CotŽ Show 036 - North-bound EA wit Matt Walburn

CotŽ: [00:00:00] So listen I don't want to lead the witness too much. I mean that's sort of my shtick. That's what I'd do. But why don't you just tell me just do some. This is the Rorschach test. I don't have it's not visual so I can't show you some diagrams but just just tell me stream of conscious. And then we'll get into some more intellectual intellectually sound stuff. So Enterprise Architect what what's your reaction. 

Matt Walburn: [00:00:30] I suppose it's kind of a visceral reaction just based off of my experiences both working with enterprise architecture and actually being a part of enterprise architecture. 

[00:00:43] To me it's maybe a term that we need to freshen up a little bit to avoid some of those knee jerk emotional reactions that I think a lot of a lot of folks have to that phrase. So I guess visceral reaction is kind of you know a centralized command that can control heavy duty you know portfolio management of technologies and applications and solutions more of a governance mechanism rather than I guess what [00:01:18] I would like to see it become which would be a mechanism for. teaching, coaching, guiding enterprises into. [10.1] 

Matt: [00:01:28] Maybe you know greener pastures or some of the approaches that we have available today. 

[00:01:32] Yeah no I think I think I think you've hit on all the stuff there right. Like we'll have to have to use that to detail out stuff. It's good it's a good agenda for discussion in an arc if you imagine a 250 page 300 page book that you just described the arc right at their end and do it to run to the conclusion I think you know so. So I've been looking around at my ongoing project here is to figure out what I'll turn out to disclaim too many cheesy terms which is just run with the cheesy terms right. Like so what is what is a cloud native architect or a devah. I mean a dev ops enterprise architect or to put it another way if you're doing sort of like maximal DevOps and you know I generally when I say the word cloud native I think of that is the superset of dev ops and a large part of figuring out what an enterprise architect does is defining what that supersets stuff is. But like if you've got like a dev ops What do they call it in the literature culture organization in place like What are the enterprise architects do. And I get this question a lot. And it's easy to theorize about and I think I think what you ended up with is the most helpful and hopeful answer that would be good to shoot for. Right which is in order to do all of this stuff this transformation we need you know it's right there in the world we need to change and it seems like Enterprise architects should be part of it and then working backwards. 

CotŽ: [00:03:01] You know it's sort of like and this and this will get us back to talking about like you or your views on the bad part of enterprise architecture. Right. [00:03:11] It seems like now what enterprise architects do and I would say this is only half straw person is governance. [8.6] And what does governance mean. Right. It's not just because we're Americans living in the Central time zone that we think poorly of governance. I think everyone does. You can you can read you can read the coverage of you know the NIH from the register and other people to see that people don't like governance. But it's just like in general governance is not about doing new things it's about doing the same thing over and over again in as safe and baroque and costly as a way possible maybe not costly. That's the negative view of it. But anyhow there's my here's my post leading the witness. 

Matt Walburn: [00:03:55] I mean I think a lot of it has to do with minimizing deviations from whatever we are they defined as as the norm. Right. So I think to your point right like it's sort of contradicts almost everything that I think a you know neoclassical dive ops type person might want to get out of you know their variants as an engineer or an. You know as an individual contributor is this idea of [00:04:23] I want to minimise change versus figure out how to support change. [3.8] 

CotŽ: [00:04:28] So so so so introduce yourself briefly and give some credentials for this topic right like you've got an interesting set of primary sources to draw upon for for this yeah sure. 

Matt Walburn: [00:04:40] So again Matt Welborne I'm currently a principal transformation consultant with AWS although a caveat that all of the views expressed from this point forward are definitely my own and are intended to represent. 

CotŽ: [00:04:57] That's right. 

Matt Walburn: [00:04:58] My employer seems to just get that out of the way right. Re way up in the US for a while prior to that I worked at pivotal software where I did a similar consulting helping organizations move from kind of traditional siloed waterfall delivery type organizations to BOP's and micro services. 

Matt Walburn: [00:05:22] Nirvana prior to that actually worked for Target large you know big box retailer based out of Minneapolis and let a lot of their work in the Obst and transformation and cloud migration space and then also. 

CotŽ: [00:05:39] Also I mean I met you at pivotal people could reverse engineer that. 

[00:05:42] But I think I think another thing was to link to it is you did some good some good collaboration on on one of the papers we have about building a platform. Right. Like building a cloud platform now obvious obviously being from pivotal. Our goal is to make you think that instead you should buy your platform from us right which which the paper does valiantly but like if you care to you can kind of strip away that call to action as they say and it does show a. Its not necessary. Well its part of enterprise architecture but it shows a method method not methodological a methodical way of thinking through. If I have these goals sort of strategically like what kind of platform would I put in place. And so thats a I mean that was a good piece of work yielded an interesting sort of thing to look at and I think I think eventually when I figure out what cloud native enterprise architecture is like that kind of platform work is a huge. Its like kind of I don't know I don't know tell me if you think this is true but I think I think that sort of some of the resulting work that an enterprise architect would sort of guide the organization towards right in the same way that in the past for better or worse in the past in the present like and they would say like we should use any to be bright like like there seems to be a lot of sort of top down saying this is the kind of platform we would use to achieve whatever goals we set out. 

