What defines success with public sector enterprise architecture

To ensure success, enterprise architecture programs in the public sector must be carried out in a somewhat different manner than in the private sector. This article looks at a series of successful enterprise architecture programs in the public sector from around the world and identifies the common characteristics that seem to have contributed to their outcomes. The findings point to the importance of political leadership, federated governance, and shared enterprise frameworks.

Jan K. Gravesen (jkgraves@us.ibm.com), Executive Industry Architect, IBM

Jan is IBM's Executive Industry Architect and Client Technical Adviser for the California Account, which serves the public, healthcare, and educational sectors in California. Previously, he was the Enterprise Architecture and Technology service area leader for Northern Europe, served on IBM's Worldwide Enterprise Architecture Council, and was the Application Innovation Services leader for IBM in the Nordics. Before joining IBM, Jan worked in a series of positions as adviser and enterprise architect to private and public sector companies in the Nordic countries.



11 December 2012

Introduction

The European debt crisis and the US "fiscal cliff" debate have intensified the interest in having a more productive public sector and spurred more focus on finding service and productivity improvements in social services, health care, education, public safety and defense. Similarly to the private sector, much productivity and service improvement can come from delivering a more customer- or, in this case, citizen-centric service: Having government self-service processes initially defined by the needs of the individual citizen rather than the needs of a functionally structured and complex set of government organizations and programs.

Because citizen-centricity and online government programs often require better coordination of programs and services, they tend to be cross-departmental and sometimes must function on a national or state level. Their central challenge becomes one not of technology but of coordinating a group of autonomous or semiautonomous states, agencies, and departments to pursue solutions and services that make life easier for citizens and businesses. This is why enterprise architecture as a managerial design discipline has seen considerable interest across most of the Western world in recent years:

  • Within health and human services organizations, enterprise architecture has created more citizen-centric programs that integrate the "citizen experience" when interacting with the public sector. Large-scale examples include:
    • Germany's e-Government 2.0 program, which coordinates online government initiatives among the 16 partly sovereign constituent states in the Federal Republic of Germany
    • Service Canada's monumental citizen service improvement program
    • Denmark's MEDCOM and online health initiatives, which serve as a unified entry point to the health care sector and store electronic patient records for all Danish citizens
    • California's FI$Cal program to create a shared platform for procurement, accounting and public budgeting processes across 120 state departments
    • Norway's implementation of a common ERP system for its health care sector
  • On a larger level, joined-up government is a key objective for many governments. The term "joined-up government" was used in the United Kingdom as part its first government strategy. This strategy's goal was to "join up" or integrate online government services so they appeared in a citizen-centric manner rather than in a departmental manner that would seem fragmented and confusing to individual citizens. The term is now accepted in many parts of the world to describe the integration of the government services, government processes, and systems that are necessary to deliver a citizen-centered government, and represents the objectives in many contemporary online government initiatives, such as Services Canada and the Dutch National Implementation Program.
     
  • Within military organizations, the need for network-centric warfare (NCW) and network-enabled defense (NED) models led to the creation of comprehensive enterprise architecture frameworks, such as the US Department of Defense Architecture Framework (DoDAF), the UK Ministry of Defense Architecture Framework (MODAF) and the NATO Architecture Framework (NAF). NAF, for instance, describes canonical data formats for the exchange of operational data among cooperating military forces from different NATO countries participating in integrated missions.

In these examples, enterprise architecture is a discipline to design and manage programs that share resources better and facilitate an efficient and coordinated exchange of information among departments.

In general, public sector organizations differ from private sector organizations in fundamental ways that lend themselves to using enterprise architecture as a managerial discipline in accomplishing many of the objectives listed previously:

  • First, public sector departments tend to be governed in a more autonomous and political manner. They apply a "power behavioral" approach or political approach to strategy, planning, and problem-solving processes and to the development of new programs.
     
  • Second, budgets for improvements, information technology, and so on are rarely centralized, because that is seen as leading to inefficient and unnecessary bureaucracy.
     
  • Third, in most big states, departments operate nearly independently of each other and with less coordination and sharing of resources than is often seen in the private sector. In Germany, the 16 partly sovereign states (or as they are called in German, Länder) that make up the republic of Germany assume independence from each other and partly from the federal government. In a large cluster of cooperating countries such as the European Union or the Eurozone, there is a constant struggle between the need for further political and economic integration and the autonomy that individual states have enjoyed for centuries.

