Alan Shimel: Hey, everyone, welcome to another DevOps Unbound. If this is your first time catching a DevOps Unbound, welcome. DevOps Unbound is a biweekly video series where we talk about relevant topics in DevOps. In addition to the biweekly show, we do, itโs every four to six weeks, a live roundtable show where we invite the audience to participate and drive the discussion with online questions and comments and so forth, so we open it up to a live audience. Unfortunately, this show is not one of our live roundtables, it is prerecorded, but we have a great panel and I think youโre gonna enjoy it.
Iโm gonna get into todayโs topic in a second, but before we do, let me introduce you to todayโs panel. First of all, I wanna introduce a business partner of mine and my friend, Jayne Groll. Jayne, why donโt you introduce yourself to the audience, if you donโt mind?
Jayne Groll: Thanks, Alan. Hi, everyone. Iโm Jayne Groll, Iโm CEO of The DevOps Institute, weโre a global member association and a certification body around topics related to DevOps. Iโm really, really excited to be here and Iโm hope everyone listening joins us at DevOps Institute and becomes a member.
Alan Shimel: Fantastic. Welcome, Jayne. Joining Jayne, Iโm happy to have Paul Bruce. Hey, Paul, welcome, and if you wouldnโt mind, you know, introducing yourself?
Paul Bruce: Yeah, hey, Alan. Good to be back. My name is Paul. In my full time job, I work with Tricentis in a number of different areas. Iโm historically a performance and reliability nerd. In my off time, I organize a number of events, help to co-organize volunteer things like DevOps Days Boston, just did o11yfest this year about observability, we had that conversation, I think, with Christine Yen a couple months ago.
Alan Shimel: Good stuff. Mm-hmm.
Paul Bruce: And generally, Iโm a people advocate. I had to pull DevOps out of my LinkedIn title because there were too many recruiters that I donโt need, because Iโve got a pretty good job. But also, because people advocate really describes what I care the most about, right? People are some of the hardest part of the DevOps people, process, and technology, really, software. So, I really focus my time on that.
Alan Shimel: Thatโs great to have you back on, too, Paulโgood stuff. Next, we have a first time guest on DevOps Unbound today, a new member of the band, Sarah Baker. Sarah, welcome to DevOps Unbound, if you wouldnโt mind introducing yourself?
Sarah Baker: Thank you. Wonderful. Welcome. My name is Sarah Baker, my pronouns are she and her. Iโve been working in Operations my entire career and Iโve moved into DevOps as something I alwaysโI needed a name for it. I always wanted the people part of Ops and technology and DevOps was that. And Iโve been doing that as soon as I learned the term. And Iโve been moved and started working with Paul Bruce in the IEEE work group 2675, who is trying to create a standard for DevOps.
Alan Shimel: Fantastic.
Sarah Baker: I work forโand today, I work as the Manager of Biztech in Airbnb.
Alan Shimel: Thank you, and thanks for joining us. That sounds like a great place for you to do this. And then the last member of our panel today is my co-host and another business partner, a long-time friend as well, Mitch Ashley. Mitch, if you wouldnโt mind introducing yourself?
Mitch Ashley: You bet. Always good to have a good panel on with you, Alan. Mitch Ashley, Iโm CEO of Accelerated Strategies, an analyst firm thatโs focused on DevOps, and also on cloud native, cyber security, and digital transformation. I come to DevOps both as a practitioner and a vendor leveraging it, so very interested in hearing the perspective of folks on our topic today. So, thanks for joining us, everybody, too.
And by the way, our DevOps Unbound is sponsored by Tricentis. I think you may have mentioned that, right, Alan?
Alan Shimel: No, I think I didnโt, Mitch, but thanks for picking it up for me. [Laughter]
Jayne Groll: [Laughter]
Mitch Ashley: I was trying to give you credit where credit may not be due.
Alan Shimel: Yeah, no, I donโt wanna take credit when I messed up. Yeah, DevOps Unbound is sponsored by Tricentis, and many thanks to them for their sponsorship. It makes all of this happen.