Matt Walburn: [00:07:15] Yeah I mean I would agree I think I mean to your point I don't exactly know what a cloud native enterprise architect is either. And so we'll explore that a little bit. 

Matt Walburn: [00:07:25] But we definitely agree that I think no matter what the answer book ends up saying that platforms that enable you know the developers in that business to move faster with less undifferentiated you know heavy lifting duplicative work. I think obviously part of the solution. 

CotŽ: [00:07:44] So you listed a bunch of things that you would do like let's let's delve into those. I mean what if you if someone was like hey man I want to pay you like half a million dollars a year plus some awesome equity and I want you to come to let's call it let's call it since you're well. We'll figure out what type of business it is as we talk. You can define that. But I want you to come work at Acme. Right. And you're going to be the head of enterprise architecture right. Like we're just like we're just a few bad martini lunches away from being a Circuit City or a blockbuster. Right. And so the board has authorized that we're just going to kind of like we're going to bring in new people and we're just going to we're going to give them like that kind of like. And is it any license to just like go crazy. Right. And I mean that in a good way. And so now you were going to give you an offer you can't refuse to be the enterprise architect. So like you know you get all jazzed. Probably like any good person you go on a month long vacation before you take the job. Right. Maybe you finally catch up on some Martin Fowler books you read. You read that DevOps handbook. Right. You're just catching up on things and maybe even go to a conference that you had previously scheduled you kind of floated about and then it's it's Monday. Right. 

[00:09:02] What are you what's going on in your head what are you going to start doing. What's happening. 

Matt Walburn: [00:09:09] I mean I think [00:09:10] the first thing that I do in any scenario like that is really try to understand the current context in the current mindset of both you know the I.T. organization as well as you know the customer whoever that may be right whether that's the business or you know actual customers that are somehow consuming I.T. services. [20.5] So really trying to understand like where like where are they starting from. To make sure that I'm being you know respectful and inclusive of whatever has happened to date. So I think that's obviously a great place to start. 

CotŽ: [00:09:46] Or. So you have sort of like I guess a baseline if you will. And just just to just to kind of like you know bullet out what you're saying and maybe this will be annoying. We'll find out. But like [00:09:56] the baseline the baseline is basically like what is the business. [5.1] Right. How you and we'll make it a business not a nonprofit or a government thing. Right. Like the point of any business as far as I can tell is to make money right. And so how is it that this company translates whatever it does into making money. What's the business. Right. Who are the customers. But then [00:10:21] you also added an interesting other part which which is internally in order to service this business who are the internal users and customers and what do they need to do. [10.1] And then and then I think as you alluded to that [00:10:34] the next step after that is sort of like so in the I.T. department like what do we do and. Right like how are things setup right. [9.3] 

Matt Walburn: [00:10:44] Right. And I guess even before we get into like how are things set up in IT. 

Matt Walburn: [00:10:49] And I think this is kind of a hiding behind and the bullets you just listed out is I really want to know. Thats an overloaded term but I'll use it anyways like what it is. [00:11:00] What is the operating model between you know new ID is getting generated in the business ID is getting funded and then ideas then translating into you know software or other solutions being built by I-T. And what does that process look like. What is what does that overall value stream an operating model look like. To get that idea out of the developers head and then ultimately into the hands of their customer. [25.1] 

CotŽ: [00:11:25] Right. 

Matt Walburn: [00:11:27] And so I think theres just a lot of there's a lot of processes from you know product development in the business from finance and how things are funded from H.R. and how you know resources are staffed and allocated to projects and then ultimately then the internal I.T. processes as into how they actually then deliver that work once it's been funded and staffed. [22.4] Yeah. So those are the things that I would look at first. 

Matt Walburn: [00:11:55] And then basically you're trying to [00:11:57] put you know that vision that strategy that roadmap of OK this is what you're doing today and you've got me in here because by some admission even if it's not much it's not working for you. [14.5] 

CotŽ: [00:12:12] Right right. I mean I mean let's say let's let's let's say in their press releases they're like we're excited to begin our new growth initiative for transformation into Bitcoin AI Watson harvesting. Right. Like they just got a word salad of awesomeness and then but behind closed doors right. You've met with the CEO and the CIO and they're like shit is fucked. Right. Like you got to fix that. So there's there's no internal illusion about like things or things are not operating well. Right. And I mean I guess I'm kind of setting up a bit of a thing here like there's a question of what a cloud cloud native enterprise architect would do in a functioning company. But let's assume that if you have a functioning E.A. function why would you change. Right. Like if it's doing everything you want then there's no reason to even have this quandary so the bullet point or you know whatever the word is patentheticalize here now and this is this is one of the what would you call it. 

