Light Bulb Interview Series - Jeff Kennedy - Part 1

Welcome to the first interview in the 2009 Light Bulb Effect interview series. In this series I will be interviewing a number of influential Enterprise Architect's focused around 3 key areas:

1. Major Trends in EA
   
2. Working in the New Business Climate
3. Hot Topics for EA

The first interview is with Jeff Kennedy - from the University of Auckland.
Jeff has enjoyed a long involvement with information systems, from a background specialising in application integration and identity management. He is currently serving as Enterprise Architect within the Information Technology Services division at The University of Auckland. Jeff is convening this year's CAUDIT Enterprise Architecture in Higher Education Symposium, to be held at The University of Auckland in November 2009.

The interview will be posted over 3 parts and I thank Jeff for agreeing to take part and insightful responses.

Light Bulb Effect Interview
Part 1 - Trends in EA and IT

MC: What do you see as the role of Enterprise Architecture as related to Business Strategy?
JK: Enterprise Architecture has become mission-critical in its role as the transmission glue between business strategy and technology strategy. Although reports of the Enterprise Architecture function migrating from information-technology-owned homes to CIO-based or central-business-unit-owned homes are both overemphasised and irrelevant, the role Enterprise Architecture must fulfill today is much more business-oriented than ever before. Key drivers for this new business outreach include Service-Oriented Architectures, Business Process Management, the global recession, and the ongoing demise of Enterprise Architecture frameworks.

MC: What is the role of Architecture Governance?
JK: With apologies to Paolo Malinverno, it has become evident that governance craves governance, and the situation today sees many enterprise struggling to understand what behaviours and structures constitute governance, struggling with administratively overburdened governance structures, or relying upon technology-based point solutions to the overall governance requirement. Trendwise, governance ought best see out 2009 as a year of deliberate consolidation in which enterprises understand the nature of the governance they have in place today, the advantages and shortcomings of that governance, and what should happen next to foster real maturity in enterprise governance initiatives. There is a big difference between design-time governance (e.g., for business processes, web services, and for all other artefacts) and the broader-scoped governance that determines the shape of an enterprise's infrastructure, applications, projects, and business-services portfolio, though the two domains must be linked strongly.

MC: Trends in Enterprise Systems?
JK: Almost without exception, vendors of enterprise systems are adopting open standards and moving towards some form of Service-Oriented Architecture foundations for their products. The drivers for this vendor adoption include internal requirements to consolidate increasingly broad ranges of acquired product solutions and external requirements to integrate with business applications from homegrown or other-third-party provenance. We have yet to see the rise of micro-service architecture (in which vendors could license individual business services or pieces of end-user functionality rather than semi-monolithic enterprise applications), but the increasing adoption of BPM-related, Portal-related, and WSDL-(and-REST)-related standards will continue apace during 2009.

MC: Trends in Infrastructure?
JK: It remains unfashionable to predict the death of the mainframe on account of the rise of Intel-based virtualisation, grid computing, and cloud computing, but it is apparent that big-iron infrastructure offers a much less sustainable (and therefore a much less attractive) value proposition than it did until even quite recently. The price-performance balance on commodity-based infrastructure makes even the best mainframe-based virtualisation increasingly difficult to justify. 2009 is likely to see the first stages take hold of local cloud infrastructure replacing traditional virtualisation.

MC: How about trends in Software/Hardware?
JK: The market share of proprietary unix systems will continue to slide as the various flavours of open-source linux gain still-wider popular adoption. Sun Microsystems has had tremendous investment and inertia in SPARC, in Solaris, and in everything related to Java, and the future of Sun is bound up tightly in all of this. Perhaps the most significant force in this concerns the future of Java, particularly Java EE, as competing light(er)weight frameworks stake their claims against the long-term evolution of the Java API. Many vendors will migrate to bare-metal delivery packages for provisioning their software to customers.

MC: Networking and Telecommunications?
JK: IPv6 continues to be deployed quite quickly (because it's the way of the future and all the replacement-round operating systems and router kit supports it), VoIP continues to be deployed relatively slowly (because it's expensive to deploy), and Google's Android will displace the iPhone completely (maybe BlackBerry too).


End of Part 1.