So, letโs turn to the topic at hand today, guys. You know, when I first started DevOps.com eight years ago, I met Jayne maybe a year, within a year of that or a year and a half of that. One of the interesting things about DevOps is, there was no manifesto. There was no hard definition of what is DevOps and what is not DevOps, you know, contrary.
And ever since then, itโs been a double-edged sword. In some ways, itโs helped DevOps grow, itโs helped DevOps stay current over the last 8 to 10 years as the world has changed, DevOps has changed along with it. On the other hand, though, weโre still dealing with the, โSo, what is DevOps?โ Right? โWhat exactly is DevOps and what exactly is not DevOps?โ which is sometimes harder than what is DevOps.
And as we sit here now, August, 2021, there really still is not a quote-unquote standard for DevOps. There is no standard DevOps. It is what you want it to be, right? As I think Andrew Clay Shafer said, โThe DevOps you get is the DevOps you deserve,โ right? But have we outgrown that stage? You know, that seems like a terrible twos stage of DevOps to some; to others, itโs what kept it cool. Has DevOps matured? Do we need standardization? Will it be something that kills the goose that lays the golden eggs?
Weโre gonna discuss that today, and Iโm interested in all of your views. Jayne, as the CEO of DevOps Institute, how many people have taken certified DevOps classes and certifications with DevOps Institute? Itโs gotta be tens of thousands.
Jayne Groll: Yeah, about 25,000 so far. So, itโs a pretty large community.
Alan Shimel: Mm-hmm.
Jayne Groll: And if you donโt mind, I wanna kind of just lay the stage, because Iโm really interested in the standardization initiative. So, you know, I had a privilege of seeing, going to DevOps Days in 2012 in California. So, if you can imagine, it was in an old, dusty data warehouse, right, and it was at the very, very early stages. And so, one of the reasons DevOps Institute emerged was because there was a lot of really cool things here culturally, from an automation perspective, right, doing things faster, collaboration.
And so, we started up DevOps Institute because we wanted to make sure the humans were represented. And I love that Paul and Sarah in particular both represented people. The problem with that is, at that point, it was very, very open, it was very kind of organic. You know, it was really a group of passionate people that were trying to just change the way IT operates, right, from a people, process, and technology perspective.
What happened after that, though, is that thereโs so much noise about DevOps that if I were an enterprise CIO right now, my head would explode, right? Now, is there one right way to do DevOps? No. Is it only gonna get you there with automation? No. You know, is there a very, very strong human elementโabsolutely.
So, itโs a double-edged sword, Alan, as you said, is that we wanna be able to kind of scale down, right, some of the common language, some of the common definitions. I always say Iโm married to a CPA, they have generally accepted accounting principles; maybe someday, we have generally accepted IT principles. But how do we distill the noise, right, and everyone having a different interpretation of how to do DevOps and what it is compared to, letโs get past the, โwhat is it?โ Weโve spent, what, 10 years trying to figure out what it is, to, โhow do we do this?โ Right?
Paul Bruce: Mm-hmm.
Jayne Groll: And it isnโt just one roadway that takes you there. So, I think itโs a challenge. I mean, we introduced foundational certifications around DevSecOps and SRE and DevOps Foundation and DevOps Engineering, really, to create kind of a common vocabulary more than the, โhereโs the single way to introduce CI/CD.โ
So, I don’t know if that helps kinda set the stage, but I think itโs a really big challenge right now that, if we standardize it, we run into the risk of what happened with ITIL where it couldnโt keep pace, right, and it became religious. I come from the ITIL space, right, so itโs like, open chapter 1, verse 3 and sing a hymn before you introduce change management to how do we then, on the other end of the spectrum, itโs so open and so undefined that itโs really complicated for legacy, I call them heritage organizations, to be able to move from this step to this step.
Paul Bruce: Mm-hmm.
Alan Shimel: The church of ITIL. Paul, whatโs your feeling on that?