CotŽ: [00:13:11] Her theories the foundational theories that I'm starting to evolve. [00:13:15] What a cloud native enterprise architect is and that is it only has to do with custom written software. [6.3] Right. Like you said the word developer a couple of times. Right. So a cloud native enterprise architect would not care about your Bring your own device initiative only in so much as that might be a UI or an interface to pump their software through and they probably don't really care about your desktop management and they probably don't care. They might start to care about your data warehousing business intelligence stuff. Right. If they wrote custom software they don't care about your ERP stuff. Probably at the moment all you're off the shelf software but anyways. So like do you think that good reduction down and obviously there's the footnotes of well you will care about them if you need to interact with them. But that's kind of like a big chunk of I.T. that I think a cloud of native enterprise architect clears off. 

Matt Walburn: [00:14:14] You know I think it all is in the details of what we're seeing. 

Matt Walburn: [00:14:18] what a cloud-native architect is and I think obviously there's a lot of shades of gray in and in that term. I guess I'm thinking get to your point if if you're trying to do if you're trying to do agile if you're trying to do automation cantinas delivery all of these modern practices like no you're probably not applying those against your legacy ERP or CRM type systems. That said I do think you know whether you're you know cloud friendly or cloud native in a lot of enterprises are looking to move all of those things to the cloud. Right. 

Matt Walburn: [00:14:54] Whether or not there those you know high velocity systems or differentiation or the slower you know systems of record they still want to move all of that stuff out of their data centers because they don't want to run data centers anymore. 

CotŽ: [00:15:05] Yeah. 

CotŽ: [00:15:07] And so I think I think you have to tease that apart right. I think there's this cloud native for all of your go fast stuff and then maybe just this cloud cloud of friendly or cloud hosted model that you have to be able to basically swing between and kind of dial you know up and down depending on the use case. 

CotŽ: [00:15:26] Yeah yeah. All right. All right. So let's delve into the first thing you said. So. So you go home on Monday night. Yeah I don't I don't know. I don't know what you eat up there. Maybe have a good hot dish. That's that's all I know. Right. Hot dish right. And and maybe maybe maybe you have some kale. I don't know. And then and then you come in on Tuesday and you're like OK job one. [00:15:47] How's the business work. So how do you figure out how the business works. 

Matt Walburn: [00:15:53] Yeah I mean I think a lot of that is building relationships and actually talking to the business and figuring out what a lot of their processes look like. [16.2] 

Matt Walburn: [00:16:04] What are the what are the experiences that their customers are interacting with that are supported by I.T. or so experience could be you know a digital experience through a you know e-commerce or mobile device. It could be you know an operations or call center experience. It could be any number of experiences and touch points with the customer regardless of whether or not they fall into if you know that whole digital experiencing that we that we like to talk about. And [00:16:36] so yeah I mean that's primarily where I start is trying to figure out what are the things that are actually powered by the that the businesses is consuming and then try to figure out again back to back to that operating model statement I made earlier. How successful or not successful are those products supporting their actual business objectives. [22.4] 

CotŽ: [00:16:59] So it's sort of like working backwards to use to use comically ancient enterprise architecture. What do you what do you call when you find caveman drawings in the cave or those. Pictographs or something. But you know in UML land you've got that stick figure. Right. The end user. And then and then the stick figure typically has a line going to a box right. And so that that's that's sort of like the first thing you want to hunt is like who are the stick figures. What is the line. And then maybe we'll start looking at this box to figure out what's going on because I think I think this is I mean I don't I don't like all things. I mean if I think about my own understanding of the history of computers it starts basically with like targeting missiles and trying to decrypt stuff and then doing payroll. And I guess the mission to the moon you know cataloguing what all the things are. And then there is there's this phase that's like there's a bunch of stuff written about like you know human computer interactions. And then Microsoft I mean Apple and Microsoft like steal the they don't steal. They evolve the Xerox PARC stuff and then you got you guys and so forth and so on. Right. And [00:18:14] but it seems like only recently there's been this like focusing back on the stick figure unlike what the humans are doing. [6.7] And so so that I don't know. 