This constant struggle between the need for political and economic integration, including the ensuing resource sharing and the continued autonomy of agencies, departments, states, countries, or arms of military organizations requires a specific kind of enterprise architecture.

In a previous article "Reasons for resistance to enterprise architecture and ways to overcome it" (see the link in Resources), I made the case that to be effective, enterprise architecture must align with the governance, architecture, organizational design and strategy processes used by the organization. It must reflect what is important to the organization and respect its political decision-making process.

In the case of the public and political sector, enterprise architecture must respect that:

  • Political return trumps economic return. Business cases should really be expressed in terms of policy cases.
  • Governance is federated and rarely centrally mandated, except in the form of policies or laws.
  • Patience is a virtue. The public sector can afford to and should think long-term. Programs must be durable across multiple election cycles, ideally.

My study of several successful large-scale online government or consolidation programs from around the world points to three common characteristics in successful public sector enterprise architecture programs:

  • Political leadership
  • Shared governance
  • Federated platforms

Note:
The programs that I refer to in the article are all described in the public domain, and the organizations involved are not necessarily IBM clients or users of IBM technology or advisory services.


Political leadership

First and foremost, successful programs are anchored in the political system, not in their IT organizations. They receive solid political backing from the highest political leadership, such as the German Chancellor, the State Ministry in Denmark, the Cabinet Office in the UK, or the California State Governor in the US. The political leaders are perceived as patient and have a long-term vision that will be implemented by the gradual (incremental) adoption of agencies and departments, often over many years, with the entire project completed long after the politician leaves office. In several cases, legislation is passed by cabinets or parliaments in support of programs and even enterprise frameworks, so they often become legislated or mandated.

For example, the MEDCOM national health information exchange (HIE) was included in the 1999 national budget agreement between the State of Denmark and its healthcare regions. Today, it is the unified entry point connecting the entire Danish health care sector: hospitals, private clinics, general practitioners (GPs), pharmacies, dentists, physiotherapists, chiropractors, psychologists, private laboratories, municipalities, and the public health insurance plan. All stakeholders in the health care system use the health data network to exchange information, using established standards. Examples include GPs sending electronic prescriptions to pharmacies, laboratories sending test results to GPs, and the exchange of medical imaging and video conferencing between health care providers and patients. In 2005, the Dutch cabinet adopted NORA (the Dutch Government Reference Architecture) as a national standard and framework reference point for domain-specific enterprise architectures to define and deliver joined-up government services to its citizens.

Rio de Janeiro mayor Eduardo Paes invested political capital in creating an integrated operations center to better enable his city to respond to weather-related disruptions. Today, the city has a state-of-the-art intelligent operations center, Centro de Operacoes Rio, which monitors dozens of data feeds concerning weather, traffic, police, and medical services on a real-time basis to anticipate problems and help city managers mitigate their impact (see Resources). Operations officials from across the city collaborate daily to manage traffic and public transportation systems, as well as the efficiency of power and water supplies. The Center relies on a system pioneered by IBM Research scientists: a high-resolution weather forecasting and hydrological modeling system for Rio de Janeiro, which can predict heavy rains up to 48 hours in advance. The forecasting system is based on a unified mathematical model that pulls data from the river basin, topographic surveys, the municipality's historical rainfall logs, and radar feeds. The system predicts rain and possible flash floods and evaluates the effects of weather incidents on urban emergencies, such as street closures or power outages.

At the center of most successful cases is a long-term strategy that starts a political movement and is ready to accept incremental results over a long time span, usually several years. Political will seems to be a key ingredient in successful outcomes. Service Canada, for instance, is part of a government-wide service transformation initiative "to improve the delivery of government programs and services to Canadians by making access to them faster, easier, and more convenient." (see Resources). The initiative improves the design and delivery of government programs and services to its citizens. Service Canada began operation in 2005 with a mandate to provide Canadians with a single point of access to a wide range of government services and benefits in person, by phone, by Internet, or by mail. However, Service Canada's origins date even further back, to 1998 when the Government of Canada developed an integrated citizen-centered service strategy based on surveys of citizens' needs and expectations. In 2007, Service Canada had partnered with 14 other departments and agencies to provide access to more than 50 government programs and services. It had established nearly 500 points of service across Canada, designed to deliver programs and services into the most rural and remote areas. The numbers continue to grow, because the goal is to provide Canadians with a single point of access to all government programs and services, regardless of where they live or how they want to interact with the government.