Paul Bruce: Well, everybodyโs on a journey, right? Your personal journey, your professional career, whatever, maybe you go do a walkabout on a particular topic to learn more about that. And so, we started this venture, there was an early round of, โWho wants to be involved in something like this?โ in 2016. I answered a call from Bob Aiello and Lynn Carter, who were kind of forming this idea of, thereโs a bunch of people who really care deeply who have practitioner practical experience in the field. And the early part of this process was just learning, kinda level setting on what we have done, right?
Jayne mentioned that itโs hard, maybe, to have accepted IT standards, even though there are standards, right, that not every organization applies the same way. I don’t think that invalidates some of the general principles that we know to be true, which is, you know, you ship broken software, you hurt people, you hurt yourself, right? But those are high level. How do we break this apart and start to make sense of the world, as you said, the heritage, Jayne, you said the heritage organizationsโwhat is so wrong, whatโs going on that we have these kind of security problems, these performance problems, this inability to be competitive in terms of acceleration and stuff.
And so, I think the purpose of a standard is not necessarily to force people to do something a certain way. I think itโs simply just kind of like, if youโve ever had the Christmas ornaments or just ornaments box that has the insert that allows for things to be put in careful places and also work together as a unit, right, to provide a normative, not a prescriptive standard, but a normative standard to say, โLook, these processes that you already know? Well, you already followed these from, letโs say, 12207, right?โ These processes around supply acquisition, knowledge management, human resources? Right, human resourcing in DevOps is a big problem, at least in the Boston area. Try to hire an SREโthereโs only about 15,000 to 20,000 in the United States, legitimately. Thereโs a reason for that demand.
So, how do we think about the processes we already know and start to say, how are these things different because there are some accepted principles and practices around DevOps? Now, the practices are not things that you would ever wanna dictate in a standard. This is not the same granularity level as some other standards that dictate how to encode time formats. Thatโs very important at that level. But the level that we were working on was how to level set across the things that, especially organizations in high compliance and high risk industries, right, risk-averse industriesโhow do they take the practices and processes that they know from a process standard and start to go, โI understand how to plug, now, what the DevOps principles are into the standard to encourage both of those practices in a more predictable manner?โ
So, when we did all this workโitโs taken years, right, of a lot of learning collaboration between a whole host of people I donโt have even time to list out, but a lot of people from a lot of various different aspects of software have a lot of experience in those aspects. You know, it took synthesis and refactoring, and I was up nights and weekends just rewriting some of the statements, you know? And each of these, itโs a line to 12207, so thereโs multiple processes, process declarations like quality management, quality assurance, verification and validation.
Like I said, these project enabling processes and then these technical processes, and each of those process sections have outcomes, activities, and tasks. Things that are likely to occur in this area, right, as well as, regardless of how you do this, right, not dictating what the DevOps looks like, itโs outcomes of these things that are auditable, that not compliance but conformance. That you can confirm at various levels to these outcomes to say, if itโs important to get this right, when you hit the piรฑata, what comes out should be candy, not garbage, right?
So, Iโll pause there, because I wanna give Sarah a chance to kinda provide perspective, because sheโs also been working a lot with us, for all those years. [Laughter]
Sarah Baker: I think I counted 27 people on the final author list for the IEEE.
Paul Bruce: Yeah.
Sarah Baker: And it was many, many weeks of arguing, reading together the exact language. And I agree very much with some parts of what Paul says, and I wanna step back and talk a little bit about how I see, you know, what, to me, was the big shifts. And so, the way we wrote the doc was to kinda look at the 12207 and say, โHow does DevOps change it? How has it changed the game? What are the perspectives that you need to have in mind in order to get the outcomes that Paulโs talking about, there?โ
And so, when we were writing it, we were writing it to that. How is this different, and to what extent do you wanna consider these areas? Itโs a framework of thinking. At its largest level, itโs a framework of, for me, the value chain of delivery of software. And itโs notโand it uses many of the concepts that come out of learnings of industrial engineering, way back in the โ40s, โ50s, โ60s. But it had never really come into a lot of the thinking of software, because software is different. And the more I was working and kind of digging into what are the concepts underneath DevOps, I came quite aligned with a lot of industrial engineering concepts about how things flow and how things move back and what the peaceful aspects are in that.
