just probably a couple months after Satya became CEO. And, you know, it was really exciting to get behind a leader, and I obviously, you know, loved and supported Microsoft before then. But it’s definitely pretty amazing to see the type of welcoming. I mean, I’ve, you know, I mean, I was in some of these communities before I joined Microsoft, so you all knew me before I joined Microsoft. But even after that, I felt just really welcomed. And, you know, the open source community has a lot of great cultural things to teach the enterprise, I think. And so it’s really fun for me to be able to spend time in both communities, in both worlds.

Jeffrey: You know, sometimes I get the question, how real is this? You know, like, was it a few months ago, maybe a month and a half ago, Satya did his cloud announcement and he had this big slide. He says, let me just put something to rest. Microsoft loves Linux, and he had this big logo.

Jessica: [00:12:41] Yeah, the big sign is up there.

Jeffrey: Exactly. Then people are like, well, is that real? I mean, what’s going on with that? The thing I like to explain to people to help them get their heads around this is think about Azure. In an Azure context, we make more money if you have 10 instances of Linux than if you have 2 instances of Windows. In Azure, we just want you to be successful, and so we’re going to offer you lots of services that are available, like all these great Azure services. Well, how are you making those available? And the answer is, by and large, they’re available through REST APIs that you can consume through Windows or Linux or Android or macOS or whatever you want. So it’s really this nice refreshing mindset, right? Like what I tell people, I said, the great thing that’s happening here is Microsoft’s becoming a software company. They say, well, we were always a software company. I said, no, we were a Windows software company. We’re becoming a no-adjective software company. And it’s really quite exciting. Exciting. But still, you know, I’m lead architect for Windows. I’m going to do great innovation in Windows. We are, you know, increasing our investment in Windows, but we’re also increasing our investment in Linux and making Windows great for Linux and having great interoperability. It’s, you know, again, I think, you know, I’ve got a lot of deep background and most of my team has a lot of deep background in Linux. So it’s just, wonderful for us.

Trevor: [00:14:09] I’m curious, so you mentioned that the, you know, the body follows the head. Were there people who were kind of resisting the open source movement? Were there people who were really, you know, dedicated to the tools that exist already and not willing to kind of accept the embracement or the embracing behavior of Satya’s vision?

Jeffrey: Certainly the previous CEO, Steve, you know, he called— he was not a fan of open source, right? He couldn’t quite get his head around it. Comments about cancer and things like that is like a nationwide commercial or something. I don’t know. But no, what was the question? Sorry.

Matty: I, I think what the question— and I’ll translate for Trevor, um, reading between the lines— I think it was just generally saying like, like you, you sort of alluded to it a little bit by saying the previous CEO, but But what’s the, you know, without, you know, telling tales out of school, but just kind of— and I have my kind of experiences with people I’ve worked with in Microsoft with seeing these, these changes, and I have my theories about that, which I’ll tell you in a minute. I wonder what you think about, like, what’s the acceptance level of that? Like, you know, there’s the acceptance of the community, of the Windows professionals, the people who are Microsoft customers, but within Microsoft, You said, you know, like Trevor said, the body follows the head. So how much of the body, like, is it— you getting that momentum, or is there still a lot more of— sorry for people listening, you’re not seeing my head motion.

Jessica: [00:15:42] And the—

Trevor: well, and then the other piece of that is those that didn’t follow, what did you do to help? Or those who followed later, what did you do to help them?

Jeffrey: Oh, well, it’s—

Matty: that’s a better question.

Jeffrey: Yeah, it’s very important to point out this is the beginning of a long journey, right? And so, you know, we have been doing some open source stuff for quite a while, right? OMI, the Open Management Interface, that’s where we took WMI and made available to everybody. And then Desired State Configuration, I’m doing that as a heterogeneous open platform. So, you know, a bunch of my teams have been on that page and I’ve publicly predicted some other open source stuff that’ll come down. And honestly, I think most people, I mean, when you explain things, right, like the example I gave about Azure, and I explained that a lot to people because, yeah, it’s not intuitively obvious. Like, wait a second, I just spent 10 years under a guy who told me this was like evil incarnate, and so now you’re telling me this is good. Like, what was that? Connect those dots for me. But you just explain it, and it’s like, oh yeah, you know, we want to be our customer’s software company. We want to meet our customers’ needs. We want to meet them where they are, and if they want to have this and if they want to have that, that’s fine. We can still help them be successful and make money from it. Again, Azure context, the fact that when you use our Azure services, they’re REST APIs, and, and that’s all goodness. And most people get their heads around it, and then it’s really just a question of, oh, okay, well, right now here’s all the things on my plate. Next cycle, we’ll talk about doing some of that or we’ll explore that. Really, that’s the only reticence I’ve seen.

Jessica: [00:17:25] Jeffrey, just a willingness to acknowledge that our customers are in different places. A lot of times we want to believe that people use the Microsoft stack fully and completely and nothing else. We know that that’s not the case, and I just love the willingness to recognize that. As you said, whatever you’re running, we can probably host it for you. So, I mean, didn’t we, didn’t we like stream the Super Bowl yesterday? I mean, there’s this, that recognition I think is pretty, pretty exciting.

Jeffrey: I was just gonna point out that one of the things I just absolutely love about this company is that we’re incapable of sustained error, which is important because we screw up all the time. But everybody screws up all the time. But the big difference is that Microsoft, we’re just very, very brutal with each other. Hey, you’re screwing up.

Trevor: Hey, I screwed up.

Jeffrey: Hey, am I screwing up? And we’re very open to those things, like screwing up is not a, okay, let’s fire this guy. It’s like, hey, you screwed up and I needed to— oh, okay. And so we have this fantastic immune system that makes us incapable of sustained error. We have the courage to admit when we’ve made mistakes and then the willingness to go fix them. And with that, you know, honestly, it’s kind of hard to— there’s no challenge that you can’t address with that attitude.

Matty: [00:18:51] For people who missed it, we talked about this on our last episode on blamelessness. No, 2 episodes ago.

Jessica: That was 2 episodes ago.