Another case from California illustrates political leadership as a factor. Through the formation of the California Technology Agency (CTA) and an executive order signed into effect by former Governor Schwarzenegger, the state is pursuing a consolidation of its data centers into a single federated data center known as the Office of Technology Services, or OTech. Currently, it has almost 200 customers from different departments and agencies within California. In recent years, OTech rolled out new services such as a federated data center (FDC) where customers can manage their own data, applications and infrastructure within OTech's state-of-the-art data center facilities. Although there are agencies and departments that are not using OTech's services and that are allowed to continue to operate their own data centers, the case exemplifies that the political will to create a shared platform for infrastructure services can break down traditional departmental barriers and do-it-yourself cultures that exist in IT management.


Shared platforms

Cross-organizational enterprise architecture programs affect many autonomous or semi-autonomous departments and agencies. They have a well-defined political objective, typically to establish a shared, open digital platform for improved resource sharing or for coordinating better citizen service across different channels, departments, or levels of government.

These platforms must be built in an open way to work across multiple government agencies. In some cases, they need to be able to integrate and coordinate information from many parts of government into a unified solution. They must be incrementally extensible to allow for more departments, agencies, states, or even countries to gradually come on board.

In the case of Rio de Janeiro's integrated operations center (IOC), it is the IOC platform that coordinates the activities of more than 30 municipal and state departments plus private utility and transportation companies. In this way, the operations center — essentially a shared platform with integration, coordination, and visualization features — has become one of the first facilities in the world that tries to integrate a majority of the functions of a city in a single digital command-and-control system.

The UK Government's GOV.UK is both an example of political leadership and a shared platform that serves as the unifying delivery point for a rapidly growing proportion of the UK's online government initiatives. Mandated by the UK's Cabinet Office, it is also an example of political leadership. It integrates a variety of formerly scattered websites and online solutions from different departments into a single, integrated entry into the UK's government sector for citizens and businesses.

Architectural guidance plays an important role in securing successful outcomes. This guidance is usually provided through formal enterprise architectural frameworks, such as the Government of Canada Strategic Reference Model (GSRM) or Whole-of-Government model, the Dutch government's NORA model which is at the center of the National Implementation Program (referred to as NUP in Holland), or Denmark's FORM model (the Common Public Reference Model), which shows all activities performed at the municipal, regional, and state levels by using a component based model (see the related article cited in Resources). In some instances, the guidance is through shared ontologies, or shared descriptions of outcomes, data, and activities. The shared framework makes it possible for departments, agencies, municipalities, and private companies to cooperate and exchange information by using a common information model, shared objectives, and a shared "language" for systems and data. Even with a shared frame of reference at an overall level (whether this is a nation, state, or county, or at a certain functional area), it is equally important that it can be localized to individual municipalities, agencies, and departments. In Holland's National Implementation Program (NUP), the NORA model serves as the shared frame of reference but is localized by the state sector (MARIJ), municipalities and even the water authorities.

Another example of a successful shared platform is NemID (Danish for Easy ID) in Denmark. NemID represents a cross-organizational solution to two common problems in online government: How to authenticate citizens that access public sector services over the Internet and how to sign service requests and official documents, such as applications and tax forms. NemID operates on a national level. It is interesting because it applies a highly scalable, replicable public-private partnership solution that includes more than 200 private banks and insurance companies as well as all municipalities, regional and state-level online government services. Citizens identify and authenticate themselves to their banks and insurance companies, sign their tax forms, and look up information about their electronic patient records by using a single online authentication method. NemID is run by the private company Nets, which is co-owned by the Danish Central Bank and several other private banks. It is a unique initiative, certainly in a European perspective, in that it outsources the responsibility for online authentication and a vital part of online government security to a private company.