CotŽ: [00:18:23] I mean I mean I'm just kind of rounding up what you're saying like it seems like it seems like that shift in thinking becomes [00:18:31] an important thing for an enterprise architect to start doing is to really think about the the the the workflows that not even the workflows but like what are the people doing and what do the people need. And you could round that up with the simple word design right. Design means a lot of things like that. [20.4] 

[00:18:53] Yeah that's what I was going to say that it's almost like the enterprise as you know architecture equivalent of like user centered design or you know design thinking it has to in it you know maybe it's an old term but we still use the term working backwards from the customer like extensively at Amazon and AWS. But I think that's really the right approach. Right because you figure out like what is the actual outcome in terms of company generating money. I think that's the ultimate outcome. But there's a lot of other you know outcomes that need to need to exist to support that ultimate outcome and then working backwards from there into you know what are the various levels of either user experience or API or master data or all of these legacy systems that ultimately connect the stick figure in the human to the technology that I.T. is delivering. 

CotŽ: [00:19:48] Yeah. 

Matt Walburn: [00:19:49] So I really try out that that end to end picture. As one of those first steps towards at least mapping out the current state. 

CotŽ: [00:19:58] Yeah I think I mean just to opine a little bit on my own. And I'll just start doing that. It seems like it seems like what you're going over is a huge sort of like philosophic underpinning of like Agile right and and agile especially in like the overly orthodox like Agile which is to say sort of like if you if you're a what they call the conservative Supreme Court people they're textualist or something. If you're like a textualist for like XP and Scrum it's basically like developers will do they will know that they will only do things that directly benefit the user. And I'm kind of reading into it but it's sort of like as a counter-example you have to go through hoops to argue that you should update the JVM. Right. Like like it's sort of like what value is updating the JVM to the user. And then you're sort of like blah blah blah blah blah. 

CotŽ: [00:20:56] Right like I mean there's certain things of like you know what value are the tires and the parts you select in your car and stuff like that which ultimately sometimes there's not a value. Like if I'm guessing whether or not you have it whatever a Hemi is in your truck. 

CotŽ: [00:21:12] It doesn't really matter. It's just it's just sort of like fun branding stuff right. Or. Or whatever. Anyways that's a bad analogy. 

CotŽ: [00:21:20] I'm trying to stop making analogies to the real world about software and follow the practice that software like software anyhow. But and then I think I think another interesting thing is I mentioned this because you see it pop up all the time and and trying to figure out if it's a required tool for like cloud native enterprise architecture is similar like [00:21:42] you do sort of like value stream mapping as sort of like let's only focus on what we need to actually do to satisfy the customer or the outcome that we want. [10.3] 

CotŽ: [00:21:53] And so I mean I think I think based on what what we're saying here that is a pretty big foundational shift and what you should care about is what is the outcome with an end user that you want and I'll wrap up my monologuing here with this but like I think when I read read through and just kind of remember anecdotally a lot of the lore from the last 20 years a huge amount of it is about optimizing the business running and there's not as much emphasis on optimizing the customer experience if you will. 

Matt Walburn: [00:22:27] I would agree that I think up until the last few years I think the job of E.A. has been optimized. 

[00:22:32] I'll rephrase it a little bit differently as to you know achieve the business outcomes but optimized on cost is optimizing for costs and efficiency rather than optimizing for uh... What I would say like throughput and responsiveness. [19.9] Yeah. So I deliver the business outcome but since I was optimizing on cost it took maybe four times or five times as long as it should of and maybe quality wasn't where I wanted it to be. Yeah. 

CotŽ: [00:23:01] So yeah and I think I think across all of you know the career that you went over right this and it's always annoying to like state this stuff because because you and I and other people in our position see this stuff ad nauseum right. Like it's the equivalent of like your doctor telling you that you should eat more fruits and vegetables. You like i get it man right. Like that's that's the way things are. But man or woman I guess you know I don't know is is is do discriminatory or has that become non - What do you say - binary. I don't know. Anyway this is sort of...you know it's sort of like your your business is likely not going to remain static. 

CotŽ: [00:23:44] So you're going to need to come up with new ways of doing your existing business and new businesses and therefore [00:23:50] in a static business where what you're doing really never changes then you're just basically optimizing and growing your market. And so that view of what an EA does is pretty good. It's sort of like I sort of figure out two things one is like how do I pull more costs out of this by doing centralization and shared resources and also ensure that the wheels don't come up right. Like my. The number one strategy for every company that's existing is what I like to call the don't fuck it up strategy right. Like don't take your current business. But then the second thing in this static view that an EPA has to do is like as we expand geographically in other markets how do we deploy that I.T. to it. [34.9] 

CotŽ: [00:24:25] Right. Like they don't figure out new widgets and stuff like that. But anyhow the whole point of being more customer centric and you could get those things by being customer centric as well. But I think I think customer centricity is extremely valuable necessary if you will to the businesses that actually need to do new things there. 

CotŽ: [00:24:43] So so as not to stump you with another question that ends in a period like I did earlier. 

