Iasa conf2016 experimenting with digital architecture - slide share

Editor's Notes

• #3: In this talk I will provide some experiments that you can use to help you practice and create digital architecture.
• #4: What is digital architecture? I don’t know … that’s the honest answer … there is no “industry definition”. Best I can do is say that it is fractal and recursive. That seems to be the way nature manages complexity. But seeing as we don’t know we need a way to explore what it might be. That’s where experimentation comes in …
• #5: Why experimenting? It’s the best approach for this kind of challenge. Tom Graves covers it really well. Bottom line - if you want to be a business anarchist you have come to he right place!
• #6: My intro – who and what am I? In terms of my career I’ve worked on both sides of the fence – the supply side and the demand side but pretty much always in the change space. I’m never happy with the status quo. Here’s a quick run though of my career … BBC – setup and operated enterprise MS Mail system for 8000 users – then spent 6 months skiing in France! Joined ARI to work in retail systems – Channel Tunnel and Bangkok – focused on system implementation and adoption. Time to specialise – jumped to supply side – Digital Professional Services. 6 great years – e-mail/Exchange world domination with Tony Redmond & co. Ireland, US, Switzerland, Belgium and Australia. Deep into Microsoft world. After returning from Oz I spent some time with SureSkills and then went back to HP to work as an infrastructure architect on a big financial services outsourcing project. I could resist no longer – 6 years with Microsoft as both a consultant and a Technical Account Manager. Supply side blues – needed to move to the middle – EY as an architecture consultant. But too much of a disruptor for Big 4  Building a new business now in Digital Architecture … early days but going well. Last but not least – I’m more than a tech architect – for example I’m an ENFJ …. More on that later!
• #8: What is digital architecture? Here are some ideas …
• #9: What is Digital Architecture? 0. Let’s start at the start. Are there any good definitions of architecture out there? How about the IASA definition? 1. The art or science etc. …. What does this phrase actually mean? You see it in lots of things. Let’s explore.
• #10: The whole is greater than the sum of its parts Architecture[edit] Architecture is often argued by design academics and those practicing in design to be a holistic enterprise.[19] Used in this context, holism tends to imply an all-inclusive design perspective. This trait is considered exclusive to architecture, distinct from other professions involved in design projects. https://en.wikipedia.org/wiki/Holism#cite_note-19 Holm, Ivar (2006). Ideas and Beliefs in Architecture: How attitudes, orientations, and underlying assumptions shape the built environment. Oslo School of Architecture and Design. ISBN 82-547-0174-1.
• #11: Another key pattern. What is strategy? How to make sure demand and supply are balanced into some kind of …. Equilibrium. Is that a good definition of a good system? A good design? A good strategy? A good architecture?
• #12: Last but not least – a human perspective … OK – totally reductionist but that’s what we do isn’t it! Left Brain, Right Brain. Using both? I like this one the most as I think it is the essence of architecture.
• #13: So that’s a deep dive in what architecture is. So what about Digital Architecture?
• #14: Let’s look at the science/supply/reductionist side first. What do we have here? So what is the purpose of Digital Architecture? Using digital technology to build value delivering, sustainable … Systems Solutions Services Apps Business Models Enterprises
• #15: So in summary. I’m saying that Digital Architecture is building sustainable, value generating systems using tech. Is there demand for this right now?
• #16: Before we can answer the demand question we first need to dive a littler deeper into what we mean by Digital Architecture. This is an area that can cause confusion. So … the word architecture can be both a A verb – the practice of architecture? A noun – the product of architecture? It’s fairly obvious that architecture as a discipline has value. Filippo Brunelleschi (1377 – April 15, 1446) anyone? https://en.wikipedia.org/wiki/Filippo_Brunelleschi So we can see that it has existed as a profession for a long time. Is that fair enough?
• #17: What about Digital Architecture? The whole area is still pretty immature to be honest. The fact that I’m having to spend time covering it in this talk is proof of that. But that’s what I can kind of get away with coming up with my own definition to add to the others that are out there  But suffice to say we have lots of maturing approaches to digital architecture (and this includes enterprise, IT, infrastructure, system, solution, data etc) – led by John Zachman’s ontology and TOGAF. But what about architecture the noun? What is that and is there demand for digital architects can build???? As a futurist this is a question you’ve got to ask every day. Has a tool ever been created before it’s time???? It’s a question technologists must continue to ask. The Newton anyone? Motorola ROKR, Microsoft Zune, Microsoft Kin, Amazon Fire Phone???
• #18: Digital Architecture – the noun But … in your world … Is there pressure from your customers for more technology? Is their Windows laptop, MS Office, SAP etc. just not enough anymore. How are you doing with that pressure? Are your existing processes working? What did Einstein say about madness again??? It’s time to change.
• #19: OK. I’ve said it before but I need to say it again. There is a catch. We don’t have a full blueprint yet for digital architecture. Some are emerging but they apply best to start-ups What about the rest of us? Can we change? Is it possible? The best way to deal with a lot of unknown unknowns – experiment, sense, respond, adjust, experiment again.