This "platform thinking" has led Denmark and its Steering Group for Public Cross-Sector Cooperation (STS) governance body to secure several similar successful outcomes. In addition to NemID, Denmark operates a common payment service used by all public organizations for making payments to citizens, and all Danish companies use a common invoicing service and gateway for sending invoices to all public sector entities. In Italy, an award-winning, common e-procurement service is used by all government agencies to procure equipment and services from the private sector that competes for bids. The Dutch NUP for joined-up government services recognizes the importance of identifying common services that are leveraged by all levels of government.


Federated governance

Successful programs have a federated governance orientation. They establish a shared platform capability, make its services available to all public sector ministries, agencies, departments, municipalities, and institutions, and, in some instances, the private sector and citizens, and they gradually seek to move more municipalities, departments, agencies, or even states to the strategy and the supporting "platform". This "movement" often occurs over the span of many years or even decades, requiring a long-term political vision and incremental implementation capable of surviving across multiple election cycles.

The initial definition of the shared service can be mandated by a central authority, but it seems that for the platform to be effective and useful to the point where departments and agencies adopt it, they must have a say in how it functions and in what services it offers them. Federated governance, whereby departments or agencies are involved in the initial decision-making and planning and then subsequently in the lifecycle management of services, is vital.

The program and shared platform are frequently governed by a federated steering group using anchors, meaning. large, prestigious states or departments that commit to and support the governance model early on. Early efforts benefit from having these committed anchors on board and, it seems they are brought into the development of successful programs from the very beginning.

Germany's approach to governance is one of the most widespread and complex enterprise architecture initiatives. The German IT Planning Council formulated a National eGovernment Strategy for 2010 - 2015 to lead the country's strategy and planning. Similar to other successful enterprise architecture projects, the strategy has a series of politically agreed upon strategic goals. Their focus is on building a shared federal infrastructure and shared services along with security and interoperability standards in support of government, as well as a cross-eGovernment framework to assist the IT Planning Council in coordinating activities across federal and state laws.

Highlighting the importance of a federated governance mode, the National eGovernment Strategy was developed in cooperation with a broad spectrum of stakeholders from the fields of administration, politics, and even science and business. Particular care was taken to involve citizens in this process by means of an online consultation that took place in September 2009. In this aspect, the strategy followed a typical Northern European pattern by which business stakeholders and citizens were involved in the development and review processes for national strategies involving information technology.

The German government created an action plan for improved IT governance, both for the different administrative areas and for IT. On December 5, 2007, the German Cabinet agreed on a Federal IT-Governance strategy to improve IT management within the government (see Resources for links). The goal was to optimize public administration services and promote IT innovation. The strategy takes into account internal and interdepartmental IT management, tackling current issues involving Germany's eGovernment strategy and IT security.

The case illustrates a wide-reaching initiative with high institutional complexity, backed directly by the German Chancellor and the German Cabinet, with implementation occurring over several years. The initiative covers all public administration areas in all of the 16 Federal States. In addition, clear and agreed-upon political measures and objectives were integrated into the strategy, and they conducted a complex review process, involving citizens and businesses in Germany.

In Denmark, the Steering Group for Public Cross-Sector Cooperation (STS), mentioned previously, is a classic example of federated governance. Its members include the Ministry of Finance; the National Association of Municipalities that represents the 99 municipalities; the association of the Danish Regions, which represents the five Danish regions that manage the country's health care sector; the Ministry of Economy and the Interior; the Ministry of Social Services; the Ministry of Health and Prevention; and the Ministry of Tax. STS is charged with developing the enterprise architecture for the public sector and is responsible for its overall priorities and objectives. Specific domain steering groups for social service, health care, corporate and business services, and public building and infrastructure implement the priorities and objectives in a cross-departmental manner. One interesting aspect about federated governance is its focus on cross-organizational outcomes anchored in the political system, not in a state agency for IT or technology or in an office of the CIO. It is not a centrally led initiative but, instead, one that is federated and works within established decision-making processes, with strong political backing from the most powerful Danish ministries.

STS receives support from the Department of Digitalization (DoD), which carries out many of the functions that often reside in an IT or technology department. However, the DoD works on a purely enterprise architectural level and manages Denmark's OIO and OIOXML standard. OIO is a white book standard that serves as guideline for the entire public sector's enterprise architecture, while OIOXML prescribes interchange standards for public agencies.