Matty: Episode 28. If you go to arresteddevops.com/28, when we talked to Mike Rembezi and Dave Zwieback about blamelessness and that idea. Jeffrey, to your point of saying, like you said, we’re going to screw up, right? It’s the idea of not how do we prevent it. We know what’s going to happen. We plan for that. And then know how we’re going to, to get smarter about it and, and be able, be able to pivot. So kind of, you, you alluded a little bit, Jeffrey, you started talking a little bit about DSC and, like, so, you know, you knew I was gonna start talking about configuration management because that’s my jam. But so as someone, as I always tell people, like, when I’m wearing my— it’s funny I’m saying when I’m wearing my work hat, and if you look at my avatar, it is my work hat. I introduce people and I say I’ve been doing Windows sysadmin for, you know, some 20 years, and I’m like, every single way Microsoft has told me to configure Windows servers, I have tried. And so it’s traditionally been challenging, and I’d like to talk a little bit about kind of the vision for aligning Windows. And I’m not— I know this is teeing up a little bit to probably— I know a little bit the answer, but kind of the next generation of CM that the OSS world uses How does Windows get aligned to that? Not even so much from a— just from a philosophy standpoint as well.

Jeffrey: [00:20:18] Yeah, so the first thing is, let’s see how to start with this. The first question would be, hey, why aren’t you just using the existing tools that are out there, the great tools?

Matty: Why am I not using ConfigMgr? Why am I not using SCCM to configure my Windows servers? And that can be a whole other episode if you want to talk about it sometime.

Jeffrey: Yeah, actually, that one. Didn’t think you were going there. I thought you’d say chef and puppet. I was going to answer the chef.

Matty: Okay, I thought you’re saying why aren’t you using the existing tool like within the Microsoft thing? Because I was going to say that’s the— well, that’s a question that we get a lot. And I’m just saying just in general, I see a lot of— and like Steve Murawski wrote a really interesting blog post about DSC and it went on the Reddit, the sysadmin subreddit, and they’re like, I’m so sick of hearing about how there’s no config management in Windows. We have GPO and SCCM.

Jeffrey: It’s true. Let’s be clear. It’s true. We have those things, full stop. Then the question is, well, what are the attributes of those things? And the answer is that they’re really good for things like enterprise client management, okay? And it turns out they’re not so good for data center management, okay? So in particular, I remember the day I had the conversation with the guy. I’m trying to remember which server it was. I think it was like passport.com, right? So this is going back a ways, but you know, the Passport service or the Live ID service. And he’s telling me like, yeah, I got all these servers and I try and configure them with Group Policy and I have no idea whether or not, you know, where they are because I like set a policy and I say make it so, and then I can’t tell who’s got it and who didn’t, where it worked, where it didn’t. That’s not— I need control. I need to be able to say you now, then you now. I need to be able to say, I need everybody within this time window. And the reality is that Group Policy didn’t give us that. It had the wrong semantics. It was very much focused in on— and SCCM, very similar— very much focused in on sort of the enterprise client. I have hundreds of thousands of admins out there, or server clients out there, and over the course of time, I need them all to realize that this is the proxy setting. Okay, so with Desired State Configuration, I’m super clear about this, we are focused in on a DevOps way to manage and configure servers. Now if that’s also helpful for clients, yay, but that’s not the criteria for success, and I won’t screw up the architecture trying to address those scenarios. So part of this, by the way, there’s a funny story. At some point many, many years ago, one of the executives came to me and said, Jeffrey, you need to solve configuration management for servers. And I said, yeah, no, go away. What do you mean? You’ve got to do this. You’re a senior guy. It’s like, yeah, it’s impossible. It’s never going to happen. And so then the question is, well, how am I doing this? And the point was that at the time, everybody was in this— every group, they felt themselves to be their own CTO. And so they did management their way. I do configuration my way. I’m going to do it this way. I’ve got my technology, my technology, my technology. And it was also an enterprise mindset. Now, when I say an enterprise mindset, I really— it’s probably not the right term, but there’s a certain gestalt I’d like to compare versus DevOps. In DevOps, it’s about simplicity, right? I just love— I just cannot tell you how many times I repeat the DevOps motto of the problem is that scale times complexity exceeds our skill set. If you want to be successful and you want to scale, it has to be simple, simple, simple. It was really when I took a look at the cloud that I said, hey, the needs of the cloud allow me to design an architecture that I can go force change across the industry, force across Microsoft and then across the industry. I can say, we need something different and it needs to be super simple. Now, if you want to play ball, you play ball. Someone said, you can’t tell SQL how to do configuration, and the answer is, Yeah, I’m gonna tell SQL to do configuration. They can like it or not, and they can do it or not, but if they’re super happy with what they have, fine, go with God. But if you wanna be DevOps, you need to do very simple configuration along the lines of desired state configuration. So that’s really how I got into it. Now from there, what I did was I said, okay, what people really need is configuration across all the elements of the data center. Again, we’re kind of shifting from Windows, Windows, Windows being a software company, in the data center, you need to configure Windows, Linux, routers, storage, you need to configure everything. So I needed a configuration management platform. And then the next thing was, you know, we wanted to take very much of a platform play. Microsoft has benefited greatly when our partners benefit greatly. Okay. And so what we wanted to do is we wanted to say, hey, how can I develop this platform so that people can build on top of it? Now, at this point, people typically ask, well, why not just use Chef or why not just use Puppet? They’ve got great products in the space. And the answer is yep, they sort of don’t work for Windows. And this is kind of the core architectural difference between Windows and Linux. And I know this because I stepped on this rake myself. When I came to the company, I wanted to just take the Unix tools. I mentioned to you, right, a deep Unix background. My team was deep Unix background. We said, you know, let’s just go get Bash and awk, grep, sed. Let’s get all that. Let’s put it on Windows and it’s going to be great. We got the problem solved. We can go do something else. Else. And we did, but, you know, it came that close to getting it included in Windows, and in the end I failed. It’s like because of IP concerns, but we made it a free download. And it turned out it just didn’t help at all, right? Because Linux and Unix is a document-oriented operating system. Everything’s a document. So awk, grep, sed, these are document editing tools. That’s why they’re management tools. On Windows, I don’t have any interesting documents. It’s all APIs, and awk didn’t work against the registry, sed didn’t work against WMI, grep didn’t work against Active Directory. So none of these tools helped, and that’s what happens. Now, it turns out there’s a false start here. I have a very— one, one very important workload which is document-oriented, and that’s IIS. So what happens is these Unix management tools come onto Windows, and they can do a good job managing IIS, and they think they can manage Windows, and then they ask questions like, hey, where’s the configuration? Configuration file for 80 users. It’s like, yeah, we got to have a talk. That’s not going to work.