• #20: 5 Experiments in Digital Architecture Keep them small Expect to fail Fail fast If you don’t learn from your failures it is a complete failure (as opposed to a partial failure!) Make sure you have built in a feedback loop Make the most of the dynamic nature of software
• #21: Think Services not Solutions. Waterfall has led us to focus on day 1 only. This was fine when we were building systems designed to last a long time with low change expectations. And that worked as long as everyone took that approach. But the market has changed. Companies are starting to figure out how to change fast, and use technology to meet customer needs cheaper and more quickly. Big Waterfalls are dead. They need to replaced by Agile/Lots of little waterfalls. Build services not solutions. (And this is not just advice for IT systems – it’s what the whole company will need to do – see Amazon …)
• #22: AKA API, Micro-service, Team etc. etc.
• #23: What is a service? One of the best service framework – ITIL – even if seen as an ITOps tool only. BUT IT ISN’T!! It’s a whole big bunch of processes (verbs) for digital architecture.
• #24: “Noun” as a service. Give me a suggestion and see if I can describe it as a service.
• #25: So what about service as the noun? How do you start to model and design things as services? Let me give you some tools.
• #27: https://strategyzer.com/ Here’s tool no. 1 – don’t just use it for businesses. You can use it for anything – a service, a team, yourself.
• #28: https://strategyzer.com/ Here’s an example for an enterprise architecture team/function which reports to the CIO (customer no. 1). It’s quick an easy to do and it will get you thinking in a completely different way.
• #29: Another great model is Tom Grave’s enterprise canvas. It has some added layers that the BMC does not include – in particular the control layer. You can use this as a check list to see if you are doing everything you need to do to deliver a great service (personally, for you team, for your department, for company .. If you are the CEO … )
• #30: They are fractal and recursive. They work everywhere!
• #32: Why do we need maturity models? 1. Is there a thing as too much functionality? Why does this happen? The beauty contest procurement battle? He who has most features at the lowest price wins. When I worked with Microsoft this was one of the biggest challenges. The products were overflowing with functionality but the users struggled to make use of it all. Maturity models can help you avoid this.
• #33: As a quick aside – if we compare Word to Medium we see 2 totally different approaches in action. One has more features that you could ever use. Those other has laser focused on the features it thinks are the most important. The only thing it doesn’t do is expose new features dynamically as it watches how you interact with it … maybe that will come?
• #34: My favourite source of maturity models is the IVI (www.ivi.ie). They have the IT-CMF.
• #35: Diving deeper here is one capability that is dear to my heart. Enterprise Architecture Management. It’s not perfect of course but I have used it in anger and it works. There are 8 building blocks and behind each building block is a set of questions, practices, outcomes and metrics.
• #36: This is the summary.
• #37: Otherwise known as “SHOW ME THE MONEY”!!!!! Value can be hard to measure but cost is something we can and should be able to do.
• #39: I don’t know about you but for much of my career I left the money conversation to the project manager. It made me feel pure! But we i.e. IT guys, are paying for this now. Big time. How do we pay? We get outsourced because the business doesn’t like us. They think we are expensive and slow.
• #40: To be honest I found this world of TBM a little bit by accident. I was trying to find “one model to rule them all”. I had dug though TOGAF and Zachman but most of what I found was either incomplete or more focused on the verb not the noun i.e. the model.
• #41: So what did I find? The Apptio TBM Unified Model.
• #42: The magic is in the apportioning of cost between layers. That is NOT easy as it requires a completely different way for costs to be managed. 99 times out of a 100 you’ll be constrained by an existing financial management structure that ties you to CAPEX oriented yearly investment cycles that are based on business cases that do a really bad job of estimating the cost of a service, resulting in a whole range of operational issues, and often ultimately resulting in major reputational damage for the IT team within an organisation.
• #43: This is a good place to start.
• #44: The book is worth a look.
• #45: I know. From money to emotional intelligence. That’s some switch! But this is doubly important – To practice successful architecture you must be somewhat good at this. You MUST. IASA calls it human dynamics. It’s a core pillar. But in digital architecture, the people in the system must be taken into account. This is deep integration of tech into our lives and we can’t do it without understanding human behaviour.
• #47: OK. It’s much, much, much more complex that this – an infinite spectrum between 2 extremes to be honest – and this is only 1 vector by which you could categorise someone. There are hundreds if not thousands more. BUT, we must have all experienced situations where people are talking the same language but very little data is getting across either way.
• #48: The path to emotional intelligence lies in self-awareness. Only by understanding your own biases will you start to be able to understand other peoples. There are lots of ways to work on this. E.g Emotional Intelligence 2.0, Flourishing – Maureen Gaffney, etc. And there are personality tests – for the reductionists among us 
• #49: There is plenty of help out there.
• #50: My favourite one. It’s free, quick and very positive. And a bit of fun. The key though is self-awareness and not to put your colleagues in a box! I’ve got a lot of people to do the test (as an experiment!) just to see: What they tested to compared to my own guess? Did they agree with the general description? Did their partner/close relatives agree with the general description??  What did I learn? It’s hard to guess There are lots of surprises Some buy it Some don’t
• #51: We’ve done the micro. We also need to start thinking about the macro. Org designs … Hierarchical org designs will not work well in the digital age. Too slow. To constrained.
• #52: New approaches are arriving. For example holocracy.
• #53: Check out Spotify too – tribes, squads, chapters and guilds.
• #54: Org Design
• #55: I’ll close on that thought. Thank you for your time. I hope you found it interesting.