Monthly Archives: June 2016

Enterprise Architecture and language

ian-ellery-head-small

 

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

Mark Heseltine was the next speaker I heard; talking about enterprise architecture for innovation and change, linking EA with Lean and Agile concepts. He made a really interesting observation early on in his talk: the very language of enterprise architecture causes problems. EA is based on an architectural metaphor, but because the people who are listening to us have physical, building architecture as their mental starting point there is a danger that they will (unconsciously) conflate the analogy with reality. A point I will certainly consider when talking to those outside IT.

The other interesting aspect of his talk for me was discussion around technical debt. He showed a slide of a four quadrant model for technical debt and used a financial debt metaphor for dealing with it. There is the interest we pay on technical debt – the effort in managing it – which has to be balanced against paying off the debt, in other words removing or replacing technology.

I also spotted a tweet which quoted another speaker in a parallel session: “don’t aspire to be the best airline in the world at recovering lost luggage”. I think this is something many organisations, and departments within those organisations are guilty of. We need to have appropriate goals. Let’s not be the IT department who are brilliant at fixing things when they break, let’s be the IT department who deliver services which don’t break.

The final two sessions I chose to attend were probably the least useful. The first was from the Swedish national electricity grid, and the second from the UK national air traffic control service. Both had huge problems and were undergoing massive change and transition programs. However, the very scale what they were trying to achieve made it difficult for me to see anything applicable to, or transferable to HE. The only comment I did like was from the Swedish speaker who pointed out that a target architecture is a direction of travel not “The Truth”. Something it’s easy to forget when it’s been through endless committees and gained the rubber stamp of SMT approval.

Overall, a very useful and enjoyable two days. This is not a cheap conference, so I am very grateful for UCISA funding my place. Would I recommend other HE architects to attend? Yes I would. Getting out of our own sector is always useful and the commercial challenges of disagreeing business units, limited funding, and pressure to deliver are no different to HE.

UCISA has an Enterprise Architecture community of practice which may be of interest.

Enterprise Architecture and the Double Diamond model

ian-ellery-head-small

 

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

The next talk I went to was entitled “from 0 to EA” and was about trying to use enterprise architecture in a small business undergoing radical transformation after being bought by a private equity company. They used a “double diamond” model, with ideas diverging and then converging on a solution. Another insight was thinking ahead and asking where did they want to be in the IT change cycle when “X” happens? We should specifically design timelines to meet external events. A couple of things resonated with me: firstly, true EA success is embedded and invisible, but changes everything. Secondly, be careful not to solve problems for people and tell them that EA is always right. It’s difficult to gain a seat at the table, but easy to lose it again.

I struggled to get much from the next talk from Andrew Swindell. He was describing a concept called “line of sight”, but with very busy sides, I lost sight of his concept. He showed lots and lots of alternative models, and my overwhelming impression was it was all simply too much. One to consider on rereading slides later perhaps.

UCISA has an Enterprise Architecture community of practice which may be of interest.

Enterprise Architecture – using a chess analogy

ian-ellery-head-small

 

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

This morning’s opening keynote is one I’ve been looking forward to – Gerben Wierda, author of “Chess and the Art of Enterprise Architecture”. The book was recommended to me, but I never got round to buying it. He is also creator of the well-known YouTube animation of “hairball architecture”.

His view is that EA has been generally unsuccessful. We do EA to try and make sense of complexity and prevent chaos. However, while this works in theory, chaos still remains. In many organisations there is significant churn of EA functions, and the business remain unconvinced. Why? In his view because the decisions, the action, takes place in projects and not at the higher EA level.

To make EA work, we should use a chess analogy. In chess each move is to make things better, to improve the current state, but with no specific end state in mind. So, he reasons, EA future state plans are a waste of time. The rules of chess are descriptive not prescriptive, and similarly he argues that EA principles can be toxic as they prevent any discussion leading to a “comply or explain” culture.