And so, thatโactually, if you read through the standard, you’ll see a lot of those flow pieces and all of the pieces that you then say, โOh, itโs the feedback loops, yeah, lots of feedback. Collaboration, itโs all there.โ And itโs just applying a lot of those learnings into software engineering, to me. And it was stuff I was like, โOh, itโs so wonderful to have this.โ
And the other thing, one of the big learnings that I think I brought into the conversation is howโand Paul will back me up on thisโhow security for a lot of us changed slightly as we started walking our way through the 2675 work. The terms of risk became a very interesting concept that flowed through the document more than I would have expected.
Paul Bruce: Mm-hmm.
Sarah Baker: And so, there was some learning within the totality of our team as we wrote this. It was quite fascinating.
Mitch Ashley: Iโm really fascinated to hear from both of you that have been working on this, because itโs not something Iโve been involved in. I sorta have a mixed history with standards, because Iโve worked in areas like networking and things like that where standards are very beneficial or really fundamental, but Iโve also learned standards are kind of what youโre gonna do, not how youโre gonna do it. Because we ran a test lab that, even though it says weโre gonna do least cost routing in a network device this way, people implement it differently and they donโt always work together. So, itโs not a nirvana, but itโs also not a pariah, either.
Paul Bruce: Mm-hmm.
Mitch Ashley: So, what I wrestle with here is that, I think some of the thinkingโand Jayne can speak to this, but innovation happens so quickly now. So many things are created without standards; maybe they evolve into standards, maybe they donโt. And it doesnโt bother me that thereโs lots of noise around DevOps, because weโve all seen the cycles, the hype cycles, right? If itโs worthwhile, everyone and their brother, sister, cousin, uncle will all jump on the bandwidth and call what they’re doing fill in the blankโitโs AI, itโs DevOps, itโs whatever. Thatโs just part of our industry. And itโll either fizzleโyou know, value stream management, is it a real thing? Weโll see. Itโll fizzle or itโll happen, right, if itโs a real thing. And DevOps has proven itself not only is it a real thing, but other parts of the organization now see an opportunity to play, to participate, to become part of the software creation process.
Paul Bruce: Mm-hmm.
Mitch Ashley: And I think thatโs whatโs so powerful about DevOpsโitโs not about a technology, itโs about how we create software. And if you were gonna start out, you know, Jayne, when you went to that meeting in 2012 when you said, โWeโre gonna go create a standard for completely doing software differently because we have cloud and we have the ability to automate and we were gonna change the software architecture,โ you couldnโt have known how to do that. But being the industry wouldโve, you know, the antibodies wouldโve crawled all over all of us and stopped it. So, I think thereโs a lot of power in the innovative, organic nature of it.
And I donโt want DevOps to turn into ITIL, [Laughter] because thatโs sort of taken on a life of its own, and now itโs trying to catch up and itโs done some good things, too. So, I wrestle with trying to constrain it when I think itโs real power is the innovation behind what weโre doing.
Paul Bruce: Yeah, so, before Iโ
Sarah Baker: If I could take a momentโ
Paul Bruce: – yeah, go ahead.
Sarah Baker: – Iโd like to take a moment and call out something. Part of the problem with ITIL is like, itโs ITIL. And, you know, a gazillion managers said, โITIL, let us get a hammer and hit you over the head with it.โ And DevOps, fortunately doesnโt have that problem. It has, you know, a little fuzzier in its definition, but the great thing about DevOps is that it has that sort of fuzzy edges. You canโt, itโs really hard to beat someone over the head with it. You can try. Iโve seen it happen, you know? Everybodyโs a DevOps engineer now, but itโs not necessary.
And what people talk about is not that, but the work, you know? Itโs not, โWeโre conforming with ITIL,โ itโs like, โWeโre doing DevOps.โ The doing part is the important part, not the compliance part. So, when you say, when you hear people talk about it, they talk about it differently, and thatโs the good thing.
I use ITIL, I just donโt tell anybody the word.