Jeff, Thank you for your great insight. Keep an eye out over the coming days for Part 2: Working in the new business climate soon...

2008 Renaissance Dinner

Last night I attended the annual Renaissance Dinner at the Graham Hotel in Port Melbourne. This is a tradition started several years ago by Peter Stubbings aimed at getting a number of like minded IT professionals together to share a meal talk about their experiences and generally have a good time. The people in attendance represented a pretty broad cross section of the industry including IBM, Oracle, SAP, Ingres, Oakton, Unico, Business Objects to name a few.

What was encouraging was the relative upbeat mood regarding the economy, although this did get a little heated when economics and politics started to cross. The general feeling is that we have been trading off a boom cycle for quite a while, and given that we live in one of the only developed economies with a budget surplus and sound economy going into this Global Financial Crisis there is a good chance we can buffer the storm.

Time will tell...

Cloud/Grid/Utility Computing

I was reading Micheal Specht's blog Cloud/Grid/Utility Computing what is it & must you have it? and It got me thinking, it is true that the various buzzwords that the industry offers up can sometimes be confusing, however, maybe it would help if we boil it down to bare essentials.

When Micheal and I caught up recently we had a long discussion that centered around these decisions being horses for courses - what is the best architecture for your application? Try to see through the buzzwords - whether it be cloud, utility, or SaaS or whatever. What is the underlying architecture and most importantly can it deliver for your respective application on the following criteria to the degree that you need:

1. Performance
2. Business Continuity (i.e., back-up and recovery)
3. Security, privacy and compliance
4. Service Level Agreements
5. Business Integration and processes
6. Affordability

That is what is important. What the solution is ultimately called is not important as long as you are getting the computing resources you need, when you need them at a price point that suits.

10 Strategies for Doing Business in a downturn

I was having a conversation with Michael Counsel (Senior Director, APAC Solution Engineering) regarding what advice he would give organisations in the current economic down turn. The view was first and foremost that organisations need to “stay in business”, but what he highlighted was 10 key areas that every organisation should evaluate in this difficult and uncertain economic climate:

1. Acquire Market Share - The number one strategy is making sure you continue to not only maintain and grow market share in your revenue generating business areas or revenue centres, but look for opportunities to acquire market share in new markets during the downturn. Get more customers and in wider markets provided you can sustain supply costs.
2. Look for Acquisition Opportunities - The downturn will drive some companies to the wall. Many become bargain priced for acquisition. If you have the means, this can be an outstanding strategy which can turn medium to large players into corporate giants when the market rebounds.
3. Opportunity to Fix Business Process - Loaded processes are hard to change in confident times and less efficient pieces are hidden in the profit margin and good revenue you enjoyed. This is an ideal opportunity to take what you knew needing fixing in a business process…and fix it. You will not only profit from the operational savings in the downturn, but the increased margin when the market rebounds.
4. Automate the Commodity Operations - Convert commodity operations which require human labour into automated business processes and services. You will need to be careful not to inadvertently wrapper a differentiating or competitive advantage process inside a commodity automation initiative.
5. Offload High Cost/Risk Customers – Analyse and understand your customer base and work out which customers are costing you money. Off load these to your competition
6. Double Investment in Innovation - The worst mistake you can make is cutting innovation in your business. Work out what differentiated you before the downturn and will still be in effect when the market picks up. Harness creative talent in your organisation to adjust and adapt quickly at the next upturn. Organisations that increase their competitive advantage and differentiation have been proven to get earlier and more rapid revenue growth in the next upswing.
7. Watch Your Suppliers - Remember in downturn, your key suppliers may go out of business and Supply chain response time can be deadly on the recovery if you are looking to rapidly get product to market. This time may present an opportunity to vertically integrate and make sure you can meet production levels on the way out of recession. Which in turn build that platform for greater profits
8. Re-platform your IT Operations - Use the opportunity to re-platform IT infrastructure to lower cost platforms and commodity computing. Take an inventory of existing applications and decommission, consolidate and virtualise where possible. Take the opportunity to rebuild and modernise key pieces of infrastructure which creates long term resilience and connectivity, while also improving functionality. Organisations should be vigilant on maintaining software and hardware currency by upgrading disks, servers and software on regular timeframes - not doing this can have disastrous medium to long term impact as outlook improves.
9. Adopt a Shared Services Strategy - Find ways to lower the cost of operations for commodity or corporate services that help you run your business, eg ERP/Back office processing. These are generally not the core business/revenue generating systems. You should attempt to share these operations with other business units or functions in your organisation.
10. Insource Your Key Systems - Outsourcing and Managed Services are good when you want to trade risk for premium pricing and is especially valuable in confident markets in equilibrium. Invest in the future of the business by insourcing the key (revenue) system operations so you can inject knowledge into the future talent pool and have creativity inhouse. Having the revenue systems under your control means rapid response without waiting for a third party.