CotŽ: [00:24:49] Now now you mentioned you mentioned one thing that I think I think and then maybe we'll get off the business side and into the technical stuff later like you use the word like outcome's many times which in business speak is a fascinating term, right. "Outcomes" are a good term. And I think I think theres some tangental collection this is this is an even more of a landmine term. Right is like strategy right. And so so weve got these. Let's talk briefly at least your thoughts if you have any about what an EPA does associated with these three terms right. So we had outcomes right. And then there's there's like strategy which will be interesting to see what you think of that what you think what you think that is. And then there's another term that we kind of talked about earlier is just sort of like what did you call it the operating model or how the business functions. Right. And these are kind of all the same thing. 

CotŽ: [00:25:43] I don't know which ones are or tails versus dogs and stuff but like so let's start with outcomes like when you're talking about outcomes like one what does that mean? And two like what does EA do with that? Like what - How is that an input to what their job is? 

Matt Walburn: [00:25:59] Right. I mean I think you might we might need to get specific about an example to talk about outcomes. But I guess I don't know let's let's use like e-commerce for lack of a better or better example. 

Matt Walburn: [00:26:13] Right the outcome is I want to be able to increase revenue by selling stuff on the Internet. Yeah. So that's life. 

Matt Walburn: [00:26:21] That's one outcome. So a second level outcome would be hey I want to be able to do that through a web browser or I want to do that through a mobile phone or maybe I don't want to do that via some third party partner where I'm actually selling stuff on their Web site. Right. And so you start kind of trying to figure out OK what are these different levels of come. Right and so I just went from one to three. I think you could start to think of those three as your top level. Almost at the top level of your architecture but also like a set of products or a set of outcomes right. All kind of use that the words interchange interchangeably. And then I think from there you start decomposing even further. Right. So if I am e-commerce regardless of where the customer is interacting with me you know whether it's you know a web browser a phone or or. Or the third party site like I mentioned like you still know that hey I need to be able to show the customer my products. So I need a catalog right. So that could be an outcome I need the customer to be able to easily discover and find my the products that they're looking for. 

Matt Walburn: [00:27:34] So that that outcome might Man it manifest itself as some kind of a search capability. 

Matt Walburn: [00:27:40] You know you can decompose any any any web site that you're looking at today into these different outcomes. Right. That all manifest and integrate together to produce the higher order outcome of selling stuff on the Internet. So I think [00:27:56] I like to look for those kind of time back to the earlier statement when on first coming in I want to really understand what are all those experiences and those could be you know it's human to machine experiences like I'm talking about or they could be machine to machine experiences in the form of API and other data services. [24.7] It's really just trying to understand the full landscape of experiences and then start decomposing those things into their smaller business outcomes and ultimately arriving at what almost looks like almost like an encyclopedia or a taxonomy is the word that I like to use for describing how to all of these pieces and parts actually connect into those those outcomes. 

CotŽ: [00:28:45] Now now now to that to that taxonomy thing now apparently speaking of e-commerce I have the TOGAF 9.1 spec arriving today. So all have one way of doing I think. I don't really know what is. But that's why I don't know I've heard of it. 

Matt Walburn: [00:29:01] I think I think I musta been a. Enterprise architecture flunky. Yes. 

Matt Walburn: [00:29:05] Well TOGAF as a thing. 

[00:29:08] As a total aside like I was I was lucky back when I was at BAMC and this must have been when that when this kind of stuff happened in 2004. So one of the documentation people or tech Bub's as they said they lent me the man I have it back here but they lent me one of the core books on I tell er ITIL ital and it was I think the service delivery book. 

CotŽ: [00:29:29] And I read it cover to cover and I feel like I feel like having that knowledge has made me hundreds of thousands of dollars over the years. Like like when I was an analyst and then about strategy and and the whole point rather than being snarky is sort of like. So I mean toga toga would you say Toga is one of these things that people who revile yays will like mention and I'm always kind of like by Modahl light. My theory is they've never actually read it. And so I think I mean it's always good to know your enemy so to speak right or to know your potential ally. Like are they a Hitler or Stalin. You don't really know what what you're going to do with them. So it will be interesting to read that and see which to it. So anyways Obviously you don't use TOGAF because you didn't tell me about that. But like how do you how would you like document this taxonomy. Like what what do you do. Like do you have like a bunch of post-it notes on a whiteboard somewhere or are you going to like without Vizio or what are you going to do. 

Matt Walburn: [00:30:25] As far as tooling goes I mean yes generally [00:30:28] I would generally sit in the room with I.T. and the business and actually walk their experience backwards. [6.2] 

Matt Walburn: [00:30:34] Right. 

Matt Walburn: [00:30:34] So I categorize things into you know things like hey what what business are you in what business unit within that business within that business. Who is the customer. Right is the customer outside the company. Is it inside the company. Is it a third party partner. What's the entry point into your company. From an experience perspective and then once we actually can get anchored in those initial things really start dissecting. So if this is your experience and I'll keep leaning on e-commerce like water all of the other outcomes that are what are the other activities that that customer needs to successfully be able to perform. 