He also talks about ensuring that EA does not leave out relevant detail. This led to, for me, a lightbulb moment: oversimplification is as bad as over complexity. If as architects we produce a very simple picture for our senior teams – usually to try and help them understand – they will think that the problem is simple. They will then wonder what all the fuss is about trying to solve this apparently simple problem. There were several other phrases and points which made me consider just what I am trying to do with architecture. Overall, this talk was a really good challenge to the status quo of enterprise architecture. And yes, I will now be buying the book.

UCISA has an Enterprise Architecture community of practice which may be of interest.

EA and project portfolio management

ian-ellery-head-small

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

 

 

My final technical session was from an enterprise architect and portfolio manager at Danfoss, a Danish engineering company. Although very focused on improving the bottom line and digital innovation for a manufacturing company, parts of this were very relevant. Before architecture and portfolio got together, projects were initiated and governed by individual business areas (faculties?) with no overall coherence. While they have now moved to a single portfolio model, they still allow business areas to think they have their own portfolio, even if they don’t really own it. They also spoke about technical IT people never wanting to engage with business staff, expecting business partners to act as the interface and go-between. This certainly sounds familiar at Christ Church. Their final insight for me was that although they had a big vision for how architecture and portfolio was going to come together, they cautioned against trying to explain this to business colleagues all at once. Eyes would simply glaze over, so they realised it was easier to explain the changes piece by piece.

The last session of the day was from Neil Mullarkey – formerly of “Whose Line Is It Anyway?” and now using the techniques of improvisation to teach businesses how to collaborate. He drew out a lot of parallels between agile development and improv, as well as getting us all to improvise a story with those sitting next to us. An excellent speaker, with some strong messages, who I would strongly suggest UCISA consider as a guest speaker at the annual conference.

UCISA has an Enterprise Architecture community of practice which may be of interest.

EA and bimodal IT

ian-ellery-head-small

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

 

The next IRM EA conference I session went to was described as looking at the linkages between EA and portfolio management. However, little was said about the PMO, with the focus really being on setting up an EA function. One telling insight is that most EA tools are for enabling architects to talk to architects, not architects to talk to humans. The speaker’s company eventually settled on a tool with both PMO and architecture visualisation capability, which also links to technology life-cycle. This latter allows them to link projects to technology obsolescence. The speaker ended by commenting negatively on the Gartner model of bimodal IT. He felt that the faster mode does not understand BAU, while the slow side is seen as boring. This linked nicely to the next session.

This session was entitled “avoiding the bimodal disaster”, so we were in no doubt about the speaker’s viewpoint. He was president of a digital transformation consultancy, and strongly believed that the digital agenda is impossible in a bimodal organisation. In his view either the slow side will hold back the faster transformation efforts, or shadow IT will be created. There were some good aspects to this, such as not seeing EA as city planners, as planned cities are not agile. They may be efficient, but they are not able to cope with rapid change. However, throughout the talk I was left feeling that while there may be issues with Gartner’s bimodal model, there were at least as many issues with the “self-organised” approach advocated by the speaker. Perhaps it’s working in HE, but I was left thinking that there has got to be a compromise position somewhere. One observation I did agree with however, was the need to be ready for citizen IT: whether it be citizen integrator, developer or data analyst. We need to accept that in the future, these areas are no longer wholly owned by IT.

UCISA has an Enterprise Architecture community of practice which may be of interest.

EA – making a first assessment

ian-ellery-head-small

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

 

The next session  I went to at the IRM enterprise architecture conference was from the Head of Architecture and a solution architect from Cambridge Assessment: the examinations and marking offshoot of Cambridge University. To start architecture, you should always look at what is failing. When the new head of architecture joined, he sat in IT operations and observed what was happening. He realised that things were breaking constantly, but being fixed instantly before user impact by a very dedicated team. Poor architecture was being hidden by passionate staff.

There is a balance to be made between architecture for operations and architecture for the users. In the former it’s about reliability, the latter for example about moving away from pure infrastructure (“how many data outlets do we need?”) to asking questions about the users experience in a new building. They finished by describing architecture as a set of services with outcomes aligned to decisions. And compared all EA Frameworks to the labours of Sisyphus – forever pushing a boulder up a hill for no benefit at all.