STS has created an overall enterprise architecture model with standards for information exchange based on the OIOXML standard. STS is responsible for the overall priorities, while a small set of domain steering committees implement them within their respective domains, such as social services or health care.

Table 1. Overview of public sector models, by country
Program or initiativeCountry Objective Political leadership Federated governance Shared frame of reference Supported by policy or passed legislation
Service Canada Canada Citizen centricity Yes Yes CGRM Yes
eGovernment 2.0 Germany Citizen centricity The German Chancellor Angela Merkel, the German Cabinet, and the 16 federal states Yes eGovernment Strategy Yes
NUP Holland Citizen centricity Dutch government Yes NORA Yes
GOV.UK UK Citizen centricity The Cabinet Office No. Central, mandated Yes Yes
IOC Rio de Janeiro, Brazil Interoperability The mayor of Rio de Janeiro n/a
STS Denmark Improve the delivery of government services to its citizens and businesses Department of Digitalization and the Ministry of Finance Yes OIO, FORM,
OIOXML
Yes
MEDCOM Denmark Health care connectivity Danish government Yes MEDCOM standards Yes
NemID Denmark One eID system for all citizens and businesses Danish National Bank, Danish Banker's Association, Department of Digitalization Yes NemID Yes

Trends

Judging from these commonalities, it is possible to identify trends that are shaping how successful enterprise architectures are developing within the public sector:

  • Enterprise architecture is practiced in an increasing variety of strategic and political circumstances. Although governance frameworks still focus on selling rather than telling, a trend toward more centrally mandated frameworks and shared services might be emerging. As examples, in the reforms undertaken in the UK, Holland, and Germany, the cabinets voted in favor of specific program elements.
     
  • The characteristics, uses, frameworks, tools, and policies of enterprise architecture seem to be increasingly eclectic, and enterprise architecture is practiced in many different ways. One common characteristic is that enterprise architecture is viewed as an approach to political enterprise transformation on a large scale, adopted to build better political outcomes, rather than an IT design and management approach limited to building better systems.
     
  • Enterprise architecture is moving up the value chain, out of the office of the CIO. It is increasingly being used to help governments align themselves to provide real value by applying the rigor of enterprise architecture to the business and political agenda.
     
  • Shared frames of reference, "languages," and ontologies such as those represented in FORM or NORA lay a foundation for governance in some initiatives, especially ones that are very broad-based, long-term, and span many levels of government, defense, or organizational borders . In these cases, common language and ontology support standardized exchanges of information but also provide a common view of how initiatives and projects relate to each other, fall within the same domain, represent duplicative investments, or address the same outcomes.
     
  • Shared platform services are becoming prevalent on a national scale, well beyond the size and scope of individual departments or agencies. Examples: National eID initiatives in Denmark, Holland, and Germany, the use of a shared web infrastructure for GOV.UK (the one portal-based entry point for citizens and businesses to the British government), and MEDCOM (the national Health Information Network in Denmark).

What works and what doesn't: Patterns and anti-patterns

Based on these observations and possible trends, some conclusions emerge as to what seems to work and what doesn't. These can be expressed as patterns for good behavior and conversely as anti-patterns for suboptimal or even counter-productive behavior.