Whilst some of these seem obvious, the key to success is the leadership and strength to pass the microscope over the organisation and take the opportunity to evaluate the current situation, and develop a plan that provides short-term operations cost savings whist laying the foundation and roadmap to prosperity when the market rebounds

Mike Counsel can be contacted michael.counsel@oracle.com

Actionable Architecture

I read following blog Architecture as diagrams is an anti-pattern where the it is positioned that the creation of diagrams an only a visualisation or representation of architecture.

I could not agree more that the key for successful architecture as to assist in the implementation of strategy. As mentioned in the blog,

It is the misguided belief that the creation of diagrams is architecture. They are a representation of architecture. For architecture to be useful it must be actionable. A diagram is not actionable, it is a visualization. The specification relating to the blob on the diagram is actionable - you can build it, deploy it, assess it, review it. A colored blob on a diagram is next to useless.

I have seen too many examples of organisations that focus on creating artifacts relating to current and future states, but under invest in the development of the roadmaps to make the transition. This step of making the architecture actionable is the difference between using visio and driving buisness change and transformation.

Comments on Change Management

In Micheal Specht's blog post Change Management and Technology Implementation he discussed how change management has been one of the most underrated and overlooked components during technology implementation

In my experiences, I have seen organisations time and time again look for short term $$$ savings by cutting project costs through reduced investment in project and change management.

This sort of skimping on leads to the following situations:
1. Poor scoping of the project
2. Mis-alignment of expectation for key users and stakeholders

The results are generally the same… Poor technology adoption, heavy re-work and sometimes re-implementation.

It is important that Change Management and in particular how the system/technology will impact users must be considered way before the first product is installed. Technologists left to their own devices will continue to implement solutions and products that meet project deadlines on the surface but not achieve the business objective, users expectations and more critically provide a positive ROI.

Without supplementing technology adoption projects early with disciplines around project and change management Technology (particularly large IT lead) projects are doomed to fail.

eDiscovery and Information Lifcycle Management

This week I attended a customer event around the impact of eDiscovery for the legal community. The key note presenter was Dr Bradley Schatz who is the Director of Forensic Technology from Vincents

eDiscovery extends the current incredibly time consuming and expensive paper based discovery process for litigation used by the legal fraternity. eDiscovery extends this to potentially include all corporate electronic records including information stored on:

• Corporate Transactional Systems (Finance, HRMS etc.),
  
• Messaging (email, instant messaging, blogs, wiki's, voice etc.),
  
• Native Files such as (word, ppt, PDF, video and audio) stored on various resources such as file servers, laptops, handheld devices (mobile phones and iPods), USBs, external stroage.

The process of eDiscovery described by Dr Schatz incluses 2 new steps Preserve and Process from the traditional approach to be:

1. Identify - strategies to locate relevant infromation (online, backup storage, etc)
2. Preserve - the integrity of the information (ie if on a cycled backup tape copy before it is destroyed)
3. Collect - de-duplication and near duplication of documents into a central location for analysis
4. Process - into searchable images or native electronic copies
5. Review - the content for relevance and priviledge
6. Produce - the material for the other side

Given the relative cheap cost of storage and the broad sources of "relevant" information, some organistaions have had a tendancy to hold on to as much data as possible, potentially resulting in incredible high cost in the event of litigation. The establishment of a well though out Information Lifecyle Management strategy and policy that defines security, catorgorisation, destruction of information and eDiscovery use cases is critically important. This can not only save money in the event of a Freedom of Infomation (FoI) request or litigation, but could also save significant storage and Data Centre costs and operational costs.