Matt Walburn: [00:31:14] Right. And I don't know if I'm answering your question but. 

CotŽ: [00:31:17] No no no. Totally. 

Matt Walburn: [00:31:18] It's it's a lot of interview. 

Matt Walburn: [00:31:20] It's a lot of whiteboarding it's a lot of what's the term we made this up a few years ago Boardio sensitive Vizio with Boardio are right it's a white board, White Board Vizio drive. 

CotŽ: [00:31:33] Yeah. Yeah. 

Matt Walburn: [00:31:34] So it's I mean it's it's a pretty organic process at first now. 

CotŽ: [00:31:40] Now just to interrupt you. I feel like if I was doing some sort of academic study of this right. Tell me tell me if you agree with these selections are ones you would add as well. I feel like they're sort of like four mindsets. Let's see if I remember all of them that I would go over. I would I would start with a classic Michael Porter. I don't know if you've ever seen this but he's got this from the early 70s like this. You've got the five forces and I figure with your suppliers competitors and God knows what sand substitutes complient. And then and then that translates into a way of diagramming a strategy and it's basically this big this big triangle and that is a big rectangle with a sideways triangle on the end. I don't know what you call that shape. It's like it's like a really blunt tipped arrow. And and in that you model out how all these forces kind of work and it sort of gives you an idea at least from a strategy perspective and what you do business. So that's that's sort of like I remember reading this great book The Lords of strategy and that was one of the breakthroughs in thinking about how a business operates that you could actually documented. And then I think fast forwarding way to the future. You've got remember that book that's like like sketching out business models. 

CotŽ: [00:32:56] I forget what it is but it was really popular about four or five years ago and it was like this tool kit of a book and it had a template and you would see all the startups would have to do this and they would sketch out what their business model was right out to look up what that book book is. And so you got that and then and then you've also got a value stream maps which which so far seems like one of the best things. I mean you basically just described well you didn't describe all of value stream map but you described a hue. The input to it are a huge part of the input to it. And then I don't really yet know what a Wardley map is but I think it's somehow associated with all of that. Right. 

CotŽ: [00:33:34] And then you might just have like whatever the fuck diagrams I make up to draw on the white board. Right. Like a lot of the I've been in a lot of these sessions and a lot of the times you just got to agree on a bunch of what your Boardio diagrams to do things in. 

Matt Walburn: [00:33:51] But I think I mean a lot of the point of doing all this and in depending on the size of the business or the portfolio you're trying to look at it can it takes a lot of these conversations but [00:34:01] you're really trying to look at least what I'm trying to look for is how to all of these components align with whatever the outcomes are that the business is trying to reach. [9.9] 

CotŽ: [00:34:11] Yes 

Matt Walburn: [00:34:11] that's number one and then number two is [00:34:14] how much stuff have I duplicated across various lines of business business teams and I.T. teams because what I've seen is that there's a lot of duplication and a lot of rework going on [13.5] and looking to rationalize those two things against us. What I'm assuming since since the enterprise hired at the expense of you know to figure out what you said that EA or CTO to come in is [00:34:43] what is that future vision look like. [2.4] 

Matt Walburn: [00:34:46] Right. So moving from you know in the heavily siloed whether it's I.T. or or business units too much more collaborative transparent end to end type working models and operating models. 

Matt Walburn: [00:35:03] So trying to. 

Matt Walburn: [00:35:04] And then [00:35:04] how do I actually get those business ideas through the I.T. delivery or software factory more quickly. [10.6] 

CotŽ: [00:35:16] Yeah yeah that makes sense. So one of the so. So talk about like the idea of finding like duplicate things like I mean I make you say a bunch of obvious stuff just to get it out there. But like why is that. Why is that bad or good. Like like what do you do with duplicate. Let's call them services. Right. Like a service being like a thing it does to make sure that to help the business right. Like what are you what are you what are you doing when you're looking around for duplicate services. 

Matt Walburn: [00:35:47] Yeah. 

Matt Walburn: [00:35:47] I mean I guess so [00:35:49] if we anchor towards this idea of we want business ideas to be able to be put into production faster. 

Matt Walburn: [00:35:57] It's a really interesting time to value and shortening the cycle time for how long it takes to deliver something. I think a big piece of that is trying to choose the right word. I'm going to say democratising for lack of a better word. 