Table 2. Patterns and anti-patterns
PatternsAnti-patterns
Successful EA programs are outcome-focused, with improved service to citizens (Service Canada, GOV.UK, NUP, or eGovernment 2.0), one entry point for the entire national health care sector (MEDCOM), improved force effectiveness and interoperability (NCW). EA programs should not be focused on strict models or even on architectures alone. In order to receive solid political backing, there must be clear expectations about which outcomes will be created and there must be measurable political or economic return attached to those outcomes that align with broader political intent.
A clear and accepted governance structure across levels and functions of government is paramount to successful EA outcomes. EA is not just about architecture. In the public sector, it is about building and then sustaining political and institutional support over many years. Unclear distributions of power, unclear mandates, and a constant struggle for political support will hinder outcomes.
Public sector organizations are able to make substantial progress with federated governance models that focus on selling rather than on telling, as long as there is a long-term strategy and adequate political will and backing for the programs across election cycles. Patience is certainly a virtue in many initiatives. Short-term programs that have unrealistic expectations will probably falter with a higher probability. To ensure patience, political and cross-party leadership seems vital. Similarly, telling behaviors may be counter-productive in a relatively large set of situations where individual departments and agencies have large
"freedom of movement."
Successful enterprise architecture agendas are anchored at the highest level of government, not in IT or in offices of the CIO, for example Service Canada, STS, eGov 2.0 (Germany), MEDCOM, UK.GOV, and NUP. EA programs should not be driven from technology centers or central, shared IT organizations although these organizations will play a large supporting role in defining models, ontologies, reference architectures, and IT policies and in providing oversight and architectural governance for implementing the actual changes.
Successful enterprise architecture programs are implemented incrementally and offer architecture frameworks that allow certain flexibility to participating organizations but are nonetheless influenced by strong standards. Rigid reference models or governance forms that focus on "telling" rather than "selling" will be hard-pressed to produce results in a political system.
The concept of shared platforms permeates successful programs. Platforms that are open, standardized, and adaptive allow participants to come on board at different times and at different levels of maturity (such as UK.GOV, NemID, MEDCOM, and Germany's eID) and allow an adaptation to local needs. EA programs cannot be overly prescriptive or stringent in their assumptions. They must allow a balanced amount of flexibility by considering that a shared platform must function on a global level, as well as a local (departmental) level.

What this means to the enterprise architecture method

Public sector strategy too often breaks down into departmental strategies that are occupied with driving program success specific to the mission of the individual department. Those strategies can resist coordinating resources with other departments. More recently, horizontal political strategies seek to drive outcomes such as better resource sharing or improved citizen-centric services across multiple agencies, departments, or states.

Perhaps more important than this level difference is the way strategy and decision-making is conducted within public sector organizations. Three different political strategy orientations can be identified, although more might certainly exist.

Prescriptive
Strategy results from a conscious, controlled, and deliberate process with appropriate resource commitments and political dedication. These types of programs often have solid support from the top of the political system.
 
Visionary
Strategy is a matter of finding new opportunities, but it is typically carried out in an experimental fashion, in the form of pilot projects. Resources are not committed until risks can be handled and certain outcomes secured.
 
Incremental
Strategy is executed fluidly and incrementally, not necessarily led by vision or political intent but because there is an overall framework or intention in place that works at a national or regional level. That framework forces local governments to take some form of action, for example the Affordable Care Act in the US.
 

In the prescriptive school, successful results come from a conscious, controlled, and deliberate process. Architecture, governance, process and outcomes are made explicit and complete as in the cases of NemID, Services Canada, Rio de Janeiro's IOC, and GOV.UK. Because it is necessary to deliver a shared platform that coordinates services across many parts of the government, the architecture must be open, adaptive, and extensible, yet use standardized message exchanges. Architectural frameworks such as the CGRM, NORA, OIO, or Federal Enterprise Architecture (US, federal) play important roles in ensuring that the shared platform can be leveraged by all relevant parts of government, yet also be localized to their specific needs and last for many years or even decades.

In the visionary school, strategy is dominated by the search for and exploitation of new technological opportunities, such as cloud computing, online government, or social business. It works well when policy is put in place to support a vision of technological opportunity, for instance for the United State's Cloud First Policy, the UK's insistence on considerations for a G-Cloud to be adopted by its ministries and departments, or Germany's National eGovernment Strategy. Shared frameworks, whether they are legal-, architecture- or policy-based, work well in supporting and embodying the vision.

In the incremental school, strategy is executed fluidly and incrementally, often on the basis of what can be negotiated from a political perspective or what can be accomplished with acceptable risk or acceptable means. Strategy is a process of learning and negotiation. Therefore, design and implementation of the enterprise architecture become indistinguishable, because design and implementation occur alongside each other. This allows for learning points to be integrated back into the decision and planning processes, until it is learning more than planning that fuels the process. MEDCOM began as an emergent strategy and worked in the form of a political and technical pilot project for some time before the strategy behind MEDCOM shifted into another gear and became prescriptive.