Matty: [00:27:09] There’s a reason that the IIS Cookbook is so good.

Jeffrey: Exactly. So anyway, so I wanted to develop a platform so that these tools could then interact with Windows and do great stuff on top of it.

Matty: Absolutely. And then one of the challenges that I’ve run into, because I’ve had this kind of, again, the why don’t I just use Conman and why don’t I just use is again thinking about, like you said, being towards DevOps, you know, kind of some of this DevOps mentality, or just even modern configuration management of systems, is at the end of the day, right, Configuration Manager, you know, SCCM, it’s a big database, and you can’t version pieces of a database. I mean, databases are hard, right? I’m a sysadmin, so I don’t have to deal with them. And so when I think about being able to treat my infrastructure as code, I can’t do that if all the the details of my infrastructure and its configuration are stored as data that works in that way. Then it’s also harder to test. The other thing I think about in the reality of the world that I’ve lived in, again, like you said, SCM has worked super awesome for enterprise clients. Then what do you have? You have teams of people in a company who have the skill set of writing amazing SCCM packages, and that’s all they do. ’Cause it’s not easy. And so what that means— or it’s a very specialized kind of task. And that makes it really hard when I think about, you know, one of the advantages of tools like DSC or, you know, where I work and all that stuff, right? Kind of it’s this idea that it’s a common— it’s a lingua franca, right, between your devs and your ops, and everybody can communicate around how they want their configuration using the same language. It’s a lot easier for someone to pick up the basics, like 80% of Chef or DSC or Puppet or anything that they need very quickly for the, you know, be able to say like, I’m just gonna throw this little piece into this to express what I need. You can’t be like, I’m gonna put in this little piece of this package kind of thing. So I think that’s a big— or again, GPO, heaven forbid, right? I mean, like, that’s, you know, how do you test GPO? Oh my God. I mean, now I’m gonna— we’re gonna get like all these tweets telling me why I’m wrong because I haven’t managed an AD in 7 years or 8 years or whatever. I’m sure there’s things I don’t know, but still, it just— I would be always scared of doing anything because I wouldn’t want to change it. So I want to kind of think a little bit about another thing. So we’ve been talking about kind of configuration. Another one of the pieces too, when we think about in DevOps, we think about measurement. Like, so if we’re going to talk about, you know, John and Damon, we got to talk about CAMS, right? Culture, Automation, Measurement, and Sharing. So we talk We talked about automation a little bit. So from a measurement thing, so when we’re talking about measuring our systems or thinking about that way, what are, in a DevOps way, how do we do that in the Microsoft world? You know, what are ways we can do that to be able to gather this type of information in a world like that? And like Jessica, what are you seeing people doing, you know, in the real world?

Jessica: [00:30:14] Well, so that’s the biggest challenge, right? So we’ve got this, you know, whole group of sysadmins, Windows sysadmins, who have been isolated from the Nix folks, and they don’t— they just don’t have the command line skills. And that’s a generalization, of course, you know, very much a generalization. But, and so we’ve been living in that GUI world where I click next, next, next, and I remember, oh, I remember where that button is. You know, I can’t— let me get in front of it, and then I can, you know, get this solved. And as you said, you can’t version that stuff. It’s not something that’s exactly equivalent that you can go and change in a script per se. And, you know, you guys— Jeffrey can speak to this more in detail. But until there’s sort of that— and I think we’re getting there, right? When I go and make a configuration change in Server 2012 R2 and I get to the end, before the Finish button, there’s a whole line of the PowerShell. So I’m going to grab that and look it over. And more often than not, I’m going to run that instead. I’m going to copy and paste that because I can then start to learn, you know, what— okay, if I’m going to do this in the GUI, now I know how to do it in PowerShell. So that kind of stuff is super helpful, but we need more of that. And, you know, Jeffrey, I think you could hopefully talk to this a little bit, but more of that kind of let’s meet them where they are. So these guys and gals are— they’re in the GUI, so we have to take that PowerShell to them. We have to bring them those tools where they are so it’s a little bit easier to to get them. And that’s just one, you know, one idea, one thought.

Trevor: [00:31:57] Do you have any suggestions for places that are kind of stuck in the past, so to speak? And, you know, they’re still on— not for long, but some of them still on 2003 and 2008, 2008 R2. How can we help them learn PowerShell?

Jessica: You should just already love PowerShell. That shouldn’t be the argument. If we’re having I mean, if we’re having that discussion, then I don’t know, that’s like a different podcast.

Trevor: That’s fair, but I still—

Jessica: Sorry, I didn’t mean to—

Trevor: Oh no, no, I meant more along the lines of getting more comfortable with PowerShell. It’s not that— I’m not seeing that people don’t like the idea of PowerShell, they don’t know how to use it.

Matty: I would say, okay, so I’d like to hear what Jeffrey says, but this is my thing that I always tell people is it’s called PowerShell, not PowerScript. And so I try to get people used to the idea of that it’s, you know, it’s, it’s not a scripting language, right? I mean, yes, you can write scripts in it, but it’s the same thing about Bash. Bash is not like a script. It’s not Ruby, right? It’s, it’s the way you interact with the system. And, and Trevor, if you think back to— I’ve been using this analogy a lot— when we had our Git episode And Emma said that the reason she doesn’t like GUIs for Git is because it’s hard to explain to somebody what to do, where it’s much easier to say, type these words. So I think what you— to get people used to is don’t start by saying, oh, you’re going to write all this automation and write these scripts that are going to make your life better. It’s more getting used to the thing of saying, hey, you know what? Even in a way of what you’ll tell that, like, when you’re working with people, maybe the answer isn’t like, okay, well, now go open up this thing and do this thing. Say, hey, crack open a PowerShell prompt for me and type this in. And then they’ll— I think people start to see like, oh boy, that was super easy, you know. And this is because I think there’s this feeling that it’s about writing scripts. I don’t know, that’s been my impression. So am I off base there, Jeffrey, about your baby?

