Modernism and Enterprise Architecture

Underlying conventional enterprise architecture theory and practice are some implicit assumptions that could be loosely characterized as modernist. Several people are offering more or less radical departures from conventional enterprise architecture, which could be loosely characterized as post-modern.

Aspects of Modernism

• Fordism- Simple decomposition of responsibility, authority, expertise and work (RAEW)
• Functionalism - Form follows function
• Linear Explanation - Simple association between cause and effect – e.g. process improvement
• Cartesian Theatre - Unified representation of what is going on (WIGO)
• Central Planning - Unified decision-making
• Unified Value System - Complete and consistent set of shared goals and objectives on behalf of The Enterprise
• Ultimate Solution - Converging towards an ideal set of systems, perfectly aligned to The Business

Limitations of Modernism

• Difficulties handling complexity, emergence and self-organization.
• Lack of agility, flexibility, evolution.
• Constrains organizational learning.
• No explicit treatment of holistic architectural properties such as balance and harmony.
• No room for pluralism and human values.

Towards Postmodernism

Obviously the labels "classical" and "modernist" can mean different things, but I associate the classical approach to architecture with Vitruvius - Firmness, Commodity and Delight - and I see the essential characteristics of the classical aesthetic as an emphasis on balance, harmony and order.

This emphasis on balance is completely absent from the modernist functionalist aesthetic, which seems to assume that if you take care of the parts and connect them together properly, the whole will look after itself.

The IT modernists (such as James Martin) told us to concentrate on Commodity (roughly translated as functional requirements), because Firmness (reliability and other non-functional) could be handled by technical design and innovation, and Delight could be handled by user design specialists (ha!). For his part, Zachman has said nothing to indicate that he doesn't fully share this mindset, and his interpretation of the "classical" is conceptually and metaphysically flawed. We may note that most of the notations favoured by enterprise architects (IDEF, UML, ArchiMate) concentrate on the functional.

In our time, it is of course Alexander and Salingaros who represent the return to balance and order, as the title of Alexander's masterwork indicates, although I suspect they would not take kindly to being labelled as neo-classical.

Meanwhile, an emphasis on self-organization and heroic agility can be seen as echoing some of the romantic aesthetic. After all, the post-modern aesthetic often combines elements of the neo-classical with elements of the neo-romantic.

Exploration of New Ideas

In response to the perceived limitations of the modernist approach to enterprise architecture, various thinkers are looking into other disciplines for ideas that can be imported into enterprise architecture.

Systems Thinking	Holism and Emergence Viable Systems Model (VSM) Wicked problems Dissipative Structures (Prigonine via @empiricator)
Design Thinking and Urbanism	Creativity Organic planning Metadesign  Pace layering Post-modernism
Organization Theory	Loosely coupled systems Self-organizing systems
Ecological Thinking	Panarchy Evolutionary change
Complexity Theory	Anticipative Systems Complex Adaptive Systems Dynamics of Strategy (I-Space, Cynefin)

have I missed anything?

The challenge here is two-fold. Firstly, how can a heterogeneous selection of ideas from different sources be combined into a coherent approach? Secondly, what kind of practical hands-on research (reflective practice) is needed to turn interesting speculation into grounded and credible knowledge?

Some Post-Modern and Other Approaches to EA

I don't know how many of the following approaches can be properly described as postmodern, but they all represent various attempts to break free from the limitations of modernism.

Philip Boxer et al, Asymmetric Leadership

Dave Duggal and William Malyk, Putting Work in Context for Enterprise Agility (December 2010)

Nicholas Gall (Gartner), Panarchitecture: Architecting a Network of Resilient Renewal (January 2011)

Nigel Green, A new context for EA: The Enterprise: An ecosystem of values and value (Sept 2010)

Dion Hinchcliffe, Post-Modern Enterprise Architecture: Service-Oriented, Agile, Decentralized (April 2005). Pragmatic new models for enterprise architecture take shape (August 2009)

Anders Jensen-Ward, Pondering a new model of architecture and postmodern complexity: Deleuze’s rhizome as the substitute for architectural layers and control. (June/July 2010) via Tom Graves. On management and systems (Aug 2009) On architecture and strategy (Aug 2009) Architectural control as a management paradox (Jan 2011), Rhizome: On Dilemmas in Enterprise Architecture Planning (April 2011)

@rlimbanda To understand human systems based on postmodern natural architecture and the transformation of 21st century social enterprise. // And to understand and share the things that make us human #thoozd (December 2009) via Tom Graves

Further reading

Max Boisot and Bill McKelvey, Complexity Science-A Bridge Between Modernist and Postmodernist Perspectives (2008)

Hubert Maturana, Metadesign (1997)

Neil McBride, Rachel Lander and Steve McRobb, Postmodernist Business Information Management 1997 (HTML, PDF)

See my previous post on Adaptation and Adaptability

See Linked-In discussion "Is Enterprise Architecture (and IT in general) trapped in the psychological prison of Modernity?" from January 2011 onwards



Updated October 21, 2012
Some links corrected 19 February 2020. Some remaining broken links under investigation - please help.

EA Talk on Managing Complexity

#entarch I shall be was one of the speakers at the Unicom EA Forum on February 24th in Central London. This post now edited to remove invitation and include slides.

---

Outline

As business becomes more complex, enterprises are faced with structural complexities that critically affect business performance. Enterprise architecture provides a way of planning and coordinating both human systems and technical systems (including IT), and thereby overcoming the structural inhibitors to business performance. I shall talk about some emerging practical methods for business architecture, and show how these can be used in tandem with the standard popular approaches to manage the true complexity of the business.

Slides

EA Talk on Managing Complexity from Richard Veryard

Following my talk, Professor Bernie Cohen talked about some related work by Philip Boxer and himself. Slides no longer available. https://www.slideshare.net/PhilipBoxer/mgt-complex-ea-feb-2011-healthcare

What is an EA vendor?

#entarch I just asked an innocent question about enterprise architecture vendors on Twitter, partly to find out which companies identified themselves as such.

Some industry analysts focus on vendors selling software tools for use by architects. For example, I find a Gartner report on Cool Vendors in Enterprise Architecture from March 2009, highlighting two small vendors with what Gartner regards as fresh approaches: "one with metrics and innovative graphics, and the other providing an affordable entry point that leverages users' Visio skills".

Even some people who insist that enterprise architecture shouldn't be thought of as merely software architecture seem to think that "tools" only means "software tools". For example Ian Louw (@bpmfundamentals) approvingly repeats @joemckendrick's headline "EA still too fixated on technology, versus business" but happily discusses Gartner's EA Magic Quadrant for 2009 without mentioning the possible contradiction.

But the EA market is much more than software. There are companies selling frameworks and methodologies and other intellectual property, which may be packaged as a knowledgebase or manuals or training. There are companies selling standard industry models, either as stand-alone products as IBM does, or embedded into application packages. There are consultancies and systems houses selling complete solutions, although this rather depends on one's notion of completeness.

Thus when an industry analyst classifies EA vendors in terms of "vision" and "ability to execute", we might reasonably expect these terms to refer to EA vision and ability to execute (thus probably having to do with frameworks and processes and collaborations as well as software) rather than solely the ability to envision and deliver lumps of software.

Who do you regard as EA vendors? Do you regard your own organization as an EA vendor or supplier?