Within incremental strategy, logical incrementalism is a method of working that uses many small, incremental changes with certain outcomes rather than a few (extensively planned, but uncertain) large jumps. Logical incrementalism implies that the steps in the process are sensible. Logical incrementalism focuses on a "power-behavioral approach to planning" rather than on a formal systems planning approach that is prevalently used in many if not most enterprise architecture methods. In public policy, incrementalism refers to the method of change by which many small policy changes are enacted over time to create a larger, broad-based policy change. One example is what is observed on a large scale in the handling of the Euro crisis in 2011 and 2012, advocated by the German approach to crisis management. From a traditional point of view, the Euro crisis illustrates how markets remain confounded that a single, massive end-solution is not deployed. But in the perspective of the German chancellor and her advisers, the Euro zone represents an immensely complex system with a significantly higher institutional complexity than any other currency union. Due to the complexity, many advisers favor an incremental approach.

Proponents of the incremental school and incrementalism argue that no complex sustainable system is the result of formal planning but that all complex systems result from incrementalism.

In this sense, logical incrementalism becomes a kind of macro-incremental-political approach to enterprise architecture analogous to agile development in software engineering. "Macro" refers to the idea that the entirety of a public sector can be seen as a system of systems that has an architecture and that this architecture can be developed to enable the public sector to better come to terms with a new set of political priorities that deal with greater mission effectiveness and improved resource sharing. "Incremental" refers to the breakdown of the "structural change" in smaller chunks that can be planned with relative ease, where outcomes are more predictable and therefore less risky or costly, but where each of the chunks are subordinated to the architecture and the objective. "Political" refers to the political realities of the system and becomes a reminder of the importance of building and sustaining political support and, indeed, for leading enterprise architecture initiatives from within the political system and not from within offices of IT.

What this means in terms of enterprise architecture is that open, adaptive, and standardized enterprise architecture frameworks might yield better results in combination with incremental and political change than traditional formal systems planning methods for enterprise architecture.

When looking at the cases referenced in this article, it is easy to see that most successful programs seem to contain a blend of the three schools. They are visionary in how they plot a course for the future. Many programs were formed in the late 1990s with a horizon for implementation that extends into the 2010 - 2015 timeframe. They are partly prescriptive in that they mandate a policy framework, a legal framework, and, in many instances, an architectural framework or reference model with interoperability standards. And in several cases, these frameworks and models are passed as laws by the government They are incremental in that they recognize the need for using pilot projects and for allowing departments and agencies to come on board at their own pace.

Thus, when it is practiced, enterprise architecture should emphasize vision, building political support, establishing federated governance models and adaptive enterprise frameworks over formal architecture planning, and lend from incremental development techniques to structure programs into smaller, incremental chunks. The method itself becomes subordinated to the framework and must be weaved into the political decision-making processes and agenda and led from the political system.

Resources

Learn

Get products and technologies

  • Download a free trial version of Rational software.
  • Evaluate other IBM software in the way that suits you best: Download it for a trial, try it online, use it in a cloud environment, or spend a few hours in the SOA Sandbox learning how to implement service-oriented architecture efficiently.

Discuss

  • Ask and answer questions and increase your expertise when you get involved in the Rational forums, cafés, and wikis.
  • Join the Rational community to share your Rational software expertise and get connected with your peers.

Comments

developerWorks: Sign in

Required fields are indicated with an asterisk (*).


Need an IBM ID?
Forgot your IBM ID?


Forgot your password?
Change your password

By clicking Submit, you agree to the developerWorks terms of use.

 


The first time you sign into developerWorks, a profile is created for you. Information in your profile (your name, country/region, and company name) is displayed to the public and will accompany any content you post, unless you opt to hide your company name. You may update your IBM account at any time.

All information submitted is secure.

Choose your display name



The first time you sign in to developerWorks, a profile is created for you, so you need to choose a display name. Your display name accompanies the content you post on developerWorks.

Please choose a display name between 3-31 characters. Your display name must be unique in the developerWorks community and should not be your email address for privacy reasons.

Required fields are indicated with an asterisk (*).

(Must be between 3 – 31 characters.)

By clicking Submit, you agree to the developerWorks terms of use.

 


All information submitted is secure.

Dig deeper into Rational software on developerWorks


static.content.url=http://www.ibm.com/developerworks/js/artrating/
SITE_ID=1
Zone=Rational
ArticleID=850778
ArticleTitle=What defines success with public sector enterprise architecture
publish-date=12112012