Matt Walburn: [00:36:12] Your internal data and your internal integrations to that data. Right [27.6] and so you're trying to create these efficiencies where you have well-known data sources you have well known interfaces and ways to connect to that data so that you're not burdened with either recreating that data or recreating the interfaces or just doing unnatural things to arrive at again to use the outcome. Trt to it at these these outcomes that are reusable and kind of ubiquitous throughout throughout the enterprise. Right and so that's not so much like an I.T. service that [00:36:48] I'm talking about duplication of although that is also there it's more about what are these what are these sources of true systems of record. You know repositories of master data et cetera that we want to figure out how to open up and make it accessible to everybody so that you don't have people recreating we all. [22.0] 

CotŽ: [00:37:11] OK. OK. Oh there's a there's a lot of I'm glad I asked a lot to pull apart there right because. So. 

CotŽ: [00:37:16] So one it's I don't I always forget if notable is the right word. But it's interesting as my mother would say that you didn't mention anything about cost reduction right like that if you have if you have duplicate things and you're paying too much about it which Which isn't to say that's not a concern but that's that's where my mind goes to when I think of ferreting out dissertation and then the second thing and you've mentioned this before I ask you about duplication and I see I see exactly how this maps now is like [00:37:49] if the goal you know we want fast cycle times or whatever a way of achieving fast cycle times is cutting down. As Jeff Bezos would finally say right like any time you have to communicate it's a failure. Right like like essentially any time the time it takes you to understand something and the time it takes you to research and comprehend things like if you let me put it another way if you don't already know something it's going to cost you a lot to figure it out. Right. And so you want to make figuring out these data sources and finding them easier and duplication is one of the leading causes of confusion I guess. 

CotŽ: [00:38:26] And therefore not having fast access and transparency and things like that so it's not only that you would want one did you say MDM. I [46.5] haven't heard that in years. That's great. Anyways like you want you want to have a never mind. Let's talk in logical terms which is to say from the perspective of someone consuming it it seems like this. Never mind what the physical implementation is. Right. [00:38:53] But in a logical way you want to have a centralized pool of data right and you want to have as few if not one way of knowing what that data is and how to access it and therefore that eliminates that comprehension time that time it takes to understand and allows the individual teams or people using it too. [17.6] 

CotŽ: [00:39:11] It just allows them to act faster. And so duplication can can cause a slowdown in the cycle of comprehension which means that takes you longer to get a product out the door and understand things. That's a fair blowing up summary of what you're saying. 

Matt Walburn: [00:39:26] Well that and [00:39:27] it prevents you from I mean if these things are well-defined and they're published and people know about them and they don't have to spend a lot of cycles like figuring out how they work they also that don't have to. In the absence of having any of that knowledge have to recreate the same basically the same solution themselves. [20.0] 

CotŽ: [00:39:47] Right right. 

Matt Walburn: [00:39:48] So I think rationalizing like reducing the amount of duplication that also enables it an able seems to focus on the things that are different and differentiating and evolve and value added rather than continuing to you know make another copy of you know the customer data or purchase history database. 

CotŽ: [00:40:09] Yeah, and then and then and then there's another thing implicit in there is like you've got a bunch of duplicate sources you're inevitably going to have conflicts and things that contradict each other. 

CotŽ: [00:40:20] And so - 

Matt Walburn: [00:40:20] Exactly. 

CotŽ: [00:40:21] - that's not cool. Right. All right. Well we're going to run out of time here hopefully hopefully we'll come back and we can talk about. We'll get to the technology part but maybe we can make this this. 

CotŽ: [00:40:32] I'll just assume the answer is yes but we can make this part of the we'll call it the the interface the we'll use network terminal is this networking What do they call it upstream and downstream. 

CotŽ: [00:40:45] I forget but you've got you've got the. 

Matt Walburn: [00:40:47] Like "north/south. 

CotŽ: [00:40:48] Yeah. This is the North interface to the business. 

CotŽ: [00:40:51] And then maybe next time if you deign to come come talk with me again we'll have the the south down to the the technology sort of thing and there may well have a third episode about in the long quiet nights of the enterprise architect. What is it like to talk to themselves. How do they counsel themselves to do okay. Anyways so 7 Let's wrap up with like. So defining this future vision. What's that all about. 

Matt Walburn: [00:41:19] So what I talked to customers a lot about a lot is moving away from thinking about the work that they do as individual activities. Right. 

Matt Walburn: [00:41:29] So I provision servers I test software I write business requirements. 

Matt Walburn: [00:41:34] I you know do all of these little individual activities that on their own pretty much provide no value. Right. And moving from the model of being all right they. 

Matt Walburn: [00:41:45] They only have value if they're done together and if they're done together in a meaningful way. And so I guess the short version of this kind of mapping to the [00:41:56] northbound API into our business is figuring out what are the collections of activities that make sense to group together in the context of a product team. [11.1] 

Matt Walburn: [00:42:08] Right so. 