Jayne Groll: Which is a shame. Itโs such a shame because, you know, I come from the ITIL space, and in the early days, this was not what the intent was to be. It was very much like DevOps in terms of, โLetโs help IT grow up,โ right, in terms of some things.
And, you know, to what youโre all saying, youโre absolutely right. A framework, whether itโs a standard, is a guideline, right? It helps you with the journey, as Paul said. And so, if I wanna kinda check, check, checkโyou know, Iโve worked with some of the ISO standards, right? So, in the ITSM space, there was ISO 20,000, 27,000 for security, right? And it basically, like, in the one or service management says, โYou should record all your changes.โ It doesnโt say how to record it, it doesnโtโitโs a really great practice, you should do this.
The problem, sometimes, with standardsโand Iโm sure you know this and itโs a riskโthe pure intention is pure. The ability to weaponize it, right, from the auditors and from others is a real risk, right? Because weโve seen that with other standards as well, where it becomes more about the compliance than it is about how having it being a guiding light. So, I think thereโs always a risk there.
And then the other risk is, you know, think about observability, right? So, observability has been trending, what, for a year, maybe two, right? Fourโokay, but really mainstreaming, like, talking about observability as kind of a mainstream practice, right? And I will share with you, Mitch, to your point, when we launched DevOps Institute in 2015, we were not welcomed with open arms, right? We were vilified for trying to put some boundaries around this.
Now, you know, move forward seven years and I think weโve overcome a lot of those and we have some that were detractors or really great supporters. But I think they were concerned that it would become weaponized or commoditized or whatever.
So, Iโm just gonna say one last thing because I think itโs important. Commoditization of knowledge, which, this is kind of in support of the standard, right? Commoditization of knowledge where we make sure itโs available to everyone, some people see that as a negativeโitโs not. What would happen if the whole world, rightโevery IT organization, every IT professional decided to take this standard and use it as a guiding light, according to the needs of their organization? They werenโt worried about the audit, they got trained, they shared knowledgeโwouldnโt that be amazing? But thatโs commoditization.
So, I think the risk of good versus evil on initiatives like this is real, but I also think that distilling it to everybody, kind of like accountants have done, right, is important because IT needs to grow up.
Paul Bruce: Yeah, wellโ
Mitch Ashley: Well, and if you look at the accounting standards, they donโt tell you how to do it, they tell you what to do.
Jayne Groll: Right.
Mitch Ashley: It isnโt prescriptive in the how very much, so itโs very much like an IEEE standard or whatever.
Paul Bruce: Yep. Well, first off, I forgot to mention when we first startedโI speak as an individual contributor. If it isnโt abundantly clear, [Laughter] I donโt speak for the IEEE, and I donโt speak for the other people who have spent years contributing to this meaningfully.
So, when I say certification processes inherit all the exact same things you just mentioned as problemsโweaponizing, actually low ROI maybe in some casesโwe see these deficiencies and these outcomes that arenโt so great. And if youโre somebody whoโs just gonna sit back and say, โIโm not willing to hold myself accountableโ to stand up and say, โWhat does this thing actually mean like a real engineer should, you know, to start to break apart the problem and see what are the components, how does this thing work and how do I want it to work? How does it work best for the people around me as well?โ If youโre not willing to do that, then I would call you the nay-sayer, you know what I mean? Like, not any one in particular, but that would be a nay-saying, like, letโs avoid accountability, letโs avoid doing what we know needs to be improved?
Continuous improvement is a huge element that is all throughout the entire standard. Itโs one of the things thatโs identified as some of the core principles in an ISO study that was done before we started in on this standard to see what are the common themes of people, when they’re really feeling like there is this DevOps mindset, this culture, these practicesโwhat do they mean by it?
And a number of things came out. One was a high degree of automation, right? Okay, well, thatโs very tactical. No matter how you do it, rightโhigh degree of automation. Duh, right? How about effective communication and collaboration? How about shifting things and continuous everything? The notion of, that these things are not this one time and then three years later, you get to come back. No, no, no. If itโs a good idea now, maybe donโt bite off a three-year apple. Maybe take a bite, but know that you started in, thereโs a continuous improvement process.