Jeffrey: [00:34:08] Well, you know, I think there’s lots of great starting tools out there. Don Jones has a fantastic book, PowerShell in a Month of Lunches. I did a Microsoft MVA. It’s in your check-it-out links at the end of the show. It’s like one of the most successful 2-day trainings we’ve ever had. And I’d say, you know, kind of step by step, people get it or they don’t. And what I would say is I would have a conversation with the managers, and I just say, hey, look at the future. Look at what automation gets you, look what lack of automation gets you, and I think year by year you ought to promote and reward the people who are giving you the IT that you want. So if you want unpredictable, unleveraged IT which is sort of random, continue to reward the people who click the next button. But if you want to make your environment one that’s repeatable, one that’s automatable, one where you can communicate, be part of a community and leverage that community and get better and better each year, then when it comes time and you got a budget to hand out promotions and a budget to hand out bonuses, you should be handing it out to the guys who are moving you in the direction you want. More and more I hear people that that’s actually happening. I’ll just point out to the other people that aren’t on that page, that aren’t getting those bonuses, Just do the right thing. You’ll look at it, just be clear. If you’re, if you’re going to retire in the next 3 years, like, forget it, you know, just click next and learn, you know, practice Bridge, right? But if you’re going to be in this industry more than 3 years, you really want to up your skills. And managers, you’re not doing them any help by continuing to tell them how important they are and that they’re valuable. They were valuable. That’s different than saying they’re part of your future. So reward the people who are part of your future.

Matty: [00:36:09] I think about, because I know we’ve had this conversation too about the click-next and this sort of the conversation about like, you know, people say, well, there’s always going to be, you know, jobs for people that aren’t doing this, this type, you know, not every shop in the world is going to become a DevOps shop and there’s ones, there’s always going to be a job for the click-next admin. And my response to that is, yeah, but is it going to be interesting work? There’s still jobs for AS/400 sysadmins. There’s still— I mean, I remember I worked at a financial institution and there was some weird mid-range system that we still had some stuff on, and there were 2 dudes that were the 2 guys who their job was to maintain it, and all they did was sit on the phone with each other and talk and gossip all day because they had no work to do. And maybe that’s a great job. I would go crazy.

Jeffrey: So here’s—

Jessica: well, that— and that’s the thing, right, is that, you know, automation gives you time, right? And time is the real currency. I mean, I don’t care what you’re doing, what industry And that’s it, time and cycle time compression. We’ve talked about this. And I don’t, I’m not gonna remember the quote of where it came from, and I’m probably gonna say it wrong, so forgive me, but this idea of, you know, being automationally challenged. I was talking to Steven Wroblewski earlier today, you know, and he used the words automationally challenged. I think that’s a good way to describe it. And in terms of incentives, you know, I’m glad you brought that up, Jeffrey, because we do have to look at how we incent our people. And in terms of rewarding behavior, you do want to reward the work that is repeatable. You know, Mitch Hill said, get it right, make it repeatable. And that just gives me chills because I think about no matter what industry you’re in, you might be in banking, you might be in used toys. I don’t know what you’re selling, what your company is doing, but you’re in the technology business and people, you know, it’s not a marriage. People change jobs. They win the lottery. Wonderful things happen to people and we need to capture that expertise. And the brilliant sysadmin guy, gal, whoever you have working there who figured that out doesn’t have to keep figuring it out every time. Oh man, how did I solve that? It’s captured. It’s a way to really version control your culture in a completely abstract sort of way.

Jeffrey: [00:38:30] I like that. Version control your culture. That’s great. There’s one other element here. You mentioned all these jobs are still going to be around. I’m not sure, sir, that’s true. Here’s my argument. If you’re just clicking next, like I got the DVD, I put it in, and I click next, next, next, next, next, here’s what you’re doing. You’re offering your employer undifferentiated IT. And my argument is, if you’re offering undifferentiated IT, that same service is going to be offered through the cloud at much better prices, at much better security with much better data protection. You know, look, that’s what the cloud’s great at, right? I can take a standardized offer and make it available to everybody at far lower cost than they can do themselves. So if all you’re doing is offering, you know, doing that, I don’t think there’s much of a future. But I will also then say, but there’s a ton of value you can add by doing differentiated IT to say, hey, I understand, again, one of the my favorite things about the DevOps mindset is, hey, understand the mission. Like, understand, talk to your business owners and say, what are we trying to do here? What drives our business? And then use that information to say, well, if that’s what we’re trying to do, then we should do things this way, and we can move faster, we can move more agilely, we can do things differently, and use IT as a competitive differentiator. And if you can do that, you are printing money for your company. You’re printing money for your company and they should reward you. And when they come and say, oh, or I could get a plain cheeseburger from the cloud, they’re going to say, well, what are you talking about? This guy’s helping me print money. I’m not going to take some generic thing from the cloud.

Jessica: [00:40:19] But if you’re just offering the same cheeseburger at, you know, really expensive prices, there’s no Yeah, I think the discussion about whether you’ll have work, and I have mixed feelings about this because it brings up this fear, uncertainty, and doubt, but I think we do need a little healthy fear. Not of automation. Automation, you need to make time to be scared of more interesting things. I think people do worry about whether or not they’re going to have work, and all the time I do You get good people coming up and saying, what should I learn right now? What do I need to work on? What do I need to learn? It’s automation. Here’s the thing, if your employer doesn’t appreciate the fact that you’ve automated many things and now you have a little more time and there’s a little more slack in your workflow, and if they lay you off for some reason or I’m I’m just speculating here. If you’ve really done that, so many companies will hire you and pay you to automate their infrastructure. I think that fear is a little bit misplaced in many cases. I think probably 3 to 5 years. Just like the number of jobs that will support the click next, next, next, finish That’s going to be just limited. And those companies that do that will be smaller and smaller, not the reverse.

Jeffrey: [00:41:55] You know, but I got to tell you, fear, uncertainty, and doubt, these are your friends, right? In the same way, like, pain is your friend. Do it more often. Well, as a kid, I did karate, and there was this one person went to a no-contact karate place. I was like, how does that work? And this one person, she’d fight and her hands would keep going down, and the instructor would punch towards her face, never touch her, and say, hands up, hands up, and they go up and then they go down. I said, you know what, just pop her. Just like deck her. Because I tell you what, you just get decked once and you keep your hands up. Well, same sort of thing with fear, uncertainty, and doubt. Fear, uncertainty, and doubt, it’s your friend. Look, let me just get personal for a second here. I don’t have a college degree. I’m a college dropout. And every day I come into work with the imposter complex. You took— people talk about the imposter complex. Let me tell you, I have the imposter complex. Every day I come into work like, like, oh my God, somebody’s going to point out that guy doesn’t know what the hell he’s talking about. And I walked with fear and certainty of doubt. You know what I do about it? I work my ass off. I work my ass off. I do creative things and I just perform, perform, perform. And it turns out that’s a pretty good habit to get into. It served me pretty well. I’m now Microsoft’s highest college dropout. And why? And it’s because I because it wasn’t comfortable. It’s because I embraced fear, uncertainty, and doubt, and it caused me to do the right things that benefited my employer. And because it— well, first, it benefits the customers. The customers then benefit the employer, and then I benefit from it as well. So I’m a fan of fear, uncertainty, and doubt.