Matt Walburn: [00:42:08] So that's my solution for [00:42:11] how do you ensure that you're actually delivering and anchoring against a business outcome. Well it starts with from organizing your people around activities to organizing them around those outcomes and the way that that manifests itself with technology is by creating technology products. [14.8] 

CotŽ: [00:42:26] So a large part of the vision is you could also call it the. So what are we going to do now. Like like just like how what does this mean as far as like how we're going to organize people. I mean I guess this is the part that starts going towards to use our vernacular the southbound interface right is like so how. To some extent you probably talk with the business about things they need to do and like how they should start changing what they do. But then also you're starting to think about what are we going to do and IT to meet these needs and what exactly. What are the things going to look like and what does this mean is for. And I think the people who complain to us about enterprise architecture I think this is where they start to think that we're stealing their lunch and we're just like you know pissin in their soup or whatever. 

Matt Walburn: [00:43:13] Right right. 

CotŽ: [00:43:14] They start to get really touchy about this. And and this is where I guess you know this is where you get all these things like the having a three year plan is meaningless right and where people don't like the word vision right. Like how can you how can you have a future vision and you need to have this iterative lean start up approach to stuff. 

CotŽ: [00:43:33] And if you set things in stone because everyone knows this vision is going to be taking to finance to set probably like a 12 month finance plan and then that's a constraint that you brought on and then you're talking about reorganizing things and essentially you're starting to think about what you're going to tell people to do or maybe. Right. And then also technology selections and things like this but this becomes this is the start of the conflict around stuff. But tell me tell me if this is wrong but I think I think to a huge extent [00:44:06] when you look northbound this is sort of like what makes the business feel comfortable. Right. Like if you can produce this you sort of know what you're doing and you have a plan. You can start to become relie or it seems like you will become reliable you'll making it you're making a promise of what you're going to do and then people go off and start start to do things. [25.3] 

Matt Walburn: [00:44:33] I think that's true. Yes. We'll get into this next time I guess. But I think it's supremely important that as you start pivoting to you know whether we call this a a cloud native operating model or and digital operating model or product operating model you know pick your buzzword I think when we talk about these product teams the businesses implicitly part of those those product teams. 

Matt Walburn: [00:44:58] Right. They're not separate. Right. They're not they're not in the north. They're not exclusively in the north API and I.T. is exclusively in the south. There's a blending of those things. That needs to happen on a nearly continuous basis in order to make sure that those products that these people are producing continue to meet the needs of the business and its customers. 

CotŽ: [00:45:19] Yeah. 

Matt Walburn: [00:45:20] And just to kind of cap it off I mean [00:45:21] I think the role of the cloud native enterprise architect is to facilitate and enable the IT journey and the business journey to that operating model. [11.6] 

CotŽ: [00:45:33] Yeah yeah. Now that makes sense I think. I think one way of summerizes that would be like one of your first tasks in the vision thing is sort of like figuring out how decoupled the business I.T. should be. Right. Which is like like what you just said is not very decoupled right. Right. And I think I would imagine in a lot of large organizations the current answer is the opposite. Right. Like the business will be very decoupled from I.T. and do you use another old trope. Right. Like we're just going to throw stuff at this point. The business would throw the vision over the wall to I-T and come back to them in six and 12 month increments and see how they're doing. But that's that's that's sort of like one of the first southbound sorts of things you do is you tell the business that they're going to have to come a little bit south of you and start to start to couple with you more frequently. 

CotŽ: [00:46:27] All right. Well that's great. Well this has been awesome. As always no need to worry about filling time. 

CotŽ: [00:46:37] So we'll have to find time for you to come back and and then and then we'll mutually lead each other down the path of figuring this out. But I think I think the next part that will be interesting is as we say the southbound stuff like So what does this mean in the I.T. Department. You have anything you want to add before I wrap up here what what's your Twitter account or other other other details you want to give people. Oh I'll go up with sure. 

Matt Walburn: [00:47:00] Yes so I'm at Twitter. @mattwalburn on Twitter. 

Matt Walburn: [00:47:04] I don't know if I have any other super personal details to share but excited to keep this conversation going. 

CotŽ: [00:47:10] Yeah I mean like I said you always have a lot of good ideas based on real life experience about what's going on here. Well as always this has been the cotŽ show variety podcast where I put conversations like this and other things. If you're interested in topics stuff like this and also some more wacky stuff every now and then I do an old drunken retired podcast recording with my old our old friend Charles if you remember that. You just go to cotŽ that show. That's the kind of world we're living in now. You've got tealeaves of show you can do an accent in the east. Just go to see otiose show and you can subscribe to it and pick through the other episodes whatever you like there. You know if you want to refer to other people or put it on iTunes or whatever. That's great. But really you should just tell me if you like the show. And also if you've got any ideas about cloud native enterprise architecture you know pro or against or whatever anything you want to send my way. I mean this is my current big project that I'll be working on for a while so I'm interested in all sorts of content and thanks again for being on Matt. And we'll see everyone next time. Bye bye.