I can say, we mentioned earlier about a holy war that might exist when you start getting into what is DevOps and what is not DevOps. Like, I startedโthe best way I can describe to people, because like I say, half of my day is spent on and off of Slack with people who are actually doing, working this out in a highly statistical, large distribution across the country, you know, and sometimes outside the U.S., too, just time zone supply.
But I can say, for sure, after seeing the complete garbage Dumpster fire where nobody is willing to stand up and say, โHey, arenโt these things common and standard?โ And oh, by the way, not vendors that have something to push, right? I mean, thatโs, I think, Jayne, where you were talking about maybe observability is now a thing. You know, the only reason why itโs a thing for the most part is because itโs in our face now because all the vendors are hash tagging the crap out of it.
Same thing happened with DevOps five, six years agoโeverybody had to get on the bandwagon. I remember CEOs way back then going, โWhat is this DevOps thing?โ Oh, man. You know, just talk to some of your engineers.
And what I can say, right, I have developed at least an internal personal perspective on what is not DevOps. So, if I can borrow the last 30 seconds of the mic for a second, here are some things I wrote as notes as Iโm going back over the standard, re-reading it, going what I violently agree with from both others and myself, we did not have a chance to see this before it went live many times, right?
So, what I got from one of the higher order sections in Section 5 was this feeling of, like, what is not DevOps. And those things are burnout. DevOps is not burnout. You know, youโve probably been burnt out if youโve worked in DevOps before, so where is this coming from? Why is this the case? Itโs not burnout, itโs not no help. You donโt get any help, go do this thing, you know? Itโs not heroics. Itโs not superheroes in capes, right? Single points of failure, as I call them. Itโs not cliquesโwhich, by cliques, I mean power silos. Huh-uh, there is no such thing as The DevOps Department, people. Thereโs no rogue decisions. Thereโs traceability in decisions. Thereโs accountability in decisions.
Leadership thrashโnot easy to, if youโre really adopting this DevOps mindset and way of thinking about how to do things better, leadership thrash doesnโt really fit in. Toil and debt, right? One of my biggest things is, you know, I seek to eradicate toil, which is not easy. [Laughter] Just reducing it a little is nice, but itโs, and itโs probably pragmatic to say so, but really eradicating toil from other peopleโs plates and from your own, thatโs a challenge, right? Thatโs kind of a life goal. And if we do this stuff together, right, if we say these are not the things we want, the question is how do we move towards something that isnโt those things, right, and that is closer to where we all align and agree, like, this is a better way of delivering software and doing it faster, right?
So, anyway, Iโll pause there.
Mitch Ashley: You know, one of the things Iโd love to hear from both Sarah and Paul is, thereโs always sort of that, โSo what?โ answerโquestion, right? โIf we did this, what is the result? Whatโs the impact of it?โ And I mean that in a positive way, Iโm not being critical.
Paul Bruce: Yep.
Mitch Ashley: If this standard was widely adopted, how would it improve? How would itโwhat things are we wanting to improve and change from where we are today? Can you describe that in any way to kinda give myself and others maybe an idea about that?
Paul Bruce: I just did. People already implemented it.
Sarah Baker: Well, at the base, thereโs the vocabulary, yeah.
Paul Bruce: Sorry, people areโ
Sarah Baker: At the base, that is part of it. Sorry. Paul and I always get [Cross talk].
Alan Shimel: Sarah, go ahead. Paul spokeโyou go.
Paul Bruce: Yeah, yeah.
Sarah Baker: Vocabulary is that first piece, making sure weโre all talking about the same framework. And so that, when weโre exchanging ideas, how did you do it, how did you do itโyou know, not any one way to do it is gonna it, but knowing what your options are and talking about it with others and trying something out and figuring out if it works or not in your environment is one way to get everybody on the same playing field. But if youโre not talking about the same thing, you canโt do that.