Jessica: [00:43:31] I am too. I am too, Jeffrey. You know, somebody, Peter Drucker, said, you know, the purpose of a business is to create a customer, right? And so, so many times, you know, the people in IT don’t always connect what they do with the customer, with that customer that’s at the point of sale, that’s at the online transaction ready to check out. Wherever that ultimate, you know, transaction is taking place, you know, we have to always be thinking of that person because without them, you know, we really don’t have any work to do. And wow, I also dropped out of college, so this is— wow. So the, you know, the imposter syndrome is real, but the solution is to work your butt off and learn every day. And if, you know, don’t be—

Matty: Is there ever anyone on this show that has a degree?

Trevor: See, you know, I never thought— I thought if I dropped out of college that I would never be able to be successful in technology. Clearly, I was wrong, and now I just have the debt to show for it.

Matty: [00:44:41] So I have a question that’s kind of come up in my head a little bit, and this is going around this kind of full-stack idea. So in a homogeneous Microsoft shop, which happens, and I’ve been in them, I’ve been in those, and it can happen. You can be homogeneous in lots of ways, not necessarily just Microsoft, but if you are in one where you say, hey, Hey, we are full-stack because that’s the Microsoft way of doing it. I’m saying as a customer, as a user. How do we kind of work around some of the challenges when, to be frank, Microsoft’s tooling might not be the most ideal for DevOpsing? Example, Lync is pretty cool, but Lync is not HipChat, right, or is really not as good for that type of ChatOps kind of work, right? Or we think about— I was going to say Git versus TFS, except TFS can handle Git.

Trevor: It does integrate well.

Matty: Anyway, I was just sort of saying, how do we kind of think about this idea? And I said in my notes, like, how does this play into the culture of nobody ever got fired for buying IBM, which probably using that phrase dates me a little bit. But a lot of times I know people will be like, well, we’re not going to consider that because we just use the Microsoft solution because that way we’ll, you know, ’Cause Microsoft, and I’ve been guilty of this too. I’m like, well, clearly all of Microsoft’s stuff is gonna work together better than anybody else’s stuff would ever work with Microsoft. Like, how can we culturally help people embrace the right tool for the right job, even if it’s a little part, like something silly like the way that you do, well, the way you do ChatOps, right? Or using Nagios instead of SCOM or something, I don’t know.

Jessica: [00:46:20] And I can’t— the only thing I would say to this is, you know, that kind of chat ops, that is something that culturally, you know, more companies need to embrace. And I don’t see it that much in my immediate, you know, team necessarily. But I think culturally it’s pretty critical to get some kind of a live chat going and, you know, maybe Maybe that’s Slack or whatever, pick your tool, right, it’s do you have that available, is it IRC, how are you going to do that, and how are you going to make it not too noisy, so I like Yammer for that kind of things, I think Yammer’s pretty, the tool that I tend to use in the Microsoft stack to kind of facilitate that more pulse of conversation, Yeah, I would say that, you know, in reality, Microsoft is getting DevOps in focus.

Jeffrey: You know, I think some got it in focus a little bit earlier than others, and some still don’t have it in focus. But again, back to— I mentioned the great thing about the company is we’re incapable of sustained error. As more and more of our own people run their own services, they’re stepping back and saying, hey, you know what, this tool doesn’t really meet what I need. Fix it because you’re screwing up. And then step by step, that has an effect, right? The great thing about Microsoft is, well, one, I mentioned before, we’re incapable of sustained error. There’s another great asset we have, and that is we invest about $10 billion a year in R&D. So when we do get our heads out of various body parts, we can move fast. I mean, we could do some amazing things, really. And, you know, at any given point you might say, hey, this guy’s got got a better product now or whatever. But over the long term, if we’re focused in the right direction, the Microsoft engineering machine is really a force of nature to be reckoned with. It’s, it’s quite impressive. So yeah, that’s all a way of saying I think in the past we haven’t done a good job at DevOps and supporting DevOps. I think various products are doing a pretty good job now. And but wait till you see what’s coming down the pipe. I think you’re going to see just a tsunami of getting it in focus and doing a better job.

Matty: [00:48:47] I guess then, which, which is good to hear, and I’m not surprised to hear that, I just want to think again like more around the aspect of the people working, the IT pros working out there, or the devs too for that matter, about being able to kind of think about like kind kind of that the answer isn’t— like I said, it’s sort of going back to the nobody got fired for buying IBM, where I’ve been in places or I’ve talked to people who are like, well, yeah, maybe this is, but that’s not the Microsoft thing, so I’m not going to use it. And that can be limiting, right? Because as much as I absolutely believe you that you will have a phenomenal, you know, chat ops tool at some point, people have work to do now, right? You know, and I’ve been just as guilty of it myself too, of going like, well, this this really doesn’t do what I need it to do, but it’s a Microsoft thing and I buy Microsoft stuff exclusively, so I’m gonna take this and I’m gonna— I mean, I could tell you a whole story about trying to make BizTalk do a thing it wasn’t supposed to do because it was the closest thing to what we needed. And, you know, and that is like, I just kind of like trying to think about how to help people understand that it’s okay to deviate, if you will. That’s even the wrong word, but just kind of think a little bit outside of the tools.

Trevor: [00:50:02] It’s okay to use the best tool for the job and not try and jerk something.