UCISA has an Enterprise Architecture community of practice which may be of interest.

EA for managing change

ian-ellery-head-small

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

 

 

And so, here I am at the IRM enterprise architecture conference  . The opening introduction promised a mixture of talks about the theoretic as well from those who have actually done it. We were also told that there were representatives from over 30 countries and all continents (except Antarctica!). A glance at the delegate list showed that I am the only person from a UK university, with just three other university representatives. This is reflected in the talks, with lots of emphasis on product delivery, profit margins and managing a business across multiple international locations. However, there was very little that I didn’t feel could be translated into the UK higher education sector in some way.

The opening keynote was from Ashley Braganza, a professor at Brunel Business School. He spoke passionately and length about the fact that, when it comes down to it, everything is about change. Business process management leads to enterprise architecture which leads to project and portfolio management: but all of these are really about managing change.

Theoretically, everything that is being done should link back to organisational strategy, but in practice it rarely does. He used an excellent analogy of the strategy being a mirror. When SMT look in the mirror, they see their strategy reflected back at them. But the mirror is then broken apart and each SMT member takes away a piece that reflects their part of the strategy. The mirror is then broken down again and again until every individual in the organisation has their own piece of the mirror reflecting their objectives. But the big picture, the reflection of SMT’s real vision, has been lost. And unfortunately, it is the enterprise architects who are usually called on to fix things! My reflection on this (pun definitely intended) is that either we fight this and try and get a coherent strategic vision, or alternatively, perhaps we embrace it and welcome the fact that EA is the place where corporate strategy becomes visible.

To finish, Ashley reflected on different change models, which he felt were all lacking. He was especially vitriolic about Kotter’s eight steps to change . Finally, one of the phrases of the day which totally resonated with me: the problems, in his view, are that we have 21st century models and methodologies, working within 20th century organisations led by people with 19th century mindsets. By the latter he meant a Dickensian obsession with counting things. Budgets, REF, TEF… sound familiar?

UCISA has an Enterprise Architecture community of practice which may be of interest.

Information management – AIIM’s state of the industry report

Sara Somerville

 

 

Sara Somerville
Information Solutions Manager
University of Glasgow

 

A Look Behind the Curtain: The State of the Industry – Bob Larrivee, AIIM’s Vice President, takes a look at the AIIM State of the Industry Report (in the U.S):

  • The main drivers to ECM are risk and compliance and costs and productivity
  • Dependencies – 47% of respondents said if they had an outage of their system for more than two hours it would cause severe disruption
  • Governance – 18% align IM/ECM system strategies with agreed Information Governance (IG) policies, 29% have no IG policies. Without governance you are just creating a digital dumpster!
  • Content creation mechanisms – 49% use a shared copy review but only 10% of organisations use concurrent editing; 78% are still circulating documents by email for review and approval
  • Email – 39% describe their email as chaotic; 16% keep everything; 17% have a dedicated archive with a defined retention and hold policy
  • Content deletion – 47% have an Information Governance policy, but 51% rely on manual deletion
  • Defensible content deletion – Half would struggle with cloud, social and file shares; 20% are following policy but are not auditing it
  • Business Processes – 46% have some paper-free processes (so 54% don’t); 30% are capturing content closer to the customer
  • Multi-channel inbound – 23% have elements of multi-channel integration, but only 5% with automated routing to multiple processes
  • Enterprise integration – 31% are integrated with content creation systems, but only 18% with multi-repository search
  • Content analytics – 15% using auto or assisted classification at creation/declaration
  • Cloud – 20% are live with cloud for all or some of their core content; 7% with selected users or content, or for collaboration and file-sharing.

Compliance in multiple repositories

Sara Somerville

 

Sara Somerville
Information Solutions Manager
University of Glasgow

 

Feedback on a content management session at the AIIM conference

Content Whack-a-Mole: Keeping Up Compliance across Multiple Repositoriespresentation by Michyle LaPedis and Jordan Jones from Cisco Systems

There are multiple tools popping up that enable users to create, share and manage documents, and these were challenging the traditional repository set-up such that users didn’t know what tools to use when. The team realised they needed an overarching strategy to address this issue.