So, itโs having the ability for a large corpus of options to apply to your situation. To me, thatโs one of the biggest things that it provides, and it also gives you ways of interacting with the rest of the business that is not IT, and kind of, โOh, this isnโtโitโs not just IT. Youโre a stakeholder, youโre a customer.โ And it includes all of those pieces for the collaboration and really helps support it.
So, to me, thatโs what the standard brings. It widens the organizationโs perspective a lot, and allows that cooperation to seep into the organization across all of these different pieces. Because to me, thatโs how it started as DevOps, DevSecOps, DevOps Observation Ops. I mean, it grows because thatโs part of its DNA is to include.
Paul Bruce: Yeah.
Sarah Baker: And so, itโs very natural that it has done that and it includes other parts of the organization into that basic thought processes. I totally could see DevOps marketing securityโI mean, you see all that. Thatโs whatโs, to me, the real advantage of thinking this way is that it includes the entire org.
Jayne Groll: Well, and first of all, utmost respect to Paul, Sarah, and the other 25 people that are working on this, because this is a gargantuan task, right? Trying to encapsulate something that creates a common framework, that has the purest intention and also, you know, if you think about the heart of why DevOps emerged in the first place, it emerged because we were not collaborating together, right?
So, you know, in the early days of ITโI always say Iโve been in IT longer than I really want to admit toโyou know, there were 25 people that sat on the same floor, in the same office building, and we all knew each other and we went to happy hour. And then, you know, the tree grew and there were lots of branches and other things. So, you know, giving a central source of truth that everybody can kind of internalize, process, speak the same languageโyou know, IT, we have so many different languages, weโre like countries that are butt up right next to each other.
And I said, I have so much respect, because this is a gargantuan task. And Sarah, I agree with what you said, the purpose is really to take this need for collaboration and optimizing automation, right, so making automation essentially members of your team, right, and then taking that out into the business because every business is a technology organization, itโsโIโm sure youโre aware of the risks. You know, weโre not telling you anything that youโre not aware of, but as I said, I just have so much respect. Because itโs not gonna be an easy road, right? There are so many different opinions from vendors and from everyone whoโs involved in this space that are going to have an opinion or whatever. And I think, you know, thatโs something that everybodyโs gonna have to deal with.
Again, the only thing I would suggest is build it in a way that itโs nimbleโdare I use the word agile, so thatโ
Paul Bruce: So, this is already done, right?
Jayne Groll: Yeah.
Paul Bruce: Like, this is not a draft, this is not an initiative. This has already been done for the past four years and is now an official standard. So, the good news is, to your point, I think, about suggestions about how to do this, we are working on the second version of this draft already.
As soon as it was out, even before it was out, we knew there were things that we just couldnโtโwe couldnโt spend the time properly addressing. Like I said before, there was a lot of refactoring, but there was a lot of careful avoidance of not already repeating what was already in a good security standard over here, or in a good risk management thing, or one of the folks that worked with us, Tafline Ramos, works at the ISO, right? And what is it, 29119, right, verification and validation, the processesโthat was already stuff thatโs already been, and great thinking and not slow, you know? Can fully be put into a modern, unicorn sneeze rainbows and fart sprinkles type approach as well as the large Fortune 100.
So, just to say, this is alreadyโweโve already put a ton of, itโs already out there, and the easiest way is, I mean, some people might be wondering, โOh, yeah, but itโs a standard, so now I gotta pay for it?โ Some standards, you do; some standards, you donโt. This one, IEEE, they cover their costs with the standards. The easiest way, if you wanted to get a free copy, become part of the new, the next version of this, right? See what weโve done, poke holes in it, have collaboration, letโs work on it together, because thatโs the point of this is to bring a framework of how we are working together in a more effective manner.
Jayne, Iโm really sorry I cut you off.
Jayne Groll: No, no, you didnโt cut me off. Just a cautionary tale. One of the reasons ITIL kind of lost a lot of its momentumโand Iโll be very frank, I owned an ITIL training company for 15 years and I was a chief evangelist, but it took 8 years between version 3 and version 4. And so, it wasnโt built, it was a monolith.
Paul Bruce: Mm-hmm.