Jeffrey: I’ll just say that it turns out that, like, you and I sitting here, that’s very easy to say. When people run companies, it gets far more complex. Like, you know what, here’s one of the things that people— why a lot of people just say, hey, I’m going to use the Microsoft thing, even if it’s not perhaps the best thing at this moment, is one, But, you know, next year it might be or probably be, etc. And the second is, whatever, like that best tool, it doesn’t help me if that company’s out of business, right? And I know Microsoft, you know, they’re gonna be around for a really long time. You know, 5 years down the road when I’m like in trouble, I know I can pick up the call and talk to somebody and they’re gonna help me. And this tool, like that might be a great tool, I love it, But what’s the longevity of that? Are they really going to be around, etc.? And as I say, it’s easy for you and me to say, yeah, pick the right tool, the best tool at the moment. But if that best tool isn’t around a year from now, maybe it wasn’t the best tool. In other words, there are more elements to the best tool than just features.

Jessica: [00:51:15] Jeffrey, I just want to echo that because before Microsoft, I was a consultant. And a lot of times I was in a position of having to choose a tool for my customer or on behalf of or with my customer. You know, majority of the time it’s got to be with the customer. And always weighing on my soul is, is this tool going to be around? And, you know, you have to do your due diligence, right? You need to look at who the founder is. You need to look at where it’s hosted. You have to look at all the underbelly. You can’t just look at that price per user per month and go, okay, yeah, that’s the solution for me. Software as a service, you know, sometimes I think I think it’s too often the person with that credit card signing up for that and God bless their soul for trying to do the right thing for their company, but you still have to look at the backend. At any given moment, you have to know that they could close. How many times do we get a notice every week, every month, it’s been a wonderful journey, we’re closing down because we want to focus development on this tool, or we’ve been acquired by this company and therefore the tool is going away. You have to plan for that. I mean, your data should be in— this is obviously a gross generalization, but oversimplification— but having the, you know, you’ve got to be ready for that and be able to switch. And I recognize that that’s ridiculously hard for a very large company. I’m not by any means saying that this is easy.

Jeffrey: [00:52:47] And just to put things in focus, prior to coming to Microsoft, I had worked at Digital Equipment Corporation, Apollo Computers, Greystone Technology, Royce Data Systems, and Storage Technology. Yeah, that’s exactly zero of those guys are around. Now I did work for IBM for a while and they’re still around, but that’s like one out of many. So, and these were, by the way, Digital. I mean, Digital was like, boy, if you thought that was a safe bet, that was a safe bet with Digital. And Apollo, we were number one workstation vendor. Storage Technology, one of the number 2 disk drive manufacturers in the world. Like, things change, and you really want to bet with people.

Matty: I guess what it kind of meant to me is that it matters on your scale. It matters on what the tool— because I think I absolutely concur, you can’t just say this is the new hotness, so I’m going to bet the farm, and I’m going to make this be the thing that I run my enterprise on. But by the same token, if it’s something that you’re like, this wouldn’t be that hard of a pivot if this company went out of business tomorrow, tomorrow, we would still be— we would figure out a way to keep doing work, but in the meantime, we can get some stuff done. We can pilot this with a small team. I think the problem is that sometimes we go too far. I absolutely agree. And again, my career has been working for some big old money companies, right? You know, I know how these things go. And it’s very easy, like you said, we sit here and it’s like, yeah, just why can’t you just use the best stuff ever? But sometimes Sometimes things are small, and I think sometimes we want to take the big thought that we do and apply it to every single thing, and that makes it hard for us to try new things because we restrict— I love piloting. That’s the best way to do anything. I tell my customers all the time. People say, how can I be successful with DevOps? I point them to— I say, remember how we did Agile 10 years ago, or whatever it was when we started doing it. How did you do it? Did you try to make the whole company Agile? I hope not because it didn’t work. You know, you stacked the deck. You took a team of people that love change, that were your best people, that were a great new product that you already knew was going to succeed, and you had them do Agile. Then all of a sudden everyone goes, why is that product team doing so good? You’re like, well, because they do Agile and then everybody wants it. Guess what? Works for DevOps. Listen to Justin Arbuckle’s talk at ChefConf last year. That’s how they did it at GE. You pilot and you show results. And you can’t pilot when you think about the whole big thing all the time, right? Like, you can try things, and then you should make your decision based upon was it effective, and partially now if I’m going to say I’m going to embrace this big, what does this mean? But it doesn’t mean you need to do 30 pages of due diligence before you do a trial of HipChat for 3 weeks with your team to see if we can learn a different way to work. And by the way, that’s not a shill for HipChat. Chat, but whatever it might be, right? I have different feels about HipChat, and they’re not a sponsor yet, so I don’t have to talk so much.

Jessica: [00:55:44] Yeah, you’re right. You have to make small experiments. And I mean, ideally, you’re making that, you know, thinking about the larger system and not just selecting a tool because, you know, someone gave you a deal on it or something like that. But you have to be able to try that, and you have to try it with people who are on fire about the underlying goal, and, you know, Jez talks about that in this, you know, continuous improvement, Kaizen, you know, the Toyota Improvement Kata. You know, it’s every day, is what I’m doing in alignment with the goal that we, you know, are all working towards the same goal? And it makes it, I think, much easier to look at, does this tool fit in with that culture? Does that tool fit in with that mindset? And as long as it’s kind of meeting one of those needs, give it a try, by all means, please. Do something.

Jeffrey: Yeah, can I offer a little dissenting view here? Because part of what the memes are, themes I hear picked up in the DevOps community, is this model of, hey, you know, let a team, give them the freedom to go do what they’re going to do, and if they have passion around it, let it go. And then I’m hearing things like, oh, you know, somebody’s writing the stuff in Erlang. I’m thinking to myself, Erlang, really? What happens when that guy gets another job? Like, okay, so he’s told you, hey, trust me, I have passion around this, this is the right technology to solve this problem. Well, everybody else is doing it in C# or whatever. And so, well, maybe it is, but again, that guy gets hit by a beer truck, who’s going to pick up the Erlang code, right? Nobody. You’re kind of screwed. And I had an experience like this, right? So at Digital, when I got hired there, I got hired to clean up this management product. And when I did an inventory, there were 18 languages that were contributing to this product. And I’m like, oh my God, how’s that possible? And I looked, there was 80 Ada! Ada! And I said, well, what happened? Did somebody like take a night course and decide to write some code in Ada? And like, yeah, that was Brad. Brad took the night course in Ada.

Matty: [00:57:47] That’s sort of how Rails exists. You know that, right? Like Ruby on Rails is a thing because some like contractor at 37signals wanted to learn Ruby.

Trevor: Oh no.