One of the other problems Cisco had was around the search tools which were returning a lot of ROT (Redundant, Obsolete and Trivial data), and if the users couldn’t find what they needed then they tended to create the document again.

Content Management IT at Cisco focuses on an open architecture with open source and open standards. They have three major on premise systems and are currently implementing a project to migrate documents in to one system/repository as a focus for records management, and to enable the application of lifecycle management to that content. (Cisco also use box as their organisational cloud based document management and collaboration tool.)

Some of the steps the team took to improve the situation included:

  • Rationalising their on and off premise services and adding a compliance layer
  • Creating a content management program management office to ensure that proposals for any new IT tools came through this office for approval
  • They defined an ILM (Information Lifecycle Management) strategy for their documents and identified their repositories
  • Cloud offerings often mean less control and make it harder to manage and delete content, but the team created a dashboard for users to move documents from the cloud (or any other location) in to the repository they had created
  • They sent out monthly emails with information about what records need to be deleted and then reminder emails were sent every month after that to remind users to take action (if the users don’t take action after six months then the data is deleted).

Some of the issues they have encountered:

  • New platforms do mean new issues
  • Changing personnel means sometimes starting over – but hopefully there is some hand-over
  • There is a code of conduct which states that it is an employee’s responsibility to manage their information responsibly
  • They have started to phase out the network file shares by making them read-only and then they will start to move the documents over to the other approved repositories.

Key takeaways:

How do you win the game (Whack-a-Mole)?

  • Remember the game will never end
  • It’s important to have a strategy and for records and information professionals to work with IT to implement it
  • There is always going to be a new technology, so it’s important to get the process and the people part working together.

 

 

Reflections before the Enterprise Architecture Conference Europe 2016

ian-ellery-head-small

 

Ian Ellery
Head of IT Architecture
Canterbury Christ Church University

 

 

 

 

 

 

I am the lucky recipient of a UCISA bursary  to attend the Enterprise Architecture Conference Europe 2016 run by IRM UK. In the week before the conference, I reflect here on my reasons for wanting to attend, what I am hoping to get out of it and how I hope to use the information.

I have worked at Canterbury Christ Church University for nearly 10 years and have recently become Head of IT Architecture. However, I am not an enterprise architect by background or training I am a senior IT manager and strategist. The potential power of enterprise or IT architecture was introduced to the University by the new IT director and my challenge is to deliver for her a light-touch architectural framework. This will allow the IT department to create a blueprint for the future, and allow the University to manage the significant levels of application change currently planned.

But unlike many universities, we cannot afford to recruit a team of architects. It’s just me, a software/ solutions architect, three business analysts and some project managers. And all too often, architecture feels theoretic and more concerned with tools and frameworks than pragmatic delivery. I was looking for an opportunity to learn how enterprise architecture and IT architecture delivers in real life. A few months ago we worked with Sally Bean, now a consultant and formally an enterprise architect, who consults on building and maintaining high performance architecture teams. She regularly speaks at this conference as does Chris Potts (author of “FruITion” ), and John Zachman, creator of the Zachman Framework  .

By attending I hope to gain insights into the practical application of enterprise architecture, into an organisation which doesn’t realise it needs it. I’m also interested in how hard-nosed commercial organisations use EA to improve their business.  So I am looking forward to talks with titles such as “What Can You Achieve with a New Architecture Team?”   and “Outcome Driven Architecture” . The conference is also co-located with a business process management conference, with attendees of both conferences allowed to drop into sessions from the other. This is also interesting, as in common with many universities, Christ Church has processes which have built up and grown over the years and are no longer necessarily suitable for a 21st century digital university.

As well as writing a daily blog for the UCISA website, I will be posting the occasional Tweet as @e11ery, conference hashtag #EACBPM. I will also attempt to collate my thoughts and experiences, both from the past six to twelve months as well as the conference into a beginner’s guide to creating and using enterprise architecture in higher education environment: without actually employing any trained architects.