Jayne Groll: And I think that wasโyou know, in the meantime, lots of other things happened, and I think thereโs a cautionary tale. Iโm not suggesting youโre looking at eight years, please donโt misunderstand, but this market is soโ
Paul Bruce: No, itโs probably about three, right?
Jayne Groll: Yeah, this marketโs pretty fast. Sorry.
Paul Bruce: Our target is about three, between where we officially started and now itโs out and the next revision, at most, would be another three, right, thatโs our target. It could be earlier or later than that.
Think about, weโre all citizens of the United States here, right? Or anybody in Canada?
Alan Shimel: Is that a HIPAA violation to answer?
Paul Bruce: Yeah, right.
Alan Shimel: Just kidding, just kidding, just kidding.
Jayne Groll: [Laughter]
Mitch Ashley: Iโm not showing you my vaccine card.
Paul Bruce: I ask because, I ask becauseโ
Alan Shimel: I didnโt wanna go there, but go ahead.
Paul Bruce: – we all know what the U.S. Constitution is and where it came from.
Alan Shimel: Mm-hmm.
Paul Bruce: Do we think that it was perfect and never has to be revised? Absolutely not. Thatโs what the Amendments are for. And arguably, I people wanna go back to the original version, there are some serious problems with that original version. Same thing with ITIL, same thingโremember with, I think you might have heard about The Agile Manifesto. You know, that was a drunken party with a bunch of white dudes, okay?
Jayne Groll: I know. [Laughter]
Paul Bruce: So, yeah, there were some good ideas that were encoded in there, but do they not have to be revised and worked out? Even Dave Thomas says agile is dead, long live agile. Like, what people have taken it to is not the point.
So, the goal of this is to really get a full perspective. To what Sarah was saying, itโs not justโIโve talked with a lot of the DevOps Days, various different people from DevOps Days, from a number of different, like, The DevOps Enterprise Summit and stuff, and there are some people out there that are like, โNo, DevOps is just about developers and operations and itโs about just these things, and if you start getting biz in thereโno, no.โ And itโs like, what a narrow perspective, because you know what, you had a good idea and then you screwed it up by not including other people, by not realizing there are meta learnings that come out of what you locally didโmaybe, arguably, you did wellโthat apply broadly. And when they do apply broadly, they bring us back to the same premise, which is, weโre all in this together, right? We all win when we all win. As Tim OโReilly might say, provide more value than you consume, right?
So, thatโs where we intend on taking this. Itโs not to turn into a bunch of agile consultants again or ITIL consultants, right?
Mitch Ashley: [Laughter]
Jayne Groll: Framework growers, thatโs what we call it.
Alan Shimel: Yep. Guys, I gotta pull the plug on ya, Iโm sorry, guys, and ladies. We are out of time. I think a great way to end this, though, isโSarah, Paul, where can people get more information on the standard from the IEEE?
Paul Bruce: Sure. Google it, Google IEEE 2675, you’ll get to that page if you wanna buy a copy.
Alan Shimel: 2675.
Paul Bruce: I think without being a member, itโs like 99 bucksโso, look, you know, if you wanna buy it for yourselfโ
Alan Shimel: Itโs not breaking the bank.
Paul Bruce: You probably also have, in your organization already somebodyโs a member, your Risk and Compliance are probably already members to IEEE.
Sarah Baker: Yeah.
Paul Bruce: And again, like I said, the best way to get a free copy is to come work with us on what does the next standard mean. Itโs not a huge effort and commitment like Sarah and I have put in. [Laughter]
Sarah Baker: [Laughter]
Paul Bruce: You can dip your toes in, see what makes sense, what works for you, and we love toโwe include people in that effort.
Alan Shimel: Great. Well, Paul, Sarah, Jayneโthanks for joining us on our panel today. Mitchell, as usual, great job. This is another edition of DevOps Unbound in the books, check it out. Go check out the IEEE standard. Weโll be back in two weeks with a fresh show and a fresh topic. Until then, this is Alan Shimel, be strong, be wellโTech Strong. Take care, everyone. Bye bye.
[End of Audio]