Matty: That’s why Ruby on Rails exists.

Jeffrey: Oh my God. Yeah. Well, anyways, that was That was a mess. I mean, that was a disaster. And at the time, probably the team thought, oh, you know, look, we’re so agile and we got this great culture. Everybody gets to do their own thing. But now fast forward a year, 2 years later, and it was a mess. It was a complete disaster.

Jessica: Well, and something like that, in as far as the language, I mean, that’s a different level of impact than choosing a chat tool, you know. So I’ve Obviously putting that into perspective. Then my husband tells me an example of, he looks at it, he’s a data scientist, so he’s looking at a code base and then he’s going off and saying, well, I can do that in 25 lines of Python. The experiment is being able to show that code and show that tool and have it be accepted by by the larger group of developers or however, whatever your structure is. But you can’t make that decision in a silo off in a vacuum, right? That’s a much different decision than a chat tool, but I absolutely agree with you.

Jeffrey: [00:59:15] Yeah, I had another guy at IBM, same thing. This guy was like a rocket scientist. This guy was great, C++. He knew everything about C++. And then when he left, found out he was the only guy who knew how to write C++ code like that. Nobody could understand his code. He used the esoteric languages, you know, but it was one of these trustier guys environments and looked like he was making great progress. But then when he left, I realized I just had crap and had to like flush it away. So that was a painful experience too.

Matty: That’s a really, I’m just gonna make sure that I wanna explore that topic much more deeply at some point, ’cause I think there’s a lot to think about where that goes, right, to that trust, but then also not giving yourself, like, you know, a shoe-shaped gun, right, or giving your organization, you know, making it so easy about that. We’re gonna have to do another episode, y’all. We’re definitely gonna have to have more of these Windows and Microsoft episodes for sure, because we are at the end of our time. We want— we’re past our time, and we want to wrap up. I have one more question because I promised Brian Barry of the Food Fight Show I would ask this, and I told him I thought this was a thing that Steven Murawski could answer. And so his question to you, Jeff— Jeffrey Snover, is why can’t I copy a file to a server over WinRM? You can answer that however you want.

Jeffrey: [01:00:48] I would say that great minds think alike, and when we get our head out of our body parts cards. Like, come, come talk to me at Build and Ignite. Okay, I bet you somebody’s job I got a good answer for you then.

Matty: It’s, it’s, it’s interesting. Yeah, Brian is one of the hosts of the Food Fight Show, which is a chef podcast, and he’s been probably over the last 7— probably about 6 or 7 months ago, he started doing a lot more Windows automation than he’s ever had to do before in his job. And so It’s a whole new world for him. That’s why we had— there were a couple of Food Fight shows just based on Windows configuration, and I think basically so that Brian could get a lot of questions answered.

Jeffrey: Actually, you know, he’s exactly right. I got a guy prototyping it, and at first we thought, well, okay, this is gonna be a— it’s gonna be horrible performance because, you know, compared to SMB, SMB is the right way to do that. But there’s reasons why you want to do it over WinRM, but the performance can be terrible. Anyway, the guy came back to me, he’s like, well, that’s actually interesting.

Matty: [01:02:00] That’s gonna potentially help something with my checkout that I’m gonna be talking about, knowing that. Yeah.

Jeffrey: All right, so not a problem.

Matty: Hey, that’s minimum viable product.

Jeffrey: You did ask the right question. Yeah, that was spot on.

Matty: All right, good job, Bob. Brian. So some community event stuff. So upcoming conferences: ChefConf is March 31st through April 2nd in Santa Clara, California. chef.io/chefconf. The code—

Jeffrey: there’s gonna be some great speakers at that. There are.

Matty: So if you use the code ADO to register, you get 10% discount. Everybody on this podcast is going to be speaking at that conference, some of us about more substantial things than others. And by that I mean everybody but me will be speaking about something with some I’m going to be talking about animated GIFs. The PowerShell Summit is April 20th in Charlotte, North Carolina. I’m going to have links to all these in the show notes. Microsoft Ignite is May 4th through 8th in Chicago. I will be there because I live in Chicago, so I get to go. Lots of DevOps Days coming up this spring. Ljubljana, if I said that right, is April 3rd and 4th. Paris is April 14th to the 15th. Denver, the first Denver is April 23rd through the 24th. Austin, which is like, I think this is the 5th or the 6th, is May 4th through 5th. Toronto is May 14th through the 15th. The ALM Forum, the Application Lifecycle Forum, Application Lifecycle Management Forum, is in Seattle May 18th through the 22nd. You can come hang out with me there. Got some CFPs open. If you happen to be listening live, OSCON CFP closes today, February 2nd. If you’re listening to this later, sorry, This is what Bridget’s doing right now, actually. Agile 2015 CFP closes February 22nd. QCon New York City closes March 1st. And there’s CFPs for lots of DevOps Days open now. Toronto, Paris is open-ended, Minneapolis. Visit devopsdays.org to submit your best ideas. So let’s go into our checkouts. So Jessica, do you have any checkouts for us?

Jessica: [01:04:15] I have, I have too many checkouts.

Matty: Okay.

Jessica: But I’m gonna limit it to 2. Okay. First, I just want to plug the Southern California Area Linux Expo, SCaLE 13. It’s going to be— we have a DevOps Days on February 20th. So if you’re gonna be in the LA area, it would be super awesome to see you there. And I have a discount code which I’m will post in the show notes for you, but would love to see you at, in LA at DevOps Days. Amazing community down here of DevOps practitioners and just generally awesome humans. And so for my 2 checkouts, I’m totally going to plug 2 books that I’m obsessed with right now and reading. I was reading this one at a bar. This is the Field Guide to Understanding Human Error. By Sidney Dekker. I just got this, this is the 3rd edition and it’s absolutely amazing. If you are struggling with your culture and trust and things like, how do you get folks to be able to share what happened in an error or disaster situation, and obviously this stuff is not life and death, a lot of these come from aviation and maritime disasters where often Often the people who were right there at the time, they’re often dead. In IT, it’s not life and death. We treat it like that sometimes. I just love the way this puts things into perspective around the topic of human error and blame and things like that. The second piece is the— I don’t know if you’ve got the Lean Enterprise book from Jez Humble and Out on a Roll. You can see that I have like 1,000 little bookmarks in there already. Somebody on Twitter commented that they had broken the highlighting portion of their Kindle reading this book.

Matty: [01:06:18] It doesn’t really count if you highlight the whole thing.

Jessica: Yeah, I know. That’s why I kind of stopped. It’s just that good. There’s just wisdom in literally every paragraph. Pick up these 2 books. I have a ton more. I think it would be neat if we all published an updated show of the stuff we’re reading right now. I think that would be neat to share, but it’s been absolutely amazing to be on this podcast and with all of you and everybody out there listening, so thank you very much for the invitation.

Jeffrey: Thanks.

Matty: Do you have any other checkouts?

Jeffrey: All right, so I put a bunch in the show, in the document, but I’m gonna talk about a couple of my favorite podcasts. Right now I’m just fixated with Dan Carlin’s Hardcore History podcast. He is just an incredible storyteller, and he’s doing currently a series on World War I, which is just fantastic. So can’t recommend that highly enough. And the second one I was just talking to somebody today about is the Brain Science Podcast. So if you’re into brain science and neuroplasticity and all that great stuff, fantastic podcast by Ginger Campbell. She’s a doctor, but she’s just an amateur brain scientist. She’s just total— by amateur, I mean, a true lover of the field. And so that’s some great stuff. So then I’ve got a couple other links about PowerShell and my Twitter feed and my personal blog in the show notes.

Matty: [01:07:51] Yeah, everybody can check those out. So Trevor, what do you got for us?

Trevor: I’ll be brief. So I was really excited by all the news around the FCC this week. Seeing that we’ve redefined broadband is really cool. Should have probably been higher, but we’ll take what we can get.

Matty: Ironic that your broadband connection right now is super slow.

Trevor: It is, it is, yes. But just some of the details around how that came to be apparently, like the— for once, I liked what Tom Wheeler was saying to counter the arguments of the internet service providers. It was interesting, as well as the other stuff around that. But also this week I found out that Windows 10 is going to run on the Raspberry Pi 2, which can lead to some interesting things. Apparently it’s just a shell, but I’ll take it.

Matty: [01:08:54] That’s cool. As in Server Core, you mean, or whatever the equivalent of Server Core? Like a core version of Windows 10. That’s kind of neat.

Jeffrey: I didn’t know that.

Trevor: There’s another version of Windows Server that I’ve heard rumors about, but I don’t know if those rumors are public yet.

Matty: We’ll find out. All right, so I got a couple. So one is on topic. So Test Kitchen for Windows is almost a thing. It’s actually been kind of a thing for a while, but like we’re getting really close to it being a full thing. But if you want to play with it and kind of be able to check out rapid development on it, if you look at the Windows cookbook that’s at github.com/opscode-cookbooks/windows, there’s a link to the show notes. The Gemfile that’s in there, that’s got some Kitchen YAMLs, you can actually spin up and you can try converging a Windows guest using Test Kitchen, which is super cool. Also, we were talking in the green room about that I use BitTorrent Sync to synchronize documents between my MacBook and my iMac. I am giving a call out for that. It’s at getsync.com. Basically, you can think about it like Dropbox, but there’s no cloud involved. It’s just between machines, but it uses the BitTorrent protocol to do it. It’s really efficient. Also, I’ve been listening to the audiobook of Yes Please by Amy Poehler. I would recommend the book. She reads it herself. Seth Meyers is reading on it. She’s got like her parents are in it and like it’s super— and Mike Schur from Parks and Rec reads some— is in there. And so I’ve been a fan of Amy Poehler back from the Upright Citizens Brigade days, and I— and it was just awesome to hear like stories about Chicago Improv and backstage kind of Parks and Rec stuff. But she’s got a lot of insight about human condition, and I really would recommend it, so that’s— it’s super fun.

Jeffrey: [01:10:50] So can I add one?

Jessica: Yeah.

Jeffrey: Well, Trevor was mentioning he heard rumors about Server, so I don’t know what they are, but I will tell you this, there are some great rumors out there. No, there are— at Build and Ignite, I’m going to be doing a session where I talk about some just amazing changes coming to Windows Server. If you like Windows Server now, put your seatbelt on. There’s just some amazing stuff coming out, and I’m going to give all the details at Build and Ignite.

Matty: SSHD?

Jessica: Yeah, yeah.

Matty: I was, I was given— when I first met Jeffrey at ChefConf last year, there were 2 things that happened. One was I was like, when can I have SSHD on Windows Server? When? And then I also told him that I thought PowerShell was as readable as Perl, and yet he somehow still talks to me.

Jeffrey: You also told me I was wasting my time with Desired State Configuration.

Matty: I don’t think I would have said maybe.

Trevor: [01:11:52] I don’t know what you said.

Jeffrey: That’s what I heard.

Matty: Okay. Wow. It’s not true. I love me some DSC. So on that note, we have a newsletter. You can sign up for it if you go to arresteddevops.com/bananastand. It is the best way to know about our upcoming podcast episodes and cool news with DevOps. We’re actually sending it out every 2 weeks. Weeks. Now we’re actually doing things with it, so sign up for it. We don’t spam you, we don’t sell your email address, we don’t hack you or whatever it is that people are afraid of with newsletters. Whatever it is, we don’t do it. Big thanks to Mandy Moore, @therubyrep on Twitter or at devreps.com. She is the best in the business for helping out software professionals from administration and research to event organization and to podcast production. And so check them out at devreps.com.

Trevor: Thanks again to our sponsors. Be sure to visit them at arresteddevops.com/pagerduty and arresteddevops.com/datadog30. Thanks, Jeffrey and Jessica, for joining us, and to our loyal listeners, if you enjoy Arrested DevOps, we would appreciate it if you would visit arresteddevops.com/itunes and leave us a review in the iTunes Store. No matter what you have to say, we’d love to get your feedback. Feedback.

Matty: [01:13:11] You can check us out at any time on arresteddevops.com because, I mean, it’s the internet, it’s not like there’s hours, or @ArrestedDevOps on Twitter. We’d love to get your tweets. And hey, any input, ideas, or feedback, do that email thing if you still like email to shows@arresteddevops.com and let us know any ideas that you have for future episodes.

Trevor: I’m Matt, @MattStratton, and I’m Trevor, @TrevorGHess. We’re Arrested DevOps. And remember, there’s always DevOps in the banana stand.

Matty: Oh my God, we’re so terrible.